B 429208

LORGE
NO..
E. D. CAMPBELL
ANN ARBOR, MICH.
AY
ARTES
LIBRARY
1817
SCIENTIA
VERITAS
OF THE
UNIVERSITY OF MICHIGAN |
ER E.PLURIOUS UNUM
TUEBOR
SI-QUÆRIS-PENINSULAM-AMŒNAM ›
CIRCUMSPICE
THE GIFT OF
Estate of Prof. E.D]. Campbell.
(
1

A
PRACTICAL TREATISE ON
METALLURGY.
TN
665
.K39
15
₤ 5
1868
V, 2
:
1
LONDON
PRINTED BY WILLIAM crookes, CHEMICAL NEWS office,
BOY COURT, Ludgate Hill, E.C.
A
PRACTICAL TREATISE
ON
METALLURGY,
ADAPTED FROM THE LAST GERMAN EDITION
OF
7:
PROFESSOR KERL'S METALLURGY,
BY
WILLIAM CROOKES,
AND
F.R.S.. &c.,
ERNST RÖHRIG, Ph.D., M.E.
IN THREE VOLUMES-VOL. II.
COPPER, IRON.
ILLUSTRATED WITH 273 WOOD ENGRAVINGS.
LONDON:
LONGMANS, GREEN, AND CO.
1869.
i
!
*
From the istitu
Flu & W
PREFACE.
IN the preface to the first volume of our Treatise on
Metallurgy we referred to the importance of the two metals
Copper and Iron, which we proposed to make the chief
subjects of our second volume; but as we proceeded with
our work, the great importance of these two metals,
particularly of Iron, which, in Ure's "Dictionary of Arts
and Mines," is justly called "the truly precious metal," and
the impossibility of doing adequate justice to the subject
within the compass we had originally intended, have de-
cided us to extend our Treatise to three volumes. This
third volume will contain an account of Steel and Fuel, a
chapter on the Construction of Furnaces, some Supple-
mentary remarks on Iron, Tables of Weights, &c., and a
Glossary of Technical Terms.
Iron is not only more useful than the other metals, but
there is also a great difference in the mode of its production.
Whilst other metals are met with in commerce nearly
chemically pure, Iron is metallurgically produced in a
combined state, forming three distinct varieties—indeed, we
might almost say three metals-so widely do their properties
differ. Each variety of Iron may be produced in various ways,
and although more study has been devoted to Iron and its
production than to any other metal, there is still great scope
for improvement. There is also a great difference between
the smelting of Iron and the smelting of other metals. In
vi
PREFACE.
most smelting operations the crude metal only is usually
produced, whilst in the smelting of Iron the producer is
almost always a manufacturer as well; he fabricates rails,
tyres, plates, even tinned plates, &c., &c.; and, in fact,
production and manufacture are so closely allied as to be
inseparable.
The manufacture of the three varieties of Iron must be
conducted with reference to the purpose for which the
product is intended, and this we may call the chemical
section of the Iron Manufacture, or Metallurgy proper;
whilst the further treatment of the Iron, such as rolling
rails, &c., forms the mechanical part. An elaborate de-
scription of the mechanical part is not in our province;
moreover, its full elucidation would require large working
drawings, which obviously could not be given with the
present Treatise. On the other hand, the production of
the different kinds of Cast- and Wrought-Iron has been
described in extenso, and we also propose to describe
elaborately the various processes for the production of Steel.
Our treatment of Iron is thus conformable to that of the
other metals. We feel sure that when the importance and
vast field of the production of Iron is considered, we shall
be excused for having exceeded our original plan of two
volumes.
In describing the processes for the production of Copper
we have bestowed due consideration to the important
question of the Copper Smoke Nuisance.
The third volume will shortly appear, and we trust
that the reception of the present and the third volume will
be as encouraging as that given to the first volume both
here and in America.
LONDON, March, 1869.
CONTENTS
OF THE
SECOND VOLUME.
CHAPTER I.
COPPER.
PAGE
Principal Ores of Copper
Native Copper, 3. Sulphides of Copper 4. Copper Ores containing
Antimony and Arsenic, 5. Oxidised Copper Ores, 6. Copper
Salts, 6. The English Copper Veins, 8.
Synopsis of Processes for the Extraction of Copper from
its Ores
The different ways of extracting Copper, 10.
I
ΙΟ
Conditions on which
Classification of
the choice between the Processes depends, II.
the Processes, 13.
Treatment of Copper Ores and Products in the Dry
Way
16
Sulphuretted Ores and Products.
16
Treatment of Sulphuretted Substances in Cupola
Furnaces.
16
Usual Treatment, 16. Operations required to work Copper Ores
free from Silver and Gold, 17.
Roasting the Ores
18
The purpose of this Operation, 18. Theory, 19.
Kernel Roasting,
21. Roasting in Heaps, 24. Roasting in Mounds, 26. Roasting
in Cupola Furnaces, 27. Roasting in Reverberatory Furnaces, 27.
Roasting Products, 28.
viii
CONTENTS.
Smelting the Roasted Ores.
Object of this Process, 28. Chemical Reactions, 28. Requirements
of a Normal Process, 31. A Suitable Temperature, 31.
A pro-
perly Reducing Reaction, 31. The Degree of Roasting, 31. The
Composition of the Ore Mixture, 32. The Construction of the
Furnaces, 38. Fuel, 42. Indications of the Process, 42. Smelting
Products, 43.
Roasting the Raw Matt
Object, 46. Concentration of the Matt, 46. Methods of Roasting,
48. Roasting Products, 49.
Concentration Smelting of the Roasted Raw Matt
Roasting the Concentrated Matt.
28
46
Object, 50. Smelting Mixture, 50. Products, 50.
50
Object, 52. Modes, 52. Products, 52.
52
duction of Black Copper.
Smelting the Roasted Concentrated Matt for the Pro-
Chemical Reactions, 52. Smelting Mixture, 53. The Conditions
influencing the Process, 54. Products, 55.
52
Refining the Black Copper .
Theory, 56. Oxidising Smelting of Black Copper, 57. Reducing
Smelting of Rosette Copper, 57.
56
Refining of Black Copper in Small Hearths
Hearth, 58. Operation, 59. Chemical Reactions, 61.
58
Modifica-
tions for accelerating the Process, 63. Fuel, 64. Indications of
the completion of the Process, 64. Products, 65.
Purifying Black Copper in Blast Reverberatory Fur-
naces
Perm,
Furnace, 67. Process and Manipulation, 68. Tajova, 68.
69. Treatment of Plumbiferous Copper, 70. Schmöllnitz, 71.
Second Refining (Toughening) of Rosette Copper in
Small Hearths
Manipulation, 72. Hot Blast, 73. Rising of the Copper, 73.
Refining of Black Copper by Poling
Products of the Toughening Process, 77. Influence of Foreign
Substances in Copper, 78.
Illustrations of the German Method Employed in Smelt-
ing Copper Ores free from Lead, Silver, and Gold.
1. Smelting in Sump Furnaces, 84. Atvidaberg, 84. Fahlun, 95.
Röraas, 98. Szaszka, 100. Altgebirg, 101. Schmöllnitz, IOI.
Agordo, 103.
2. Smelting in Channel Furnaces with two open eyes, 107. Sterne
and St. Josephsberg near Linz, on the Rhine, 107. Smelting
Copper Pyrites at the Upper Hartz, 108. Kupferberg in Upper
Silesia, 109. Isabellenhütte at Dillenburg in Nassau, 110.
67
71
76
84
CONTENTS.
ix
Smelting of Copper Slate at Riechelsdorf and Friedrichshütte, III.
Thalitter in the Grand Duchy of Hesse, 116. Nischnetagilsk, 116.
3. Smelting in Crucible Furnaces, 116.
Illustrations of Smelting Argentiferous and Auriferous.
Copper Ores by the German Method
117
1. Extraction of Copper in the Desilverisation of Ores
and Matts
117
Nagybanya, 117. Fahlun, 119. Müsen, 119.
2. Extraction of Copper at the Desilverisation of Matt 124
Oeblarn in Upper Styria, 124.
3. Extraction of Copper at the Desilverisation of Black
Copper
124
Oker, 124.
Treatment of the Cupriferous Intermediate Products.
and Residues
129
Upper Hartz, 130. Csiklowa, 131.
Process of Treating Sulphuretted Ores, &c., in Rever-
beratory Furnaces
132
Different Operations of this Process, 132. The Operations required
by the English Copper Smelting Process. 133. Gurlt's suggestion
with regard to the number of operations required, 134.
Calcining the Sulphuretted Ores.
135
Furnaces, 136. Chemical Reactions, 138. Vivian's Experiments of
Condensing the Roasting Smoke, 139. Improvements of the Roast-
ing Process, 140.
143
The Copper Smoke Question
Gurlt's Methods of Roasting, 144. Mr. Thomas Bell's Method, 147.
Gerstenhöffer's Furnace, 148. Mr. Peter Spence's Furnace, 152.
Production of Raw Matt (Regulus) from Roasted and
Raw Ores
Purpose of the Process, 160.
tions of the Process, 163.
Roasting the Raw Matt
Production of Blue Metal
160
Theory, 160. Furnaces, 161. Opera-
Products, 165.
168
169
Production of White Metal.
170
Production of Pimple Metal
171
Smelting of the Slags with Pyritic Ores for the Pro-
duction of White and Red Matt
171
Roasting Concentrated Matt for the Production of Black
Copper
172
+
The Production of the best selected Copper, 173. Refining Copper
Bottoms, 174.
VOL. II.
b
X
CONTENTS.
Refining the Black Copper
Furnaces, 174. Process, 174. The Loss of Copper in the Reverbera-
174
tory Process, 177.
Illustrations of the Reverberatory Process
Smelting Process as described by Le Play.
Calcining the Copper Pyrites, 178. Raw Smelting, 178. Roasting
of the Raw Matt (Bronze Matt), 178. Production of White Con-
centrated Matt, 179. Production of the Blue Metal, 180. Smelt-
ing the Slags, 180. Production of White Extra Metal, 181. Pro-
duction of the Concentrated Matt (Regulus) 181. Black Copper
Smelting, 182. Refining the Black Copper, 182.
Smelting Processes according to Percy, Napier, and
Hyde Clark
Smelting of partially Oxidised Ores, 183. Smelting Ores with 7 or
8 per cent of Copper, 184. Treatment of Ores containing 9 per
cent of Copper, 185. Smelting Modifications in the case of Ores
containing 9 per cent of Copper, 185.
Copper Works at Steinwerder near Hamburg
Copper Works at Duisburg.
177
177
183
187
189
Extraction of Copper by combined Processes in Cupola
and Reverberatory Furnaces
192
Comparison of the Processes in Cupola Furnaces with those in
Reverberatory Furnaces, 192.
Illustrations of the Combined Processes
193
Mansfeld, 193. Schmöllnitz, 211. Lower Hungary (Schemnitz,
Kremnitz, Neusohl, Zsarnowitz, Tajova), 213. Freiberg, 214.
Boston, 215. Copper Works on the Rhone, 217. Bogos-
lowsk's Works at Siberia, 219. The Total Production of Copper
in Russia, 220.
Processes Employed when Smelting Oxidised Ores.
220
Different Methods in Cupola and Reverberatory Furnaces, 221.
Illustrations of Smelting Oxidised Copper Ores.
Chessy, 221. Bouc, 222. Perm, 222.
221
Smelting of Native Copper
226
•
Different Methods, 226. Detroit, 226. French Copper Works, 229.
Extraction of Copper in the Wet Way
229
Ores best fitted for the Process, 230.
The Chief Operations of the Hydro-Metallurgical Ex-
traction of Copper
231
A. The Transformation of the Copper into a Soluble
State
231
Agents Water, 231. Dilute Muriatic Acid or Sulphuric Acid, 233.
Sulphurous and Muriatic Acid Gas together with Steam, 235.
Acid Fluids containing Sulphate of Iron, 236. Chlorination
CONTENTS.
xi
without employing Dilute Muriatic Acid, 236. Chlorination in
the Wet Way by Chloride and Subchloride of Iron, 237. By a
Solution of Common Salt, 238. Chlorination in the Dry Way by
Roasting with Common Salt, 238. Chlorination in the Dry and
Wet Way, 241. Agents for Chlorination : Concentrated Muriatic
Acid, 241.
Lixiviums containing Chloride and Subchloride of
Iron, 241. Chloride of Magnesium and Lixiviation, 242. Am-
monia, 242. A Solution of Sulphite and Hyposulphite of Soda,
242.
B. The Lixiviation of the Copper Salts
243
C. Purification of the Lixiviums before Precipitating
from Iron
243
D. Precipitation of the Copper (Cementation)
E. Treatment of the Products of Precipitation
244
248
Cement Copper, 248. Copper Vitriol, 248. Metallic Copper, 248.
Sulphide of Copper, 249. Oxychloride of Copper and Hydrated
Oxide, 249. Liquids, 249.
Extraction of Cement Copper from Native Solutions.
Schmöllnitz, 250. Amlwch in the Isle of Anglesea, 252.
Hydro-Metallurgical Extraction of Copper from Oxi-
dised Ores
Illustrations of this Process
Sternerhütte, near Linz on the Rhine, 252. Stadtbergen, 253. Alderley
Edge, 255.
Hydro-Metallurgical Extraction of Copper from Sul-
phuretted Ores and Products
The Method of Forming Sulphates, 258.
250
252
252
256
Illustration of this Process of Extraction
Riotinto, 256. Agordo, 257. Foldal in Norway, 259. Sinding's
Method of Producing Sulphuretted Hydrogen Gas, 260.
and Haupt's Process, 262. Bischof's Method, 262. The Annual
Production of Copper, 263.
256
Becchi
CHAPTER II.
Classification of Iron
•
IRON.
Pure Iron, 264. Cast-Iron or Pig-Iron, 264. Wrought or Malleable
Iron, 264. Steel, 265. Classification of the Metallurgy of Iron,
265. Estimation of the Annual Production of Pig-Iron in the
whole World, 265. The Consumption of Iron for each Inhabitant
of different Countries, 266. Statistics on the Quantity of Pig-Iron
produced in Great Britain, 266.
Production of Pig-Iron
264
267
xii
CONTENTS.
Different kinds of Pig-Iron; its Properties and Forma-
tion.
White Pig-Iron, 267. Spiegeleisen, 269. Flowery and Cellular
Pig-Iron, 276. White Pig-Iron produced by the Regular Process,
279. White Pig-Iron produced by an Irregular Process, 280.
Chilled White Iron, 281. Grey Iron, 282, Light Grey and
Mottled Fine Granular Pig-Iron, 286. Dark Grey Pig-Iron, 287.
The.Properties of Grey Pig-Iron.
267
287
Texture and Strength, 287. Specific Gravity, 289. Melting Tem-
perature, 290. The Application of Grey Pig-Iron, 294.
Influence of Foreign Substances on Pig-Iron
296
Silicon, 296. Sulphur, 301. Phosphorus, 306. Nitrogen, 310.
Manganese, 311. Copper, 314. Arsenic, 314. Tin, 315. Zinc,
315. Lead, 315. Vanadium, Molybdenum, Chromium, Tungsten,
Titanium, 316. Nickel and Cobalt, 318. Aluminium, Calcium,
and Magnesium, 318. Alkali Metals, 319.
Chemical Constitution of Pig-Iron
Limit between Pig-Iron, Wrought-Iron, and Steel, 322.
1. Production of Pig-Iron from Iron Ores.
Smelting Materials
A. Iron Ores and their Preparation.
Oxides of Iron-Magnetic Iron Ores, 324. Chrome Iron Ore, 327.
Titaniferous Iron Ore, 327. Franklinite, 327.
Hæmatite, 328.
Specular Iron Ore, 328. Red Hæmatite, 328. (Kidney Ore, Red
Ochre, Puddler's Ore).
Hydrates of Oxides of Iron-Brown Iron Ores, 331. Yellow Iron
Ore, 333.
Aluminous Yellow Iron Ore, 334. Bog Iron Ore,
Limonite, Lake Ore, 334.
Carbonate of Iron:-Spathic Ore, 336. Clay Ironstone, 338. Black-
band, 341. Cleveland Iron Ores, 343.
Siliceous Iron Ores :-Chamoisite, 344. Iron Garnet, 344. Non-
tronite, 344. Yellow Earth, 344. Slags, 345. Scrap Iron, 347.
320
323
323
323
The Adaptability of Iron Ores to the Smelting Process 347
With regard to their Percentage of Iron, 347. The State of Aggre-
gation of the Ore, and the State of the Iron contained in it, 348.
The Foreign Associates, 349.
The Dressing of the Ores.
352
Purpose of the Dressing, 352. Sorting by Hand, 352. Weathering,
353. Crushing and Washing the Ores, 353.
Roasting or Calcining Iron Ores.
355
The Purpose of Roasting, 355. Roasting Methods, 358. Roasting
in Open Heaps, 358. Roasting in Mounds, 363. Roasting in
Furnaces or Kilns, 363. Furnaces without a Grate, 365. Fur-
naces with a Plane Grate, 365. Furnaces with a Grate of Conical
Shape, 372. Furnaces with Step Grate, 372. Furnaces Heated
by Flame, 375. Furnaces Heated with Gas, 377.
CONTENTS.
xiii
Weathering of the Roasted Iron Ores.
380
Breaking-up the Iron Ores.
382
Mixing the Iron Ores.
383
Purpose, 383. Behaviour of too Rich and too Poor Mixtures, 384.
The Mixture with regard to the Formation of Slag
385
The Composition of Blast Furnace Slags, 389.
The Mixing of the Iron Ores with regard to the Quality
of Fuel employed
396
Charcoal, 396. Coke, 397. Mineral Coal, 401.
The Ore Mixture with regard to Temperature and
Pressure of the Blast
401
The Ore Mixture with regard to the Quality of the Pig-
Iron
403
Fluxes
405
Basic Fluxes, 406. Acid Fluxes, 407. Neutral Fluxes, 407.
Fuel
Action of the Fuel, 409. Circumstances Regulating the Consump-
tion of Fuel, 409. Kirchweger's Conclusions, 409.
Carbonised Fuel .
408
411
Advantages, 418. Disadvantages, 418. Wood, 419.
Charcoal, 411. Coke, 412. Chief Differences in Smelting with Coke
and with Charcoal, 414. Composition of the Waste Gases, 416.
Temperature of the different parts of Furnaces for Hot and Cold
Blast, 417. Turf Coal, 417.
Fuel in a Raw State
Mineral Coal, 420. Anthracite, 422.
418
Turf, 419.
The Blast Furnace and its Accessories
424
Lifts
424
Vertical Lifts by means of Steam Power, 424. The Inclined
Planes, 425. The Jacob's Ladder or Endless Chain System, 426.
The Water Balance, 426. Pneumatic Lifts, 429.
Blast Furnaces
432
Requirements of Furnaces for the Production of Pig-Iron from
Ores, 432. Different kinds of Furnaces, 432.
433
The Older Fur-
Construction of the Furnaces
Chief Requirements in the Construction, 433.
naces, 433.
The more Modern Construction of the Blast Fur-
naces, 439. A Combination of both kinds, 441. General Rules
concerning the Erection of Blast Furnaces, 441. Foundation, 441.
Corner Pillars, 441. Rough Walling, 442. Chimney or Tunnel
Head, 442. The Inner Shaft Linings, 442. Furnace Hearth, 443.
Furnaces provided with a Second Fore Hearth, 444. Furnaces
with a Closed Front Wall (Blauöfen), 444. Hearth Casing
formed of Clay and Sand, +++.
xiv
CONTENTS.
Collecting the Waste Gases
Apparatus formed of a Cylinder suspended in the Furnace Shaft, 448.
Apparatus in Siegen, 449. Cup and Cone Charger, 450. Langen's
Apparatus, 451.
Relations between the Different Parts of Iron Blast
Furnaces and the circumstances influencing these.
Relations.
The Interior Form of Blast Furnaces, 452. The Shape of the Fur-
nace Shaft for easily fusible Ore Mixtures, 453. The Form of
Furnaces for Ore Mixtures difficult to fuse, 454. The Section of
Blast Furnaces, 456. Rachette's Furnace, 458. Truran's Fur-
nace, 460. Dilla's Furnace, 463. The absolute Size of the In-
terior of Blast Furnaces, 463.
Theoretical Rules for Fixing the Proportion of Blast
Furnaces.
The Furnace Belly, 464. Lindauer's Formula, 466. The Position
of the Belly, 469. The Height of the Furnaces, 470. The Width
of the Furnace Mouth, 472. The Boshes, 474. The Shape of
Furnaces which have been blown out, 475. The Hearth, 476.
The Lower Hearth or Crucible, 478. The Tymp, 478. The
Dam, 479.
The Tuyeres, 479. The Number of the Tuyeres,
479. The Quantity of Blast which a Furnace requires, 480.
The Distribution of the Tuyeres, 480. Buschbeck's Determina-
tion of the Quantities of Blast, 481. The Quantity of Moisture
introduced along with the Blast, 484. The usual Tuyeres for
Charcoal Furnaces, 485. The Tuyeres and Blast Pipes for Coke
Furnaces, 485. The Pressure of the Blast, 486. The Position
of the Tuyeres, 487.
Example of the Projection of a Coke Blast Furnace.
according to Lindauer's Formula
Blowing Machines
Cylinder Blast Engines, 491. Manometers, 492. Influences on the
Velocity of the Blast, 494. Calculation of the Quantity of Blast,
496. Karsten's Formula, 496. Scheerer's Formula, 497. Schwind's
Contrivance, 498. Weisbach's New Formula, 499. Bornemann's
Graphic Tabula, 500. Examples explaining the Application of
the Graphic Table, 500. Other Formulæ by Weisbach, 502.
Table of Blast calculated from one of those Formulæ, 503.
Blast Regulators
Blast Heating Apparatus
447
452
464
488
491
505
Wasseralfingen Apparatus, 512.
Advantages of Hot Blast, 506. Disadvantages of Hot Blast, 508.
506
Pipe, 514. Cowper's Apparatus, 516.
Calder Apparatus, 513. Pistol
Iron Blast Furnace Process
Theory of the Process.
518
518
CONTENTS.
XV
Different Zones of the Furnace, 519. The Preparatory Heating
Zone, 519. Zone of Reduction, 521. Zone of Carbonisation,
523. Smelting Zone, 524. Zone of Combustion or Oxida-
tion, 526. The Lower Hearth, 529. The Reasons of the Fluc-
tuation of the Temperature in Blast Furnaces, 530. Time occu-
pied by the Charges in passing through the Furnace, 531.
The Manipulations in the Blast Furnace Process
Drying and Warming the Furnace, 532. Blowing In, 534. Charging
Ore and Fuel, 537. Manipulation in the Hearth, 541.
The
Removal of the Slag, 541. The Removal of the Pig-Iron, 542.
The Cleansing of the Hearth, 545. Repairs of the Hearth 546.
Stopping the Furnace, 547. Blowing Out, 547. The Length of
Operations, 548.
Conduct of the Blast Furnace Process and the Indica-
tions of the Working Condition of the Furnace
The Nature of the Pig-Iron, 549. Grey Iron, 551. Scheerer's
Classification of Differences in the Regular Process, 551. The
Normal Process, 552. Examples for the Production of Mottled
and Grey Iron, 555. White Pig-Iron, 559.
White Pig-Iron, 559. Spiegeleisen, 559.
Flowery and Cellular White Iron, 561. White Pig produced at
a Regular Process, 562. White Pig produced at an Irregular
Process, 563. The Nature of the Slags, 565. Slags resulting
from the Regular Process, 566. Slags resulting at the Irregular
Process, 567. Behaviour of the Smelting Mass before the Tuyere,
568. Behaviour of the Flame from the Furnace Mouth and from
below the Tymp, 569. Analyses of Deposited Smoke, 570.
The Products of Iron Blast Furnaces.
Pig-Iron, 571. Slags, 572. Deposits in the Furnaces and Segre-
gations in the Lower Part of the Furnace, 573. Ferruginous Bears,
574. Cyanide of Potassium, 574. Cyano-nitride of Titanium,
575. Waste Gases, 576.
The Examination of a Blast Furnace Establishment
Tables giving Details of some of the mentioned Points
in different Blast Furnaces
Founding and Moulding
The Adaptability of Pig-Iron for Castings, 578. Circumstances
making advisable to Re-melt Pig-Iron, 579.
Re-melting of Pig-Iron
Different Methods, 580. Literature, 580.
532
548
57I
•
576
578
578
580
581
582
583
Re-melting Pig-Iron in Crucibles
Applicability, 581. Manipulation, 581.
Re-melting Pig-Iron in Cupola Furnaces.
Applicability, 582.
Smelting Materials
Pig-Iron, 583. Scraps of Cast- and Wrought-Iron, 583. Fluxes, 584.
Fuel, 585.
..
xvi
CONTENTS.
Smelting Apparatus
586
Charcoal Cupola
Hot Blast Appa-
Blowing Machines,
Furnaces, 586, 588. Portable Furnaces, 588.
Furnaces, 592. Coke Cupola Furnaces, 593.
ratus, 599. Advantages of Hot Blast, 599.
601. Fans, 601. Quantity of Blast which a Cupola Furnace
requires, 603. Determination of the Amount of Blast, 604.
The Cupola Furnace Process
Charcoal Cupola Furnaces, 604. Blowing-in, 604. Regular Process,
605. Irregularities of the Process, 606. Tapping or Ladling the
Iron, 607. Coke Cupola Furnaces, 608. Products, 608. Illus-
trations of the Cupola Furnace Process, 609.
604
Re-melting of the Pig-Iron in Reverberatory Furnaces 614
Smelting Materials
Pig-Iron, 615. Fluxes, 616. Fuel, 616.
Smelting Furnaces
615
616
Furnaces having a Hearth inclined towards the Flue, 617. Furnaces
with the Hearth inclined towards the Fire-bridge, 617. Employing
Blast under the Grate. Corbin-Desboissières' Parabolic Furnace,
619. Staffordshire Furnaces, 620.
The Process in Reverberatory Furnaces
621
Manipulation, 621. Examples of Melting in Reverberatory Fur-
naces, 622.
Gas Reverberatory Furnaces
623
Moulding and Casting.
625
Operation, 625.
Materials used in Moulding.
625
Properties of the Materials, 625. Green Sand, 625. Analyses of
some varieties of Green Sand, 627. The Sand of the London
Basin, 627. Dry Sand, 628. Clay, 628. Metallic Mould, 629.
Blacking and Coal Dust, 629.
Patterns and Cores
630
Apparatus, Contrivances, Tools, and Implements used
in Moulding
632
Pits, 632. Drying Chambers, 634. Cranes, 634. Cases or Boxes
for containing the Sand Moulds, 635.
Methods of Moulding
635
Moulding in Green Sand
635
Moulding without the use of Boxes, 635. The Moulding in Boxes, 636,
Moulding in Baked or Used Sand
639
Dr. Mallet's Views on the Constitution of Cast-Iron,
and the Influence of Shape and Size of the Castings
upon the Physical Properties of the Iron
640
Moulding in Loam
660
Casting in Metal Moulds or Chills
667
CONTENTS.
xvii
Improvements in Moulding.
Machine Moulding, 668. Plate Casting, 668. Shell Casting, 669.
Finishing the Castings
Grinding, Polishing, Boring, &c., 670. Blackening, 670. Coating
with Oxide, 670. Welding of Cast-Iron, 671. Tempering, 671.
Bronzing, 672. Gilding, 672. Silvering, 673. Coppering, 675.
Tinning, 674. Enamelling, 675. Coating Iron with a Glass-like
Material, 684.
Production of Wrought-Iron
General Characters of Wrought-Iron, 685.
The Properties of Wrought-Iron.
668
669
685
686
Colour, 686. Texture, 686. Specific Gravity, 688. Tenacity, 688.
Hardness, 689. Behaviour of Wrought-Iron upon Heating, 689.
Behaviour of Wrought-Iron towards other Substances 690
Oxygen, 690. Nitrogen, 691. Sulphur, 691. Phosphorus, 691. Silicon,
692. Aluminium, Calcium, &c., 692. Tin, Antimony, Arsenic, 692.
Nickel and Cobalt, 692. Manganese, 692. Copper, 693.
Ascertaining the Quality of Wrought-Iron .
693
Mechanical Tests, 693.
Classification of Wrought-Iron according to certain
Characteristics.
694
Fine-grained Iron, 694. Fibrous Iron, 695. Red-Short Iron, 695.
Raw-Short Iron, 695. Cold-Short Iron, 695. Burnt Iron, 696.
Applicability of the Process, 697.
Methods of Making Wrought-Iron directly from the Ores 697
The German Process in Bloomeries
Catalan or French Process .
The Trompe, 699. The Catalan Forge in America, 700.
The Method used in Corsica and Italy
The Method used in East India and Finland
Newer Methods
•
Production of Wrought-Iron from Pig-Iron.
697
697
702
702
703
704
The Process, 704.
The Chief Methods of Producing Wrought-Iron from
Pig-Iron.
The Production of Wrought-Iron at a Glowing Heat.
Analyses of Malleable Cast-Iron, 705. The Bessemer Process, 706.
The Production of Malleable Iron in Fineries, 707. The Pro-
duction of Malleable Iron in Reverberatory Furnaces, 707. Com-
bined Reverberatory and Finery Process, 707. Comparison of the
Finery with the Puddling Process, 707. Combinations of the Finery
and Puddling Processes with different Re-heating Operations.
1. Production of Wrought-Iron in Hearth Fineries or
Open Fires
705
705
710
xviii
CONTENTS.
Materials used in the Finery Process.
Cast-Iron, 710. White Cellular Iron, 711. Flowery or Radiated
White Pig, 711. Spiegeleisen, 712. White Iron of the Regular
White Iron of the Irregular Process, 712. White
Chilled Iron of Radiated Texture, 712. The Varieties of Grey
Iron, 713.
Process, 712.
Preparation of the Pig-Iron
Preparation of the Iron in Blast Furnaces, 714. Raising the Pig-
Iron to a Glowing Heat, 715. Refining the Iron by Treating it
in the Liquid State in Hearths or Finery Fires, 715. Various
Plans for the Employment of Fluxes, 716. Mr. George Parry of
the Ebbw Vale Works, Purification of Iron by means of highly
Heated Steam, 718. The Refineries, 720. Analyses of Refinery
Slags, 723.
The Refinery Process, 723. Modification of the
Refinery Process, 724. Refining in Reverberatory Furnaces, 726.
Eck's Furnace, 726.
Fuel
Fluxes and Agents
710
714
735
736
Sand and Clay, 739.
Slags, 737. Analyses of Finery Cinders, 737.
Water, 739. Carbonate of Lime, 739. Schafhäutl's Powder, 740.
Apparatus used in the Fining Process.
The Finery Fires or Hearths, 740. The Construction of a Finery
Hearth, 743. The Tuyeres, 744. The Dimensions of Finery
Hearths, 745. The Blowing Machines, 746. The Quantities of
Blast required, 746. Hot Blast, 746. The Application of Steam,
The Lancashire Forge, 747-
747.
Apparatus for Re-heating the Balls or Blooms.
740
747
Open Fires, 748. Hollow Fires, 748. Re-heating Furnaces, 748.
Apparatus for Forging the Iron.
749
Lift Hammers, 750. Tilt Hammers, 752. Forge Hammers, 752.
The Methods of the Finery Process
754
Differences between the Methods, 754. Extremes and Varieties of
756
the Process, 755.
The German or Breaking-up Process.
Second Period:
First Period: Melting-Down of the Pig-Iron, 756.
the Fining Proper, 757- The Lump Fining, 758. The Breaking-
up Process, 758. A Combination of the Lump Fining and the
Breaking-up Process, 758. The Bohemian Process, 759. The
Rhonitz Process, 760. The French Process, 760. The Sulu
Process, 760. Half Walloon Process, 760. Third Period: Ham-
mering the Loupe, 760.
Walloon Process-Twice Melting-Down Process.
Eifel Walloon Process, 762. Swedish Walloon Process, 762. Eng-
lish Walloon Process: a. Lancashire Process, 763. b. South
Wales Process, 764. Styrian Walloon Process, 764.
761
CONTENTS.
xix
The Once Melting-Down Process
Modifications of this Process :-Austrian Slag Process, 765.
Styrian Process, 765. The Siegen Process, 766.
Methods requiring a Preparation of the Pig-Iron.
765
The
766
Different Modifications, 766.
Products of the Finery Process
767
The Puddling Process.
768
Advantages of the Process, 768. Its History, 768. Literature, 768.
Puddling Materials
769
Different Sorts of Pig-iron, 769.
Fettling, 771. Treatment of Manganiferous Pig-iron, 771.
Preservation of the Hearth
Fuel
•
771
I. In the Refining of Pig-Iron, 771. Gas, Coal, Coke, Charcoal, 771.
2. In the Puddling of Iron :-Coal, 772. Brown Coal, 772. Turf,
773. Wood, 773. Gases Produced with Wood, 775. Gases
Produced from Turf, 775. Brown Coal Gases, 776. Gases Pro-
duced from Coal, 777.
3. In the Re-heating of the Puddled Iron, 777. Coal, 777. Brown
Coal, 778. Turf, 778. Wood, 778. Wood Gases, 779. Char-
coal Gases, 779. Turf Gases, 779.
Fluxes.
a. In the Refining of Pig-Iron :-Refinery Slags, Fluor Spar, Lime,
779.
b. In the Puddling Process :-Slags, Hammer Slag, Rich and Pure
Iron Ores, &c., 780. Wrought-Iron Scraps, 780. Schafhäutl's
Powder, 780. Steam, 781.
c. In the Re-heating Process :-Sand, Water-Glass, 781.
Puddling Apparatus
The Plant of Puddling Works, 781.
Puddling Furnaces
Different Constructions, 782. Principal Parts of Puddling Fur-
naces :-Fire-Place, 782. Plane Grates more or less Inclined,
782. Clinker Grates, 783. Puddling Furnace at Königshütte in
the Hartz, 783. Puddling Furnace for the Production of Fine-
grained Iron, 787. Puddling Furnace for the Common Puddling
Process, 788. Puddling Furnaces with Brown Coal Firing, 788.
Step Grates, 789. Siemens's Regenerative Gas Furnace, 789.
The Advantages of Siemens's Furnace, 790. Faraday's Descrip-
tion of the Furnace, 790. The Ash-pit, Soo. The Furnace Hearth,
800. The Side Walls of the Hearth, Soo. The Fire-bridge, So1.
The Flue-bridge, 802. The Furnace Roof, 802. Double Fur-
naces, 802. Chimney, 805. Dimensions of Puddling Furnaces,
806. 'Tools used in the Puddling Process, S07.
Apparatus for Re-heating the Iron
Re-heating Furnaces, 807.
779
781
782
S07
XX
CONTENTS.
Machines for Forging and Condensing Iron.
Hammers, 813. Steam Hammer, 813. Stamp Hammer, 815.
Hydraulic Hammers, 815. Friction Hammer, 815. Brown's
Patent Bloom Squeezer, 815. Squeezers, 817. Rotary Squeezer,
818.
813
Rolls or Cylinders
818
Compound or Universal Rolling Mill, 824. Finishing Rolls for
Rails, 826. Rolls for Slitting Sheets of Metal, 826.
Shears
827
The Production of Wrought-iron in Puddling Furnaces:
The Puddling Process
828
Different Methods of the Puddling Process, 828. The Preparatory
Refining of the Pig-Iron, 828.
Dry Puddling in Furnaces with a Sole of Sand or of
Iron Plates
830
The Boiling or Wet Puddling Process.
Modifications of the Process, 832. Analyses Showing the Gradual
Modifications of the Pig-Iron during its Transformation into
Wrought-Iron, 833.
The Puddling Process aiming at the Production of
Fibrous Iron
Pig-Iron Employed, 833. Operations comprising the Process:
1. The Formation of the Hearth, 834.
2. Charging the Puddling Furnace, 834.
3. Melting-Down the Pig-Iron, 835. Chemical Reactions at the
Melting-Down, 836. Siemens's Experiments concerning the
Behaviour of Silicon and Carbon, 837-
4. Boiling and Stirring the Fused Mass, 842. Chemical Reactions,
842. Mechanical Appliances Substituting the Manual Puddling,
843.
5. Consolidation of the Iron into Masses or Balls fit for Hammer-
ing and Further Decarbonisation, 844. Different Manipulations
of this Part of the Process, 845.
6. Shingling the Balls, 847. Irregularities of the Normal Process,
848.
The Results of Different Puddling Works
•
a. Puddling Works using Mineral Coal as Fuel:-Staffordshire,
848. Scotland, 849. Cleveland and Durham, 849. Belgium,
849. Phoenixhütte in Ruhrort, 850. Königshütte in the Hartz,
850.
b. Puddling Works using Brown Coal as Fuel:-Leoben, 850.
Franzenshütte, 851.
c. The Puddling Process when Turf is used as Fuel :-Maximilians-
hütte, 851.
d. The Puddling Process when Wood is used as Fuel:-Suraham-
mer and Nyby in Sweden, 851.
832
833
848
CONTENTS.
xxi
e. The Puddling Process with Gases Produced with Brown Coal:-
Prävali, 851. Kerms, 852.
f. The Puddling Process with Turf Gases:-Nothburgahütte near
Freudenberg, 852. Buchscheiden near Klagenfurt, 852.
g. The Puddling Process with Wood Gases :-Zorge in the Hartz,
853. Lippitzbach, 853. Neuberg, 853.
h. The Puddling Process in Siemens's Regenerative Gas Furnace.
The Puddling Process for the Production of Fine-
grained Iron
Illustrations of
Its Differences from the Preceding Process, 858.
the Process:-Alvenslebenhütte in Silesia, 860.
Upper Silesia, 861. Low Moor, 861. Vierzon, 862.
Analyses of Puddling Cinder
Re-heating the Blooms, &c.
858
Pielahütte in
862
862
Iron No. 1, No. 2, &c., 863. The Piling, 863. Notice concerning
Rails, 863. Beattie's System of Piling, 864. Piling the Bars
crosswise, 864. Formation of Heavy Piles, 864.
Re-heating in Hollow Fires
Upper Hartz, 865.
Re-heating in Reverberatory Furnaces
Manipulation, S66. The Loss of Iron in the Operation of Re-heat-
ing, 867. Analyses of the Re-heating Cinder, S67. Illustrations
of the Process:-England, S6S. Karolihütte near Dernoe (Hun-
gary, 868.
Manufacture of Finished Wrought-Iron of Different
Kinds
Manufacture of Bar-Iron, S69. Manufacture of Plates and Sheets,
870. Manufacture of Tin Plate, 871. Manufacture of Wire, 874.
S65
866
869
LIST
OF
WOOD ENGRAVINGS.
COPPER.
FIGS.
I-3
7-9
Small Hearth for Refining Black Copper
Bredberg's Smelting Furnace
10-12 Carlberg's Smelting Furnace
13, 14
15, 16
17, 18
19-21
22-24
25-27
28, 29
30, 31
32, 33
34
35, 36
Bredberg's Furnace for Smelting Roasted Matt
Hearth for Refining Black Copper
Copper Smelting Furnace at Röraas
Mound for Roasting Copper Ores
Copper Smelting Furnace at Riechelsdorf
Furnaces at Oker
Refining Furnace at Oker
Reverberatory Furnace for Roasting Copper Ores
The Hearth of such a Furnace
•
Parkes's Double Roasting Furnaces
37-40 Calcining Furnace at Swansea
41, 42 Gerstenhöffer's Calcining Furnace
43-45 Spence's Calcining Furnace.
46, 47
Copper Smelting Furnace in England.
Combined Calcining and Smelting Furnace.
49-52 Copper Smelting Furnace at Mansfeld
48
►
•
•
PAGES.
58,59
86, 87
87-89
91
97
98
104
II2
•
125, 126
127, 128
137
138
138
140
142
150, 151
153, 154
162
163
196
53-55 Reverberatory Furnace at Mansfeld for the Concentration
56, 57
58, 59
60, 61
of Matt
Gas Reverberatory Furnace at Hettstädt for Refining Black
Copper
•
Copper Refining Furnace at Perm
Apparatus at Schmöllnitz for Precipitating Copper
62-64 Sinding's Apparatus for Producing Sulphuretted Hydrogen
Gas
198, 199
•
201, 202
224
250, 251
260
65
மம்ம்
66
67, 68
69
70, 71
IRON.
Open Heap for Roasting Iron Ores
Roasting Furnace without a Grate at Lerbach.
Roasting Furnace with a Plane Grate at Upper Silesia
Vordernberg
Siegen
•
359
365
368
•
369
•
370, 371
LIST OF WOOD ENGRAVINGS.
xxiii
FIGS.
72
73
74
75, 76
77,78
79-81
82
83, 84
356
85
∞ ∞ ∞ ∞ ∞ ∞¤·
86
87
88-96
98, 99
100
IOI, 102
103, 104
Roasting Furnace with a Grate of Conical Shape
Step Grate
at Gollrath
Norway and Sweden, &c. .
Gas Roasting Furnace at Dannemora
Hof
An Inclined Plane (Lift)
•
Jacob's Ladder, or Endless Chain System
Water Balance .
Hydraulic Lift at Derby
Pneumatic Lift in Upper Silesia
Iron Blast Furnace
•
of Newer Construction
Section of a Charcoal Blast Furnace with a Second Fore
Hearth.
Construction of Blauöfen.
Hearth Casing of a Mixture of Sand and Clay
•
PAGES.
372
373
374
376
•
378, 379
380-382
425
426
427
428
430
434-439
440
444
448
446
448
449
450
•
450, 454, 455
455
·
455, 456
457
457
457,458
105
тоб, 107
108
109-117
118
119-122
123
124
Apparatus for Collecting the Waste Gases from Iron Blast
Furnaces
Similar Apparatus at Siegen and in the Hartz
Cup and Cone Charger
Furnace Shafts for Easily Fusible Ores
Treating somewhat less Fusible Ores
Furnaces for Ores Difficult to Fuse
Furnace in Belgium
France
•
125-128 Furnaces in Germany and Sweden
129-131 Employed chiefly since the Application of Hot
•
458, 459
Blast
132
Alger and Abt's Furnace
460
133-138
139
Rachette's Furnace
Truran's Furnace
•
•
460-462
•
463
140-144
145
Shape of Blast Furnaces which have been Blown Out
Tuyeres for Charcoal Furnaces
•
475, 476
484
146, 147
485
148, 149
•
150-152
Calder
Tuyere for Coke Furnaces
Wasseralfingen Hot Blast Apparatus
153-155 Cowper's
Blast Furnace Showing the Ideal Zones
157-159 Charcoal Cupola Furnace at Lerbach
Coke Cupola Furnace at Berlin.
Königshütte.
Wasseralfingen
Gleiwitz
in England.
Maillard's Cupola Furnace
Movable Furnace
156
160
161, 162
163
164
165-167
168, 169
170
171
Boccard's Furnace
•
172-174
Hot Blast Apparatus.
175, 176
Schwarzkopf's Fan
•
177
178, 179
180, 181
182-192
•
•
•
·
•
•
•
•
513,514
•
•
515,516
517,518
520
593
594
594
595
595
596
597
598
598
•
боо
бол
602
619
620
193-202
203, 204
205, 206
207, 208
209, 210
Braithwaite and Ericson's Excentric Fan.
Corbin Desboissières Parabolic Reverberatory Furnace
for Re-melting Cast-Iron
Reverberatory Furnace at Berlin
Sections Showing the Crystalline Arrangement of Cast-
Iron
•
Drawings Explaining the Moulding in Loam
Blowing Machine (Trompe) used in Navarre
Catalan Forge
•
Refinery Fire in England, &c.
at Mariazell
642-645, 647, 656.
661, 662, 664 665
699
•
700, 701
721, 272
724, 725
xxiv
LIST OF WOOD ENGRAVINGS.
FIGS.
211-213
214-224
225, 226
227, 228
Double Refinery Fire at Mariazell
Eck's Gas Reverberatory Furnace
Finery Hearth
Hollow Fire
•
Puddling Furnace at Königshütte
•
Alvenslebenshütte for Fine-Grained Iron.
Fibrous Iron
for the Application of Turf as Fuel
Siemens's Regenerative Gas Furnace
Re-heating Furnace
at Creuzot
Dernoe
229, 230
Lift Hammer
231
Tilt Hammer
232
Forge Hammer.
233-238
239-241
242, 243
244
245-248
249-251
252-254
255, 256
257,258
259
260
261, 262
263
264
265, 266
267
268
269
270, 271
272, 273
Groebe's Re-heating Furnace
Steam Hammer.
Brown's Patent Bloom Squeezer
Squeezers
•
Rotary Squeezer
Rolls
Compound or Universal Rolling Mili
Rolls for Birkenshaw's Rails
Slitting Sheets
Shear.
•
Apparatus for Tinning Iron Plates
Drawing Wire.
·
PAGES.
726, 727
727, 729-733
•
741, 742
749
750
752
753
•
783-785
•
785, 786
•
787, 788
788
•
794, 796, 797
809
810, 811
811, 812
812
814
816
•
816, 817
818
822
825
2 567
2 2 2
826
827
827
872
•
874, 875
A PRACTICAL
TREATISE ON METALLURGY.
CHAPTER I.
COPPER.
PRINCIPAL ORES OF COPPER.
THE ores* containing this metal are found in different
geological formations; chiefly in crystalline slates and in
more modern rocks, up to the new red sandstone; in rare
cases in still more recent formations, as, for instance,
in Hungary, where copper ores occur in the triassic for-
mation, or on Lake Superior in North America, where
native copper is found in the alluvium. They are found in
veins, in beds, and as impregnations, associated with quartz,
heavy spar, fluor spar, chlorite, amphibole, pyroxene, &c.
Frequently they occur in admixture with other ores, chiefly
with lead, zinc, and silver ores; also with iron pyrites,
cobalt, nickel, and tin ores. The copper slates at Mansfeld
form an interesting cupreous formation in a geological point
of view. These curious strata of bituminous schist are among
the most ancient, containing the exuviæ of organised bodies
not Testaceæ. From among their tabular slabs is extracted
the vast multitude of fossil fish which have rendered the
* COTTA'S Erzlagerstätten, i., 20, 49; ii., 600. LAMBORN, Metallurgy of
Copper, 1860. p. 44. Rivor, Métallurgie du Cuivre, 1859, p. 23. Dr. URE'S
Dictionary of Arts, 1867, i., p. 870.
VOL. II.
B
2
COPPER.
cantons of Mansfeld, Eisleben, Ilmenau and other places
in Thuringia and Voigtland so celebrated; many of the
fish are transformed into copper pyrites. Here, also,
have been found the fossil remains of the lizard family,
called monitors.
A judicious administration is so necessary for the pros-
perity of these mines, that the study of a thin layer of slate
in this formation, of which one hundred pounds commonly
contain but one pound and a half of copper, occasionally
argentiferous, led to the establishment of smelting works,
worked now for several centuries, of the greatest importance
to the adjoining country.
The frequent derangements of this deposit soon led skilful
directors of the underground operations to study the order
of the super-position of the accompanying rocks. From
these observations a system of facts resulted which has
served to guide, not only the miners of Mansfeld, but of
great part of Germany, and of other countries where the
same series of rocks, forming the envelope of the cupreous
schists, occur in the same order of super-position.
Copper ores occur in many varieties. Scheerer of Freiberg
has arranged them as follows:-
1. Copper Glance (Kupferglaserz), Cu,S, containing 79.7
per cent of copper.
2. Copper Pyrites, Cu₂S, Fe,S,, containing 34.8 per cent.
of copper.
3. Variegated Copper Ore, 3Cu₂S, Fe,S,, containing 55'7
per cent of copper.
4. Fahlerz, 4(Cu₂S, FeS, ZnS, AgS), (SbS,,AsS,), containing
14-41 per cent of copper.
5. Red Copper Ore, Cu₂O, containing 88.5 per cent of
copper.
6. Malachite 2CuO,CO₂+HO, containing 57'4 per cent
of copper.
7. Azurite 2(CuO,CO₂)+CuO,HO, containing 55'3 per
cent of copper.
The ordinary copper ores are sulphides and oxides, the
sulphides predominating. Copper is also found combined
with arsenic, selenium, antimony, iron, silver, and various
PRINCIPAL ORES OF COPPER.
3
acids. The minerals containing copper may be classified
according to their state of combination as follows:-
1. Native Copper, which in a few localities only occurs in
large quantities. It is found in Peru and Chili in small
grains, together with other copper ores and earths, containing
about 70, at the highest 90, per cent of copper; it is called
Copper Sand or Copper Barilla, and is smelted chiefly in
England and France. A sample of native Corocoro copper
contained-
Cu
SiO3
Fe₂03.
MgO,CO, and CaO,CO,
S
73.81
20'94
136
2:48
trace
A sample of native copper from Brazils contained—
Cu
Ag
Au
Fe
99'56
0°30
0*08
ΟΙΟ
The most remarkable masses of native copper hitherto
discovered are those found in the mines of Lake Superior,*
some of which exceeded 150 tons in weight; copper occurs
here in trap rock, in the form of injected veins, and is
associated with quartz, chlorite, calc spar, epidote, prehnite,
&c. In 1854, a mass of native copper weighing about 500
tons was found in Minnesota; its division occupied forty men
for twelve months. It consisted, according to Haute-
feuille,t of-
Cu
Ag
Hg.
Gangue
69 280
5'453
0'019
25'248
Crystals of native silver sometimes occur in native copper
otherwise free from silver, so intimately mixed that neither
a dressing operation nor a smelting process will effect their
P. 2.
* B. u. h. Ztg., 1856, p. 261; 1858, p. 47; 1861, p. 305. BERGGEIST, 1858,
RIVOT, Métallurgie du Cuivre, 1859, p. 24. Bulletin de la Société de
l'Industrie Minérale, vi.
+ BERGGEIST, 1857, p. 160. B. u. h. Ztg., 1859, P. 447.
R 2
4
COPPER.
separation, which must be performed by a treatment in the
wet way.
Native copper sometimes occurs at Lake Superior, finely
disseminated through quartz; it then requires dressing.*
Native copper is also found in New England, New Jersey,
Pennsylvania, Virginia, and North Carolina. It occurs
abundantly in the mines of Siberia, Tourinski, Hungary,
Fundo-Moldavia in Gallicia, Fahlun in Sweden, and
Cornwall.
The copper produced from native copper ranks among the
best in the market.
2. Sulphides of Copper.-These frequently contain suffi-
cient silver and gold to be profitably extracted, and they
yield copper varying in purity according to the foreign im-
purities of the ores. Fieldt found bismuth in almost all
ores containing sulphide of copper, and in all such Cornish
ores. The dressing of these ores increases in difficulty the
greater the specific gravity of the foreign associates. Sella‡
has used a magnet with advantage to extract magnetic iron
from ores containing copper pyrites. The following are the
ores belonging to this class.
a. Copper Pyrites, Cu,S,Fe,S,, containing 344 per cent of
copper. This is the most common sulphide used in smelting
works; it occurs in large masses and in extended veins in
primitive and transition districts, and is commonly accom-
panied by grey copper, sulphide of iron, sparry iron, sul-
phides of lead and zinc. It yields a copper varying in
purity according to the quantity and quality of the associates.
The copper produced in Norway, Sweden, and Dillenburg, &c.,
from pyrites, is of excellent quality. The pyrites occurring
in Sibenbürgen and Fahlun is sometimes auriferous, and in
rare cases argentiferous.
Copper pyrites is the principal ore of the English smelting
works and of the North American works along the Atlantic.
coast. The largest amount of copper is manufactured from
this ore.
B. u. h. Ztg., 1861, p. 306.
+ Polyt. Centr., 1862, p. 410.
B. u. h. Ztg., 1862, p. 101.
COPPER ORES CONTAINING ANTIMONY, ETC.
5
b. Copper Glance, Cu₂S, containing 79'7 per cent of copper,
and usually silver and iron. This is one of the varieties.
frequently met with in copper ore veins; it is often found in
masses. It is profitably mined in Cornwall, where it occurs
together with red iron ore, and sometimes with copper
pyrites; also in Tuscany, Chili, South Australia, Cape of
Good Hope, and in Connecticut, New York, Virginia, Mary-
land, and other of the United States.
c. Variegated Copper Ore (Buntkupfererz), 3Cu₂S,Fe₂S¸,
containing 55'7 per cent of copper. It seldom occurs in
masses, and is usually associated with copper glance, and
also with copper pyrites, as in Cornwall, Tuscany, &c.
d. Indigo Copper (Kupferindig), CuS, containing 66'5 per
cent of copper. It is found in large quantities in Chili only.
e. Silver-Copper Glance, Cu,S,AgS, containing 53 per cent.
of silver, 31 per cent of copper, and sometimes a little iron.
It is found associated with iron pyrites in Siberia, Chili, Peru,
and in the silver mines of Arizona.
3. Copper Ores containing Antimony and Arsenic.—
These ores mostly require complicated smelting operations.
in order effectually to separate the antimony and arsenic,
which constitute the worst impurities, rendering it scarcely
possible to produce a very good copper from these ores.
Arsenic is more easily separated than antimony; these ores
sometimes contain an amount of silver which increases their
value, but they then require very complicated smelting pro-
cesses for the extraction of the silver. As some of these ores
have a low specific gravity, and often contain a considerable
amount of silver, they must be dressed with the utmost
care.
a. Fallow Ores, 4(Cu₂S, FeS, ZnS, AgS, HgS), (SbS¸, AsS₂).
-Some varieties contain both antimony and arsenic, others
one only of these substances. Their average amount of
copper is from 30 to 48 per cent. Those which are very rich
in silver (31 per cent) contain only 15 per cent of copper,
whilst ores free from silver contain the highest amount of
copper. Some of these ores occurring in Hungary, Tyrol,
and Tuscany contain from 0'52 to 17:27 per cent of mercury,
which is sometimes extracted.
6
COPPER.
b. Bournonite, зCu₂S, SbS¸ + 3(PbS, SbS₂), containing
12.7 per cent of copper, 41'7 per cent of lead, and sometimes.
a little iron. It is free from silver, and only occurs in small
quantities together with other copper ores, as in Cornwall,
Devonshire, on the Hartz, Chili, &c.
c. Enargite, 3Cu₂S,AsS,, containing 48.3 per cent of copper,
and sometimes also ZnS, FeS, and SbS,. It is found in
large quantities in the Cordillera mountains in Peru at a
height of 15,000 feet.
4. Oxidised Copper Ores.-When found in sufficiently
large masses, which very seldom occurs, these require very
simple metallurgical processes, and yield an excellent
copper.
a. Red Copper Ore, Cu,O, containing 88.8 per cent of copper.
It is found in large masses in South Australia (the celebrated
Burra-Burra mine), in Spain, Peru,* &c. It also occurs in
Tuscany, New Jersey, and at Lake Superior.
b. Black Oxide of Copper, CuO, containing 79.8 per cent of
copper, and generally also a certain amount of oxide of iron
and manganese. It occurs in large quantities at Copper
Harbour on Lake Superior, and at the Burra-Burra mine.
Besides these oxides, an admixture of oxide of copper with
hydrated oxide of iron is frequently associated with copper
ores, but in small quantities only.
5. Copper Salts. These ores are mostly a secondary
formation of other copper ores, and usually occur only on
the head of lodes. The quality of the copper produced
from them depends on the foreign substances with which
the original ores were associated. Sometimes they contain
silver and gold, as the green and blue malachites of Algeria,†
or only silver, as the blue malachite of Spain.‡
2
a. Green Malachite, 2CuO, CO₂+ HO, containing 57'3 per
cent of copper, is found in large masses in Ural, South
America, South Australia, in the Portuguese colonies on the
coast of Africa, &c.
* B. u. h. Ztg., 1859, p. 242.
+ Ibid., 1859, p. 242.
Ibid., 1859, P. 242.
COPPER SALTS.
7
2
b. Blue Malachite (Azurite), 2CuO, CO₂ + CuO, HO, with
55'1 per cent of copper, is only occasionally found in large
quantities, as at Chessy, near Lyon, in Bohemia, and at
Kanmantoo in South Australia. At the latter place it con-
tains an injurious amount of antimony and bismuth. A
brown anhydrous carbonate of copper, associated with some
malachite and red iron ore, is imported into Europe from
India.
3
c. Sulphate of Copper, Blue Vitriol, CuO, SO, + 5HO,
containing 25 per cent of copper; it is imported into South
Wales from Mexico and South America; it is also frequently
formed artificially in copper ore mines, as in Anglesea,
Wicklow, Hungary, and Rammelsberg on the Hartz.
39
d. Silicate of Copper, Dioptase, 3CuO, 2SiO3 + 3HO, con-
taining 399 per cent of copper, and sometimes also small
quantities of Fe₂O,, Al₂O, CaO, and MgO. Some varieties
are Kupferblau, CuO, SiO3, + HO, containing 36.3 per cent
of copper, and Kupfergrün, 3CuO, 2SiO3 + 6HO, containing
35'7 per cent of copper. They are well adapted to the
smelting process, rendered easily fluid, and occur in large
quantities on the Ural, and in South and North America.
e. Chloride of Copper (Atacamite), CuCl+3CuO+3HO, con-
taining 59'4 per cent of copper, and 2(CuCl + 3CuO) + 9HO,
containing 56 per cent of copper. This is imported in con-
siderable quantities from Chili and other parts of the west.
coast of South America into South Wales, where it is worked,
yielding an excellent copper.
f. Phosphate of Copper, containing a variable amount of
hydrated oxide of copper, and from 50 to 56 per cent of
copper. It occurs in Chili, Bolivia, Hungary, &c. The
phosphorus it contains, like antimony and arsenic, is hurtful
to the resulting copper.
g. Arseniate of Copper, also containing a variable amount
of hydrated oxide of copper, and from 30 to 35 per cent of
copper. This is imported from Chili to South Wales. The
usual impurities are,-oxide of iron, quartz, heavy spar, and
calc spar, and sometimes oxide of zinc.
8
COPPER.
The most various copper ores, of both foreign and home
production, are smelted in the copper works of South
Wales,* in the South of France,t at Boston, &c.
+
The following account of the English copper veins is
given in Dr. Ure's "Dictionary of Arts," i., p. 872-
"The deposits of copper in Cornwall occur as veins in
granite, or in the schistose rocks which surround and cover
it; and hence, the Cornish miners work mostly in the granite
and clay slate, the former of which, when metalliferous, is
usually in a coarse and often a disintegrated state; this they
call growan, the latter killas.
(C
Copper veins are abundant in killas and more rare in
granite, but most numerous near the line of junction of the
two rocks. The different kinds of mineral veins in Cornwall
may be classed as follows:
1. Veins of elvan, elvan courses, or elvan channels.
2. Tin veins, or tin lodes.
3. Copper veins, generally running east and west.
4. Second system of copper veins, or contra lodes.
5. Crossing veins; cross courses.
6. Clay veins, called cross flookans or slides.
"The width of these veins does not often exceed 6 feet, though
occasional enlargements to the extent of 12 and more feet
sometimes take place. Their length is unknown, but one
explored in the United Mines has been traced over an extent
of seven miles. The gangue of these veins is generally
quartz, either pure or mixed with green particles analogous
to chlorite. They contain iron pyrites, blende, sulphide,
and several other compounds of copper, such as the car-
bonate, phosphate, arseniate, chloride, &c. Most of the
copper lodes are accompanied by small argillaceous veins,
called by the miners flookans of the lode.
These are
often found on both sides of the vein, so as to form cheeks or
walls.
"When two veins intersect each other, the direction of the
one thrown out becomes an object of interest to the miner.
* PERCY'S Metallurgy, i., 322. B. u. h. Ztg., 1862, No. 40.
↑ B. u. h. Ztg., 1859, p. 242.
Ibid, 1859, p. 309.
THE ENGLISH COPPER VEINS.
9
"The copper mines of Anglesea, those of North Wales, of
Westmoreland, the adjacent parts of Lancashire and Cumber-
land, of the south-west of Scotland, of the Isle of Man, and
of the south-east of Ireland, also occur in primitive or transi-
tion rocks. The ores lie sometimes in masses, but more
frequently in veins. The mine of Ecton in Staffordshire,
and that of Cross-gill-burn near Alston-moor in Cumberland,
occur in transition or mountain limestone.
"The copper ores extracted both from the granitic and
schistose localities, as well as from the calcareous, are uni-
formly copper pyrites more or less mixed with mundic; the
red oxide, carbonate, arseniate, phosphate, and chloride of
copper, are very rare in these districts.
The working of copper mines in the Isle of Anglesea may
be traced to a very remote era. It appears that the Romans
were acquainted with the Amlwch mine near Holyhead;
but it was worked with little activity till about seventy years
since. This deposit lies in a greenish clay slate, passing
into talc slate; a rock associated with serpentine and eupho-
tide. The veins of copper are from one to two yards
thick, and converge towards a point where their union forms
a considerable mass of ore. On this the mine was first
commenced by an open excavation, which is now upwards
of 300 feet deep, and appears from above like a vast funnel.
Galleries are formed at different levels upon the flanks of
the excavation to follow the several smaller veins, which
run in all directions and diverge from a common centre like
so many radii. The ore receives in these galleries a kind of
sorting, and is raised by means of hand-windlasses to the
summit of a hill, where it is cleaned by breaking and
jigging.
"The water is so scanty in this mine that it is pumped up
by a small steam engine. A great proportion of it is charged
with sulphate of copper. It is conveyed into reservoirs
containing pieces of old iron; the sulphate is thus decom-
posed into copper of cementation. The Anglesea ore is
poor, yielding only two or three per cent of copper; a portion
of its sulphur is collected in roasting the ore.
ΙΟ
COPPER.
"The copper mines, now so important, were so little
worked until a recent period, that in 1799 we are told, in a
report on the Cornish mines, 'it was not until the beginning
of the last century that copper was discovered in Britain.'
This is incorrect, for in 1250 a copper mine was worked
near Keswick, in Cumberland. Edward III. granted an
indenture to John Ballanter and Walter Bolbolter for
working all mines of gold, silver, and copper;' but that the
quantity found was very small is proved from the fact that
Acts of Parliament were passed in the reigns of Henry VIII.
and Edward VI. to prevent the exportation of brass and
copper, lest there should not be metal enough left in the
kingdom fit for making guns and other engines of war, and
for household utensils; and in 1665, the Calamine works
were encouraged by the government as the continuing these
works in England will occasion plenty of rough copper to be
brought in.'
"At the end of the seventeenth century, some gentlemen.
from Bristol made it their business to inspect the Cornish
mines, and bought the copper ore principally for £2 10s. per
ton, seldom giving more than £4 a ton.
"In 1700, one Mr. John Costor introduced an hydraulic
engine into Cornwall, by which he succeeded in draining the
mines, and he taught the people of Cornwall also a better
way of assaying and dressing the ore.'
6
"The value and importance of copper mines since that
period has been regularly increasing."
SYNOPSIS OF PROCESSES FOR THE EXTRACTION OF
COPPER FROM ITS ORES.
The extraction is mostly effected in the dry way, either by
a smelting process in cupola furnaces (German process), or by
treating the ores in reverberatory furnaces (English process).
The wet way is employed when the ores are too poor for pro-
fitable smelting,
Whether the German or English process is to be employed
depends chiefly on the following conditions:-
PROCESSES FOR THE EXTRACTION OF COPPER.
II
a. The price of the Smelting Materials, chiefly the fuel.
-The process in reverberatory furnaces consumes much fuel,
(from 16 to 18 parts of coal to I part of copper), and can
therefore be applied only where good and cheap coal is
attainable. Cupola furnaces require good and cheap char-
coal and coke, but they concentrate the heat better, and
thus economise the fuel. The wet way is applicable when
fuel is too expensive, as in Riotinto aud Huelva in Southern
Spain. The English process does not invariably give good
results, for instance in France, at Boston,† in Isabellenhütte,
in Nassau, &c., owing to the very great consumption of fuel,
which is at least three times as large as in the German pro-
The advantages afforded by the process in England
are founded less on a cheaper production than on the greater
ability and facility of buying ores and selling the produced
copper which the English copper smelters possess.
cess.
*
b. The Quality of the Ores.-Rich and pure ores may
be worked in either smelting apparatus with equal facility.
The cupola furnace is preferable for treating poorer ores, as
it permits of a more complete extraction of the copper, and
is also preferable for the treatment of very rich pure ores, as
it effects a quicker extraction; the richer the ores are, the
less fuel is used in proportion to the copper produced. In
England, from 16 to 18 tons of coal are used for the production
of I ton of copper, therefore the English process cannot
be introduced into the south of France, where the price
of coal is three or four times as high, and the ores under
treatment only half as rich as those worked in England.
The reverberatory furnace is preferable, if the ores are of
a variable composition, embracing sulphides and oxides with
different gangues and impurities, as the intelligent smelter
is able to regulate the more open process if variations in the
quality of the ore require it. If the ores are of a constant.
quality they allow of a regular and continuous process in
cupola furnaces.
B. u. h. Ztg., 1859, p. 251.
+ Ibid. 1859. P. 365.
I2
COPPER.
When treating pure ores in cupola furnaces, a quicker and
better yield of copper results, and wages and fuel are better
economised than when treating these ores in reverberatory
furnaces. The presence of lead, antimony, arsenic, and tin
in the ores makes the production of good commercial copper
difficult in both kind of smelting apparatus, and necessitates
more expensive operations. A smaller amount of antimony
and arsenic is more completely removed by the roasting and
reduction processes required for treatment in cupola fur-
naces than by working such ores in reverberatory furnaces,
where only the roasting processes act in volatilising the
arsenic and antimony. A larger amount of these substances
makes the production of a copper of middling quality in
cupola furnaces possible only with a great loss of copper. But
this may be done by the reverberatory furnace process, though
requiring more operations. Arsenic is less injurious than
antimony, as the arseniates formed by the roasting process
are easier to decompose, and as the arsenic combined with
copper may be more easily extracted from it by volatilisation.
Gold and silver may sometimes be profitably extracted.
c. The Time required for the Extraction of the Copper.
The more quickly the copper is produced and sold the
higher will be the profit on the working capital. A quicker
extraction also allows the smelter better to avail himself of
the favourable turns of the market. We have before men-
tioned that cupola furnaces effect the extraction.
quickly in proportion to the purity of the ores.
more
d. The Extent of the Production.-Reverberatory fur-
naces are preferable for larger, and cupola furnaces for
smaller, productions. According to Percy, copper works
carried on with reverberatory furnaces must produce at least
1,100 tons of copper a year to be profitable.
The process in reverberatory furnaces, therefore, is chiefly
adapted to sufficiently cheap ores, varying in quality, con-
taining a medium amount of copper, and when coals are very
cheap, the building materials for the furnace are refractory, and
a large production intended; smelting copper ores in rever-
beratory furnaces requires less skill than smelting lead ores.
REDUCTION IN THE DRY WAY.
13
The chief modes at present adopted for extracting copper
from its ores may be classified under the two following
divisions:
I. REDUCTION IN THE DRY WAY.
1. Sulphuretted Ores and Products.
These may be treated—
a. In Cupola Furnaces.-In this case the ores and pro-
ducts are repeatedly roasted in heaps or mounds, and after-
wards submitted to a reducing and purifying smelting in
cupola furnaces for the production of black copper. At this
smelting coal is the chief reducing agent. The resulting
black copper is refined in small hearths or in blast rever-
beratory furnaces.
If the ores are free from silver and gold they are smelted
either in sump furnaces, as in Atvidaberg, Fahlun, Röraas,
Szaska, and Agordo, or in cupola furnaces with two open
eyes, which are adopted in Sternerhütte, Dillenburg, Upper
Hartz, Kupferberg, and Riechelsdorf; in India and Japan
these ores are smelted in crucible furnaces. If silver and
gold are contained in the ores they are extracted either direct
from the ores and the matt, as in Nagybanya, Fahlun, and
Müsen, or from the matt only, as in Oeblarn ; or this extrac-
tion is effected from the black copper; the latter method is.
adopted in the Hartz and at Sziklowa.
b. In Reverberatory Furnaces.-The ores and matt are
roasted either in mounds or in reverberatory furnaces, and
smelted in reverberatory furnaces for the production of black
copper, the sulphur serving in the latter process to reduce
the oxides formed in the roasting process; the black copper
is then refined in air reverberatory furnaces. The ores
treated are mostly free from gold and silver. This process
is used in South Wales, Dillenburg, Elbkupferwerk, Duis-
burg, Freiberg, Bendorf, and Kaafjord.
c. By a combined Smelting in Cupola and Reverbera-
tory Furnaces.-The manner in which this is effected
varies in different localities. At Mansfeld the copper slate
is smelted in cupola furnaces, the resulting matts are con-
centrated in cupola and reverberatory furnaces, the black
14
COPPER.
copper is smelted in cupola furnaces, then the first refining
is effected in small hearths, and the final refining in rever-
beratory furnaces.
At Schmöllnitz and Altwasser, in Upper Hungary, the
production of black copper from fallow ores is effected in
cupola furnaces; the residues of the amalgamation of the
black copper are smelted in reverberatory furnaces, and the
resulting black copper is refined in blast reverberatory fur-
naces.
At Schemnitz, Kremnitz, Neusohl, and Tajova, in Lower
Hungary, the copper-lead matts are concentrated in cupola
furnaces. The resulting concentrated matt, after being
treated by Augustin's process, is converted into black copper
by treatment in a reverberatory furnace; the black copper is
refined in a blast reverberatory furnace.
At Freiberg and Grünthal, copper-lead matt is concentrated
in cupola and reverberatory furnaces; the desilverised matt
is worked in reverberatory furnaces, and the black copper is
refined also in reverberatory furnaces.
At Boston and the copper works on the Rhone, the ore,
matt, and black copper are smelted in cupola furnaces, and
the black copper is refined in reverberatory furnaces.
At Bogoslowsk, ore and matt are smelted in cupola furnaces,
black copper in blast reverberatory furnaces, and the refining
is effected in reverberatory furnaces.
2. Oxidised Ores and Products.
These are treated either in cupola furnaces, as in Chessy
and Perm, or in reverberatory furnaces.
3. Native Copper.
At Grünthal this is smelted in cupola furnaces, and at
Detroit, in France, in reverberatory furnaces.
II. REDUCTION IN THE WET WAY.
This treatment embraces first the decomposition of
cupreous solutions by iron, as in Anglesea, Schmöllnitz,
Rammelsberg, Wicklow, Maidenpek, and Riotinto.
REDUCTION IN THE WET WAY.
15
Oxidised Ores are treated-
a. By decomposing a solution of chloride of copper by
means of iron. This may be effected, as in Alderley Edge,
Linz, and Stadtbergen, by dissolving the substance in liquid
muriatic acid, or by treating it with a liquid containing
chloride of iron, or lastly, by heating the ores with con-
centrated muriatic acid, chloride of iron, or chloride of
magnesium.
b. By decomposing a solution of sulphate of copper by
means of iron. For this purpose the ore is treated either
with sulphuric acid or with a solution of sulphate of iron.
c. By decomposing an ammoniacal solution containing
copper.
d. By decomposing a solution of sulphite or hyposulphite
of copper by means of sulphide of sodium.
Sulphuretted Ores are treated—
a. By decomposing a solution of chloride of copper by
iron, sulphide of calcium, or lime.
For this purpose the roasted ores are treated either with
liquid muriatic acid at the common temperature (seldom at the
roasting temperature), or with muriatic acid in the form of
gas. In some cases raw pyrites is roasted with common
salt, or the pyrites is first subjected to an oxidising roasting
and afterwards chlorinated.
b. By decomposing a solution of sulphate of copper by
iron or sulphuretted hydrogen. At Schmöllnitz and Riotinto
the disintegrated pyrites is lixiviated with water to obtain a
solution of sulphate of copper. At other places the ores are
roasted in heaps, or in cupola or reverberatory furnaces, and
afterwards lixiviated with water. According to Dähne's
method, the roasted ores are heated with sulphate of iron
and then lixiviated. Levis and Robert treat the roasted
ores or matt with liquid sulphuric acid; this has also been
used in the form of gas.
Black Copper is treated by dilute sulphuric acid in the
presence of atmospheric air, as the ores treated in Mansfeld
are nickeliferous, and those at the Lower Hartz auriferous
and argentiferous.
16
COPPER,
DIVISION I.
REDUCTION OF COPPER ORES AND PRODUCTS IN THE
DRY WAY.
I. Sulphuretted Ores and Products.
The smelting processes for sulphuretted ores are based
upon the chemical behaviour of copper with regard to its
affinity to sulphur and oxygen; its affinity to sulphur being
greater than that of its associated metals, except gold and
silver, and its affinity to oxygen being less. These processes
are performed either in cupola or in reverberatory furnaces,
and they must be repeated more or less often according to the
amount of foreign substances present; if the ores contain
enough gold, silver, or lead to be worth extracting, some
special processes are introduced for that purpose. Gold and
silver are extracted either direct from the ores or from the
intermediate products, such as matt, black copper, and speiss,
and by either the wet or the dry way, whilst plumbiferous
copper ores are always first treated for the extraction of the
lead, and the resulting copper matt is passed on to the
copper smelting process. Mercury in fallow ores is some-
times extracted during the roasting process.
a. Reduction of Sulphuretted Substances in Cupola
Furnaces. This is usually effected by first roasting the
ores, thus volatilising part of the sulphur, antimony, arsenic,
&c., and transforming part of the contained metals into
oxides, sulphates, antimoniates, and arseniates, whilst some
of the ore remains undecomposed. The roasted mass is then
submitted to a reduction smelting in cupola furnaces (raw
smelting), with an addition of suitable fluxes when they are
not contained in the ores. This smelting first reduces the
oxide of copper to metallic copper; it then transforms the
greater part of the metallic sulphates into sulphides,
sulphurous acid being volatilised; these sulphides combine
with the metallic copper formed, and also with any sulphides
not decomposed by the roasting process, forming raw matt.
The metallic antimoniates and arseniates are reduced to
TREATMENT OF SULPHURETTED SUBSTANCES.
17
antimonides and arsenides (speiss), while part of the anti-
mony and arsenic is volatilised. The foreign oxides (chiefly
oxide of iron) after being reduced to protoxides, combine with
the earths which are contained in the ores, or added as
fluxes, forming slags.
The greater part of the foreign substances, antimony,
arsenic, and (chiefly) sulphur, are then extracted by repeated
roasting and reducing smeltings (concentration smeltings), by
which they are either volatilised or scorified; and at the
last of these concentration smeltings metallic copper results,
called blister, coarse, or black copper, from the dark colour which
is produced by a coating of oxide; besides this copper, a
small quantity of matt is produced. The coarse copper still
contains more or less of foreign substances, iron, lead, zinc,
antimony, arsenic, &c., from which it is separated by an
oxidising smelting; the foreign substances having a greater
affinity to oxygen than copper are oxidised at this smelting,
and are then either scorified or volatilised. If this oxidising
process is carried on sufficiently far to extract all the
foreign substances, a great deal of sub-oxide of copper
is also formed, which is partly scorified and partly absorbed
by the resulting copper, impairing its ductility. A quick
reducing smelting converts the sub-oxide of copper into
metallic copper, and renders the copper ductile.
Both smelting processes may be combined by refining black
copper in reverberatory furnaces.
It has been before mentioned that the processes become
much more complicated if the copper ores contain enough
silver and gold to be worth extracting; moreover, the copper
resulting from such ores is of inferior quality.
The following operations may be required in smelting
copper ores free from silver and gold in cupola furnaces :-
1. Roasting the ores.
2. Smelting the roasted ores for the production of raw
matt.
3. Roasting the raw matt.
4. Smelting the roasted raw matt for concentration:
5. Roasting the concentrated matt.
VOL. II.
C
18
COPPER.
6. Smelting the latter, when roasted, for the production
of black copper.
7. Oxidising smelting of the black copper.
8. Reduction smelting of the black copper.
9. Refining the black copper.
10. Working up the cupriferous residues of the pre-
ceding processes.
Roasting the Ores.-This process partly effects the re-
moval of all volatile substances, such as sulphur, antimony,
arsenic, bitumen, water, &c., either by the application of a
high temperature alone, or by an admission of atmospheric
air at the same time, and on the other hand it aims at an
oxidation of the metals contained in the ore, to fit them for
scorification at the raw smelting process; also, small
particles of ore are caked together by the roasting process,
and this is mostly advantageous at the subsequent smelting.
The purity of the ore, and chiefly its amount of sulphur,
indicate how far the roasting may be carried on for pro-
ducing a metal of good quality with as small a loss of copper
as possible. The quality of copper produced from ores con-
taining arsenic and antimony is better, the less the first
roasting is carried on and the oftener the roasting and
reduction smelting of the matt is repeated, as the foreign
substances then have more opportunity of volatilising and
scorifying; but this repetition of the processes is limited by
their great expense.
It is, therefore, bad policy to roast impure ores until,
owing to a want of sulphur at the raw smelting, black
copper results instead of matt, as in this case the black
copper contains a great part of the foreign impurities; and,
at the same time, a loss of copper by scorification will take
place. Such an impure copper can be refined only with
great difficulty and loss, and will never yield a product of
good quality. The roasting of pure ores may be carried on
farther without fear of injuring the quality, but it also causes
a loss of copper by scorification.
When working copper ores containing much antimony and
arsenic, the formation of a cupriferous speiss at the raw
smelting is unavoidable, even when they are roasted but
ROASTING THE ORES.
19
slightly, as antimony and arsenic behave like sulphur. The
extraction of these substances, and also the decomposition
of zinc blende, may be facilitated by mixing some
small coal with the roasting mass, the coal decomposing
the formed antimoniates, sulphates, and arseniates.* It
is also advisable to mix such ores intimately with iron
pyrites, as this causes the antimony and arsenic to vola-
tilise as sulphides.
The suitable amount of roasting must be ascertained by
the smelting results, as no judgment can be formed from
the appearance of the roasted mass.
The chemical reactions at the roasting process vary
according to the state in which the ores are roasted, whether
pulverulent or in fragments, and also according to the
quantity of foreign substances contained in the ores.
If the ore consists of copper glance (Cu,S), sulphurous
acid and sub-oxide of copper will be formed at the com-
mencement of the roasting, and afterwards sulphate of
copper; the oxidation of sub-oxide of copper may be effected
either by the oxygen of the atmosphere or by the sulphuric
acid formed. As long as sulphurous acid is developed, the
whole of the sub-oxide of copper cannot be transformed into
oxide, as the oxide is again reduced to sub-oxide by the acid.
At a higher temperature, sulphate of copper is decomposed
into sulphuric acid and oxide of copper, and the sulphate of
copper, together with the oxygen of the air, transform the
sub-oxide of copper present into oxide. Oxide of copper is
therefore the product, if the roasting process is well carried
on. If the temperature is raised too quickly, a smelting
and an incomplete oxidation of the sulphide of copper will
take place, by which a mixture of Cu₂O, CuOSO,, Cu₂S, and
Cu results from the reaction of oxidised substances upon
the sulphides. An admixture of sulphide of iron facilitates
the roasting of the sulphide of copper, forming sulphuric
acid more abundantly. A mixture of sulphide of iron and
copper may be so roasted as to produce a mixture of oxide of
iron, oxide of copper, and sulphate of copper, as the sulphate
*
Bgwkfd., xiii., 401.
B. u. h. Ztg., 1859, P. 70.
2 C
20
COPPER.
of copper decomposes at a higher temperature than sulphate
of iron.
Copper pyrites, (Cu₂S, Fe₂S₂) at a dull red heat, when
repeatedly stirred, is transformed into a powder, consisting
of peroxide of iron, sulphate of iron, and sulphate of copper,
which is friable and dark red-coloured when cold; sulphurous
acid is evolved at the same time. At a little lower red
heat the proto-sulphate of iron is transformed into per-
sulphate of iron, which then loses the greater part of its
sulphuric acid, and a mixture results consisting of free oxides.
of iron and copper, a great deal of sulphate of copper, and a
little sulphate of iron. If the temperature is still increased,
oxides of iron and copper only will remain.
Copper ores, chiefly containing copper pyrites, (Cu₂S,Fe₂S₂)
and iron pyrites, (FeS2), form sulphurous acid and protoxide
of iron at the beginning of the roasting process, owing to an
oxidation of the sulphide of iron on the admission of atmo-
spheric air; part of the sulphurous acid escapes, and part is
transformed by oxidation into sulphuric acid, which then
combines with a part of the protoxide of iron; the rest of
the protoxide of iron is converted into proto- and per-oxide
of iron. The sulphate of iron and the magnetic oxide are
gradually converted into per-oxide of iron, which forms a
porous coating over the fragments of the roasting mass,
allowing the oxygen of the air and the sulphuric acid
gas formed again to decompose the sulphide of iron in
the manner described, until at last the whole is trans-
formed into oxide of iron with a little sulphate of iron. The
reaction of the sulphuric acid increases, and that of the
atmospheric oxygen decreases, the thicker the coating of the
oxide of iron becomes. If the sulphuric acid is at last pre-
vented from reacting upon the roasting mass, or combining
with the free oxides, it then escapes as sulphurous and sul-
phuric acids; the sulphuric acid, absorbing water from the
atmosphere, forms white vapours. When the air has not
had free admission to a part of the roasting mass, the sul-
phurous acid enclosed in the pores of the hot peroxide of
iron may transform part of the peroxide into magnetic oxide,
and the sulphurous acid in contact with sulphide of iron
ROASTING THE ORES.
21
may be decomposed into sulphuric acid and sulphur; the
sulphur vapour, combining with the sulphide of iron, pro-
duces magnetic pyrites.
When most of the iron is to be scorified by the subsequent
smelting of the roasted ore, it must be converted into per-
oxide; on coming into contact with coal, this peroxide is
reduced to protoxide, and scorified as such. If at the
commencement iron is present as a protoxide, part of it will
be reduced to metallic iron, which is not usually de-
sirable.
When oxidising sulphide of iron and sulphide of copper,
the silver, lead, zinc, nickel, cobalt, manganese, &c.,
which may be present are transformed into free oxides and
neutral or basic sulphates, and they can then be partly ex-
tracted with water. If the ores contain some antimony, a
combination of teroxide of antimony, antimonious acid, and
metallic antimoniates will be formed, whilst part of the
antimony is volatilised. Arsenic is partly volatilised, and
some of it deposited upon the surface of the roasting mass
as crystalline arsenious acid; this acid is frequently
coloured orange yellow or red by an admixture of sul-
phide of arsenic. Part of the sulphur always escapes
from the iron pyrites in the form of gas at the beginning of
the roasting, when the low temperature of the roasting mass
prevents its complete oxidation, and it condenses upon the
surface of the roasting mass, and may be collected. If the
roasting mass contains arsenic, this sulphur always contains
some sulphide of arsenic.
Copper ores containing but little sulphide of copper in
proportion to their sulphide of iron have the following
peculiar behaviour in the roasting process. Upon properly
roasting these ores, the copper concentrates in the core of
the fragments, showing the appearance of copper pyrites or
variegated copper ore, while the porous crust of peroxide of iron
contains little or no copper. The core or kernel contains most
copper at its points of contact with the peroxide of iron;
the core does not always form the centre of the roasted
piece, but is contained sometimes in its lower, and now and
then even in its upper part. If the roasting is carried on
22
COPPER.
until no sulphide of iron (Fe,S,) remains, or no sulphur is
volatilised in the form of gas, some sub-oxide of copper will
be formed besides peroxide of iron, and the core will have
the composition of a rich copper matt, from which metallic
copper may be extracted by a further continuation of the
roasting. This mode of roasting is still in use at Agordo in
the Venetian Àlps, for enriching poor copper ores. The
crust is easily separated and then thrown away, or used for
the production of vitriol.
Karsten,* Werther,t and Von Lürzer I have published
theories of this peculiar roasting (core-roasting, kernel
roasting), and Plattner § explains the process as follows:
Sulphurous acid and protoxide of iron are formed by the
reaction of the atmospheric air upon fragments of ore which
have been heated until a small part of their sulphur has
escaped in form of vapour. The greater part of the sul-
phurous acid escapes, but part of it is transformed into sulphuric
acid, forming magnetic oxide of iron and sulphate of iron;
the sulphate of iron and the magnetic oxide of iron are then
converted into oxide of iron by a continued reaction of the
air and sulphuric acid. The sulphide of copper being less oxi-
disable, and also protected from oxidation by sulphur vapour,
which escapes from the interior of the ore fragments, fuses
and combines with the sulphides with which it comes into
contact. As the porous crust allows a continued admission
of air to the sulphide of iron, causing the above-mentioned
transformations, the crust of peroxide of iron grows thicker
and thicker, while the sulphide of copper becomes more and
more concentrated in the core. The thicker the crust
grows, the less air will be admitted, and the oxidation is at
last chiefly effected by the sulphuric acid formed in the crust.
After the roasting has been carried on for some time the
undecomposed inner part of the ore will be found sur-
rounded by a small strip having the appearance of copper
pyrites (Cu,S, Fe,S,); a combination similar in appearance
* KARSTEN, System der Metallurgie, iii., 433.
+
ERDMANN, J. f. pr. Ch., tom. 58, p. 321. B. u. h. Ztg, 1853, P. 439.
TUNNER'S Jahrb., 1853, p. 339; 1854, p. 242. B. u. h. Ztg., 1853, P. 440.
§ PLATTNER'S Röstprocesse, p. 195.
KERNEL ROASTING.
23
to variegated copper ore, (3Cu₂S, Fe₂S,), is formed later;
and a further concentration of the copper up to the compo-
sition of copper matt (Cu₂S, FeS) may take place as long as
sulphur vapour escapes. Some pieces are found with cores
consisting of masses similar to copper pyrites; these cores
are surrounded by a mass like variegated copper ore, and
this again by a mass of the composition of copper matt, and
the whole is enclosed in the crust of peroxide of iron. An
equal division of the sulphide of copper in the sulphide of
iron takes place at a high temperature when the develop-
ment of sulphur ceases, and the oxidation of the sulphide
may be carried on without attacking the sulphide of copper
until a combination of the composition of copper matt
results; if the roasting is carried on still further, sub-oxide.
of copper, protoxide of iron and sulphurous acid will be
formed. Sub-oxide of copper cannot be transformed into
oxide so long as sulphurous acid escapes, which would
again reduce the oxide; metallic copper and sulphurous
acid may be formed by the reaction of sub-oxide of copper
upon sulphides of copper and iron. The greater part of
the sub-oxide of copper and metallic copper remains with the
oxide of iron, and is transformed later into free oxide and
sulphate of copper by the reaction of sulphuric acid. A
well-roasted piece of copper ore may contain, therefore, per-
oxide of iron, magnetic oxide of iron, oxide of copper,
sulphate of copper, sub-oxide of copper, and metallic copper.
The more sub-oxide of copper there is present, the more
metallic copper is separated. If the ores contain sulphide of
lead, a formation of sulphate and oxide of lead will take
place; both substances remain with the oxide of iron, and
as they cake easily, they facilitate the formation of magnetic
oxide of iron. Sulphide of zinc is transformed into free
oxide and sulphate of zinc. If sulphide of silver is present,
a great deal of metallic silver may be extracted by concen-
tration into the cupriferous core.
From this it is obvious that the core-roasting must not be
carried on too far. If, in consequence of mechanical impe-
diments, such as an admixture of gangue, or in consequence
of too low a temperature, particles of sulphide of copper
24
COPPER.
remain in the crust of peroxide of iron, they will be trans-
formed into sulphate of copper by a surplus of sulphurous
and sulphuric acid present, or metallic copper is formed by
the reaction of sub-oxide of copper upon sulphide of copper,
and a gradual increase of copper takes place in the crust.
from the outside to its inner part. As the concentrated
sulphide of copper fuses, it may permeate the porous crust,
thus explaining a fact often observed, namely, that the
cupriferous portion is sometimes contained in the lower part
of the lumps of ore and sometimes even in the upper part
of the lower layers of ore.
Copper slate, a combination of silicates, carbonates, and
bitumen, with some disseminated copper ore, and containing
only
a few per cents of copper, loses its bitumen at the begin-
ning of the roasting, provided there is an ample admission
of air; by combustion this bitumen forms carbonic acid and
steam. At the same time the oxidising reaction of the air
upon the sulphides begins to form sulphates, and the car-
bonates of calcium and magnesium are also transformed into
sulphates. The gypsum formed is liable to cake if the tem-
perature is too high.
Regnault and A. Dick* have found by experiments that Cu₂S
may be partially decomposed at a higher temperature by
the action of steam and by the exclusion of atmospheric air,
forming HS and CuO, which then reacts decomposingly upon
Cu₂S, whilst metallic copper is separated; the result is,
therefore, Cu₂S in fused globules and metallic copper.
The roasting of copper ores is performed-
1. In Heaps, chiefly when roasting ores rich in sulphur in
fragments, and when a perfect roasting is not required. If
the ores are in the form of powder (schlich) they are either
put on the sole of the heaps or upon them as a cover. They
are also sometimes moulded into bricks in admixture with
lime or with a solution of sulphate of iron. More or less
wood is used as foundation of the heaps according to the
amount of sulphur the ore contains. Fuel is never put into
the heaps themselves unless the decomposition of antimony,
* PERCY, Metallurgy, i., 256.
ROASTING ORES IN HEAPS.
25
arsenic, or zinc compounds is intended. The fuel not only
reduces the antimoniates and arseniates, but it also increases
the temperature, thus evolving them as sublimates. The
formation of these heaps is described in Vol. I., Chapter I.
If
If a kernel-roasting in heaps is intended, the heap must be
most carefully covered so as to prolong the process to three
or four months. According to Forbes,* when thus roasting
argentiferous, and perhaps also auriferous ores, the outside
of the lumps of ore is the richer, so that part of the silver
remaining in the crusts is lost when smelting the cores.
a roasting heap suffers by the cooling influence of the atmo-
sphere, owing to too small an amount of sulphur in the ores,
and if a lixiviation of vitriols, causing a loss of copper, is
to be prevented, the heaps must be covered by a shed. Ores
sometimes require to be roasted more than once, according
to the sulphur, &c., they contain. Copper ores containing
antimony and arsenic, and enough silver to be worth ex-
tracting, must be roasted in sufficiently high and steep
heaps covered with a finely divided substance, and the ore
must be in pieces not larger than the size of the fist; these
modifications will somewhat lessen the loss of silver by
volatilisation.
A comparison of the results of roasting in heaps with
roasting in reverberatory furnaces shows that the first mode
is well adapted for ores rich in sulphur and not requiring a
complete roasting, as it economises fuel and labour, sometimes
being only one-fourth as expensive as roasting in furnaces.
Other advantages are that the ores may be roasted in heaps.
in fragments, and no expensive apparatus is required.
However, the superiority of this process depends on the
degree to which the roasting is to be carried, and on the
facility of the roasting. In places where coal is dear a
roasting in heaps with two fires will not cost more than one
roasting in reverberatory furnaces. The reverberatory fur-
naces again have the advantage of yielding a more uniform
product, and the roasting process may be regulated according
to requirement. On the other hand, roasting in heaps has
* PERCY, Metallurgy, i., 447.
26
COPPER.
the great disadvantage of requiring a longer time for roasting,
sometimes nine or ten months, requiring, therefore, a con-
siderable store of material and a large space of ground;
good wood fuel also is necessary. This led in the Lower
Hartz to experiments on roasting in mounds, but without
satisfactory results. If previously mixed with small coal
ores containing zinc blende* may be roasted in heaps almost
as perfectly as in mounds, and less of deposits and skumnas
(a sort of matt containing oxysulphide) results in the sub-
sequent smelting.
2. In Mounds, or in a Walled Area. The mode of
surrounding the area on which the roasting takes place with
three or four little walls, leaving a door in the one in front,
has been adopted in order to get more control over the fire
and to save fuel. It is used for ores in the form of schlich,
as well as for ores in lumps. Ores are used in lumps chiefly
when they are rich in copper and silver, to prevent mechanical
loss as much as possible. These mounds are of different form,
sometimes rectangular, oval, horse-shoe-like, &c. The
rectangular form is most usual, and several such mounds are
frequently erected, connected with each other by their lateral
walls, and all terminated by a common wall which forms the
posterior part. Sometimes they are covered by a shed partly
supported by the back wall, and built sufficiently high for
this purpose; the dimensions of the mounds vary. The more
heat the roasting mass can sustain without caking, the larger
may be the mounds employed. The mounds for roasting sul-
phuretted ores and matt must be smaller than those for
roasting iron ore, as the roasting process can be regulated
better in smaller mounds. The smaller mounds of the
capacity of 2 to 50 tons of matt or sulphuretted ores are from
3 to 8 feet broad, 8 to 20 feet long, and 3 to 6 feet high.
While, for instance, copper matt cannot be roasted in larger
quantities than 5 tons on account of its easy fusibility, the
more refractory raw matt may be roasted in quantities of
50 tons and more. An area of 8 feet long, 4 feet broad, and
3 feet high, is sufficient for roasting 5 tons of matt. Large
Bgwkfd., xiii., 404. B. u. h. Ztg., 1859, p. 70.
ROASTING ORES IN REVERBERATORY FURNACES.
27
mounds, as used for roasting iron ores, are sometimes 20 feet
high, 25 feet long, and 15 feet broad, and occasionally even
larger.
Sometimes the mounds are provided with condensation
flues for collecting the escaping sulphur,* and in Hungary
they have flues for condensing the escaping mercury; kernel
roasting is carried on in mounds at Wicklow.t
3. In Cupola Furnaces.-The construction and size of
these furnaces, formerly used on the Continent, has been
modified according to the English kilns, and satisfactory
results have been obtained. These furnaces consume con-
siderably less fuel than either heaps or mounds, and they
are employed chiefly when the collection of most of the
escaping sulphur is intended. The sulphur escaping in the
form of sulphurous acid is mostly used for the manufacture of
sulphuric acid.
Nordenskjöld tried to employ steam for the roasting pro-
cess, and Brunfaut|| describes a method of completely
desulphurising sulphuretted ores by mixing them with coal
and by an alternate treatment with hot blast and steam.
Richardson applied the galvanic current to the roasting
process; but all these modifications have no practical appli-
cation.
4. In Reverberatory Furnaces.-This method is the
most expensive with regard to labour and the consumption
of fuel, and is only employed when pure and easily fusible
ores are to be roasted as completely as possible.
The reverberatory furnace affords one of the best means.
of roasting where it is necessary to employ the simultaneous
action of heat and atmospheric air to destroy certain com-
binations, or to decompose sulphides, arsenides, &c.
is also evident that the facilities thus offered for stirring the
matters spread out on the sole, for renewing the surface, for
*
It
WEHRLE, Hüttenkunde, i., 226, 346; ii., 193, 307. Oesterr. Ztschr., 1861,
No. 5, &c. RITTINGER'S Erfahrungen, 1853. B. u. h. Ztg., 1S59, p. 91.
↑ B. u. h. Ztg., 1858, p. 25.
+ Ibid., 1856, p. 138.
+
|| Ibid., 1861, p. 446.
§ Ibid., 1852, p. 302; 1854, p. 187.
28
COPPER.
observing their appearances, for augmenting or diminishing
the degree of heat, &c., promise surer results and better
executed roasting than any other process. We know also
that flame mingled with much unburnt air issuing from the
furnace is highly oxidising, and very fit for burning the sul-
phur and oxidising the metals. Finally, this is almost the
only method suitable for roasting finely powdered ores.
The ore, whether roasted by one or more fires, is usually a
mixture of oxides, sulphates, some undecomposed sulphides, or
sulphides reduced to a lower state of sulphuration, and some
earths more or less transformed (for instance, CaO,CO₂ will
be CaO after roasting). If the ores contain antimony and
arsenic, antimoniates and arseniates will also be present.
Well roasted ore is porous, earthy, and of a reddish brown
colour. When black it contains magnetic oxide of iron,
owing to an insufficient admission of air, and this is liable
to cause a caking. The core of many pieces is proportionally
rich in copper.
2
Besides the roasted ore there may result from the roasting
process-
Raw Sulphur, containing more or less arsenic and sele-
nium; this must be refined; and
Sulphurous Acid, which is in some places used for the
production of sulphuric acid.
Smelting the Roasted Ores.
This process aims at the scorification or volatilisation
of the foreign oxides and earths contained in the roasted
ore and the concentration of the copper in a matt. Upon
submitting well roasted and suitably mixed ores to a
purifying and reducing smelting at an appropriate tempera-
ture, in cupola furnaces, the following reactions take
place :-
Ore mixture and fuel are not charged alternately in layers,
but the ore mixture is constantly put in the front of the
furnace while the fuel is put at the back wall. When the blast
enters the furnace by the nose (which is kept 6 or 8 inches
long) and comes in contact with the fuel, carbonic oxide and
carbonic acid are formed; the greater part of the carbonic
SMELTING THE ROASTED ORes.
29
acid is converted into carbonic oxide by passing through the
glowing coals. The carbonic oxide reduces the ore more or
less according to the size of the fuel and ore, and according
to the construction of the furnace. The ore being heated in
the upper part of the furnace gradually enters the hot lower
parts, and the rising carbonic oxide reacts upon those oxides
and salts which are easier to reduce; these are copper and
lead in preference to iron, manganese, and zinc. The oxide
of copper is first reduced to suboxide and then to metal, and
the oxide of lead to metal, which partly volatises. Peroxide
of iron is mostly converted into protoxide, and in less quantity
into metal, except when the temperature is too high; if
present, oxide of tin is partly reduced to metal. Metallic
copper may also be produced in the lower parts of the furnace
by the mutual reaction of Cu₂S, Cu₂O, and CuO,SO,, upon
each other. Sulphates, antimoniates, and arseniates are
transformed partly into oxides and partly into sulphides,
antimonides, and arsenides, while sulphurous, antimonious,
arsenious acid volatilise; and the volatilisation of these
substances is more complete the slower the reduction takes
place at a very gradually rising temperature; it is, therefore,
best performed in high and wide furnaces. The undecomposed
sulphides contained in the ore either remain unchanged or
they lose part of their sulphur, and then again sulphurise
any metallic copper and iron which may be reduced. If the
smelting mass comes near the zone of fusion, which begins
somewhat above the tuyere, any oxide of zinc present is also
reduced to metallic zinc; this partly escapes from the fur-
nace mouth in the form of gas, and partly forms deposits on
the front wall of the furnace. Part of the oxide of zinc is
scorified later on. The masses now entering the zone of
fusion begin to soften, the silica begins to combine with the
unreduced metallic oxides and with the earthy substances,
forming a slag. This slag melts before the tuyere and more
or less covers the metals, sulphides, antimonides, and
arsenides which drop down at the same time before the
tuyere, but it does not entirely protect them from the influ-
ence of the blast; part of the antimony, arsenic, sulphur,
lead, and zinc is therefore volatilised. Any sulphates,
30
COPPER.
antimoniates and arseniates still present, are likewise decom-
posed by silica. The fused slags then form in the hearth a
suitable medium for the collection and separation of the me-
tallic masses according to their specific gravity. The crude
copper settles on the sole of the hearth; upon this the speiss,
(metallic antimonides and arsenides), then the sulphides
(matt), and finally the slag. As before mentioned (page 18)
the formation of crude copper and speiss in this process
must be avoided as much as possible. The formation of
black copper may be prevented by a well conducted roasting;
a sufficient quantity of sulphides will then remain in the
ore; this readily combines with the copper, forming a matt
rich in copper, on account of its great affinity to sulphur.
As long as the ore mixture contains more sulphur than the
copper requires to become Cu₂S, the copper present will
combine with the matt. Scorified sub-oxide of copper is
transformed, by the reaction of the fused matt upon the slag,
into mono-silicate of iron and sulphide of copper, the latter
then combines with the matt ;-
3Cu₂O+SiO3+3FeS=3FeO,SiO3+3Cu₂S.
A certain amount of copper in the slags originates, there-
fore, less from scorified copper than from matt mechanically
enclosed in them. Sulphide of calcium reduced from sulphate
of calcium, and sulphide of barium* reduced from sulphate
of barium react like sulphide of iron. If iron from the ore
mixture is carbonised above and fused before the tuyere,
permeating the liquid slag in the hearth, it causes the
separation of metallic copper if any has been scorified, with
formation of silicate of iron. Tin reacts less energetically
upon proto-silicate of copper than iron. The extraction of
the copper contained in the black copper and refined copper
slags added to the ore smelting mixture is based upon these
reactions. A reduction of suboxide of copper contained in
the slags still takes place by keeping coal below and behind
the fore stone, as the fused masses, chiefly the slags, have to
pass through them before flowing out of the furnace. There-
fore, when carrying on the process normally, slags and raw
*B. u. h. Ztg., 1860. p. 184.
SMELTING THE ROASTED ORES.
31
matt only will result; but if the ores contain a larger
quantity of antimony and arsenic, a formation of some
spciss is almost unavoidable.
A normal process chiefly requires—
I. A suitable Temperature at the Smelting.-If too high
other oxides beside the oxide of copper will be reduced
abundantly, causing irregularities of the process; for instance
iron will cause the formation of iron deposits and of matt
richer in iron and poorer in copper; zinc gives rise to the
formation of much deposit on the front wall,* and both
injures the process and shortens the operations; zinc is the
most injurious, as zinc deposits will absorb valuable metals,
such as gold, silver, and copper, which can only be extracted
from them with great loss and expense. If the temperature
is too low a partial scorification will take place, owing to the
more incomplete reduction of the oxidised copper. Although
it is impossible to regulate perfectly the temperature of these
furnaces, still the operators must do what they can to regu-
late it. The temperature is usually too high; the height
and width of the furnaces, the proportion of the charges of
ore and fuel, &c., influence the temperature.
2. A Properly Reducing Reaction.-If this reaction is
too strong, a great deal of the foreign metals will be reduced,
and either form deposits of iron or produce a very impure
copper; if too feeble, a scorification of oxidised copper will
take place.
The greater the amount of foreign oxides
present, the more the reducing reaction must be limited by
the suitable height of the furnace, &c.
3. The Degree of Roasting. The external appearance
of the roasting mass does not always indicate the proper
degree of roasting, but it must be ascertained from the
smelting process. If the roasting has been properly con-
ducted, a poor slag easily separating from the matt, and a
matt containing 25 or 35 per cent and not more than 40 per
cent of copper, will result from the smelting process, but
no black copper. If the roasting has been carried on too
*
P. 74.
KERL, Rammelsberger Hüttenpr., 1861, pp. 42, 185. B. u. h. Ztg., 1859,
32
COPPER.
far, copper becomes scorified, and a great deal of impure
black copper is produced, which cannot be well purified so as
to give an excellent refined copper without great loss, if the
ore contains much antimony and arsenic. If, on the other
hand, the roasting has not been carried on far enough, a matt
poor in copper and containing much sulphur and iron will
result; this matt is difficult to separate from the slags owing
to its low specific gravity, thus causing a considerable
mechanical loss of copper. Antimonial and arsenical ores
must be only slightly roasted, in order to avoid the extraction
of copper rich in these metals. The antimony and arsenic
not volatised by the roasting and reduction smelting process
collect in the matt, and may be removed from it by repeated
roasting and smelting. Ores containing zinc blende require
a careful roasting and mixing with coal in the roasting mass
in order to decompose the sulphate of zinc formed; other-
wise the sulphate of zinc will be reduced at the reduction
smelting to sulphide of zinc. This, together with the zinc
blende which has remained undecomposed, will not enter the
matt, but will sulphurise part of the protoxide of iron con-
tained in the slag; the sulphide of iron formed then enters
the matt. This causes the slag to become less fluid, and the
matt of less specific gravity, scarcely separable from the
tough slag.
4. The Composition of the Ore Mixture. The reduction
smelting requires the formation of slags which are either bi-
silicates containing protoxide of iron and up to 45 per cent
of silica, or silicates standing between bi- and mono-silicates.
If the nature of the ore will allow it, it is mixed so as to
form these slags they are sufficiently liquid and of a
sufficiently low specific gravity to allow a separation of the
raw matt, they do not chill too quickly, and they form
small deposits in the hearth, and they attack the furnace
walls but slightly. To prevent the scorification of copper
caused by the larger amount of silica in the slags, the
roasting must not be carried on too far and some sulphur
must be left in the roasting mass which will then combine
with the copper. Thus the formation of deposits of iron is
also avoided, and a grey, green, or blackish slag, with an
COMPOSITION OF COPPER SLAGS.
33
inclination to the vitreous state, and poor in copper results.
The following analyses show the composition of copper
1
slags:-
I.
II.
III.
IV.
V.
VI.
SiO 3
Al₂03
CaO
MgO
38.77 33'95 34'67
3'33 10'16
4.38
4.87 4°32 353
0*22
23.50 2162 39'75
3'50 5'15 6.65
334
2:57 3'00
2.00
FeO
45*32 49*38 48 25 66.85
65.62 40*98
Cu₂O
0'19
1.69
2:26 0'02
ZnO
2.23
2.89
3.65
РЬО
1.88
I'07
MnO
2'0
FeS
3'70
Cu,S
o'91
ZnS
3.81
S
1.85
3'60 2.08
Nos. 1 to 3 are raw slags obtained in smelting copper pyrites
at the Upper Hartz; they are of an iron black colour, and
according to Plattner, of the formula-
m(FeO,CaO,PbO)3,2SiO3+n(FeO,ZnO);,SiO3+Al2O3,SiO3.
No. I is a slag from Lautenthal analysed by Bodemann,
and according to him consisting of-
7(FeO,CaO,PbO)3,2SiO3+(FeO,ZnO,,SiO3+Al₂O₂SiO¸.
No. 2 slag from Altenau, by Hampe. No. 3 from the same
place by Streng. No. 4 and 5 are slags from the Lower Hartz,
mixtures from mono- and sub-silicates. No. 4, slag from Oker
smelting works, by Breymann-
7(FeO,CaO),SiO3+Al₂O3SiO3.
No. 5, slag from the same works, by Kerl-
4(FeO,CaO),SiO3+Al₂O3SiO¸.
No. 6, slag from Gilsaa smelting works in Lapland, by Stro-
meyer.
Slags still more silicated so as to stand between bi- and
trisilicates are seldom formed, and only when the earthy
components of the sulphuretted ores are nearly of the same
composition as such slags, as, for instance, the Mansfeld
copper slate slags. They are more difficult to fuse, and
still more inclined to scorify copper, but they allow long
VOL. II.
D
34
COPPER.
smelting operations. They are mostly of a light colour,
vitreous, and tough when semi-liquid.
The slags of Mansfeld are, usually, trisilicates; the
alumina is calculated as united with the silica. Their com-
position is shown by the following analyses :-
IV.
V.
VI.
54 13 53.83 57°43
I.
II.
III.
SiO3
Al2O3
49*80 48*22 5000
12:20 16:35
15.67
10*53 4°43 783
CaO
19*20 19:29
20:29
19:41
33*10 23'40
MgO
2'40 3°23
4:37
I'79
I'67
0.87
FeO
13°20 10*75
8.73 10.83
4'37 7°47
Cu₂0.
ZnO
0*75 0.67 2°03 0*25
I.26 I'II
0'30
Fe
I'IO
2°09 I'97
If the ores contain much iron pyrites, forming peroxide of
iron during roasting, the formation of mono-silicate slags,
or of mixtures of mono- and sub-silicates must be attempted,
in order to scorify the excess of iron.
In such cases it is advisable to form at the same time
monosilicate slags containing earthy and metallic oxides as
bases. Slags poor in earthy bases and too rich in protoxide
of iron are thin when fused, of a dark colour, and of more
or less metallic lustre; they crack when cooling, and give
rise to frequent deposits, owing to their inclination to chill;
they strongly corrode the furnace walls, and therefore allow
only short smelting operations. The raw matt only separates
with difficulty from them, as they chill easily and as their
specific gravity is rendered high by the protoxide of iron
they contain; a surplus of iron, on the other hand, prevents
scorification of copper. If these disadvantages are avoided
by an abundant addition of silicious substances, other dis-
advantages, such as the formation of iron deposits, and scori-
fication of copper, will ensue. By limiting the formation of
peroxide of iron by a less roasting, matts poor in copper will
be formed, the concentration of which is more expensive.
The formation of a suitable admixture requires a thorough
mineralogical and chemical knowledge of the ores, and the
observation of the following general rules.
FORMATION OF COPPER SLAGS.
35
a. Well-roasted ores with a considerable amount of copper
are mixed either with roasted copper ores rich in oxide of
iron, or with basic fluxes, such as slags of the matt smelting,
or the smelting of black copper, and also of the refining of
copper; they are seldom mixed with iron ore. These fluxes
are added in such quantities that either bisilicates or an ad-
mixture of bi- and mono-silicates will result; fluor spar is
usually a very appropriate flux. Observations of the copper
smelting in reverberatory furnaces show that fluor spar
assists the removal of arsenic by forming fluoride of arsenic,
and also causes a loss of copper, probably by forming volatile
fluoride of copper. If the ores contain silica in quantity suffi-
cient for the formation of slags, they are smelted without any
addition, or with neutral flux, to facilitate the fusibility of
the mass, or to protect the matt. But an excess of this
addition augments the smelting mass, and increases the
consumption of fuel and the loss of copper by the slags.
According to Le Play,† some copper slags always contain
more sulphur for certain quantities of copper than the matt
produced at the same time, and it is supposed that the
excess of sulphur is contained in the FeS dissolved in the slag
owing to the tendency of the protosilicate of iron to form
a chemical combination (sulphosilicate), which is said to
prevent the presence of oxidised copper in the slag, as it
becomes sulphurised by the sulphide of iron. Le Play has
not proved the existence of the sulpho-silicate of iron, and
as the sulphurising of oxidised copper in the slags may be
effected by sulphide of iron, it is probable that Percy's ‡
theory is right, which supposes the amount of sulphur in the
slags to originate from intermixed matt. It may also be
the case that the sulphur contained in some iron slags is
derived from mechanically mixed sulphides, though it seems
that sulpho-silicate is contained in some natural minerals,
for instance, helvin.
b. If the ores contain a considerable amount of oxide of
iron, they are mixed either with silicious ores or similar
PERCY, Metallurgy, i., 371, 496.
↑ LE PLAY, Beschreibung der Waleser Kupferhüttenprocesse.
PERCY, Metallurgy, i., 344.
+-
D 2
36
COPPER.
fluxes in sufficient quantity. It is reported that the natives
of South America, at Viquintipa, use tinkal as a flux.*
c. Ores too strongly roasted, so that separation of black
copper is to be apprehended, require to be mixed with raw
ores or even with raw iron pyrites. Oxidised ores if present,
or residues of the desilverising processes, require the same
flux.
d. If ores are not sufficiently roasted, they must be either
roasted again or mixed with strongly roasted ores or
oxidised products, such as copper refinery slags.
e. Arsenical and antimonial ores, if too strongly roasted,
are mixed with raw ores or with iron pyrites, in order to
prevent the formation of speiss, and to combine with the raw
matt the portion of antimony and arsenic which has not
been volatilised by the smelting process. It is usually possible,
by a repeated roasting and smelting, to extract these
substances sufficiently to produce a saleable copper. If the
ores contain a considerable amount of antimony and arsenic,
the formation of speiss cannot always be avoided by a slight
roasting; and if the amount of those substances is so large
as to require too extensive a roasting and smelting process
to extract them sufficiently, the formation of speiss even
may be resorted to by giving suitable fluxes.
Beudant and Benoit have proposed a method for ex-
tracting antimony and arsenic from copper ores by means of
iron and lead.
The reactions of the components of the mixture which
may take place upon sulphuretted and oxidised copper are
clearly explained by the following results of experiments
made by Berthier, Percy,|| Smith,§ Dick, T and others :-
a. Sulphide of copper has a great tendency to combine.
with all metallic sulphides.
b. Sulphide of copper forms with suboxide, oxide, and
* V. LEONHARD, Hüttenerzeugnisse, p. 76.
+ BERGGEIST, 1856, p. 237.
+
BERTHIER, Handb. d. Metallurg., Analyt. Chemie, 1835.
PERCY, Metallurgy, i., 243.
§ Ibid.
Ibid.
REACTIONS OF THE COMPONENTS OF ORE MIXTURE. 37
sulphate of copper, according to the proportions employed,
either metallic, oxidised, or sulphuretted copper, as shown by
the following formulæ :-
Cu₂S + 2Cu₂O
Cu₂S+2CuO
Cu₂S + зCuO
Cu₂S + 6CuO
6Cu
+ SO₂
4Cu
+ SO₂
= 3Cu + Cu₂O + SO₂
= 4Cu₂O + SO₂
Cu₂S + CuO,SO3 = 3Cu +2SO,
Cu₂S+2CuO,SO, = 2Cu₂O + 3SO₂
Cu₂S + 4CuO,SO, = 6CuO + 5SO₂
2
c. Metallic copper forms with oxide of lead a more or less
cupriferous lead, besides fused slag; if the litharge employed
amounts to twenty-one times the quantity of copper, only
one-sixth of the copper will remain unoxidised; when copper
is treated with protoxide of iron it becomes oxidised on the
surface.
d. Oxide of copper with sulphide of lead form slag and
regulus; the amount of copper in the regulus increases.
according to the increase of oxide of copper.
e. This reaction, 3Cu₂O+3FeS+SiO3=3Cu₂S+3FeO,SiO3,
is the chief basis for the raw smelting.
f. If sulphide of copper is heated to a high temperature
with coal a little metallic copper will be formed, while
sulphide of carbon is evolved.
g. If sulphide of copper is heated with iron, the copper is
partially extracted, as sulphur salts of iron and copper will
be formed, upon which the iron no longer reacts. Sulphide
of copper and zinc form a cupriferous matt, and an alloy of
copper and zinc containing up to So per cent of copper; part
of the zinc is volatilised. If sulphide of copper is heated
with tin a regulus containing one-fifth or one-sixth of the
copper and some matt will be formed.
2Cu₂S+2Sn = CuSn + [Cu₂S + (Cu,Sn)S].
Sulphide of copper with antimony forms a matt and a com-
bination of copper and antimony. Sulphide of antimony
may also be decomposed by copper, forming matt and anti-
monial copper. The amount of metals in the products varies
38
COPPER.
according to the quantities of sulphide of antimony and
copper employed; for instance-
Mixture.
Matt.
Cu
Sb
S
Cu
Regulus.
Sb
S
3Cu + SbS3
47'53
29°32
23 15
3.86 95'97 017
6Cu + SbS,
66°44
16'91
16.65
42'54 57'06 0'40
75°90 22.80
I'30
12Cu + SbS.
18Cu + SbS3
66.72 32.98 0*30
I'33 75'90 24'03 0'07
77.36 2131
Sulphide of copper may be desulphurised by employing
twenty times the quantity of oxide of lead; cupriferous lead
and some slag then result. If less oxide of lead is taken
some matt will also be formed. Four parts of oxide of lead
transform one part of sulphide of copper into oxide of copper
and sulphurous acid, but in practice more oxide of lead is
required.
5. The Construction of the Furnaces.-The form of the
interior of the furnace exerts essential influence upon its
reducing powers.
This reaction may be increased by providing the furnaces
with more or less distinct boshes,* or with perpendicular
walls. Both these modifications cause the reducing gases
to be longer in contact with the smelting mass. By giving
the front and back wall a backward inclination this reaction
will decrease.
The height of the furnaces is also important, as the re-
ducing reaction increases in proportion to the height of the
furnace; therefore mixtures containing much oxide of iron.
require low furnaces to avoid the formation of iron deposits.
If the mixture contains less iron pyrites and not too much
earthy matters, higher furnaces may be employed; and mix-
tures with a little iron and much earthy matter being difficult
to fuse, high furnaces, similar in construction to the iron blast
furnaces, are suitable.
Sulphates, antimoniates, and arseniates are more com-
pletely reduced in high furnaces, owing to the longer contact
with carbonic oxide gas, unless the process in low furnaces
is carried on much slower; it is therefore advisable to smelt
ores poor in iron, and containing much antimony and arsenic,
Bgwkfd., ii., 257.
THE CONSTRUCTION OF THE FURNACE.
39
in high furnaces at a low pressure of blast. If ores con-
taining those substances are smelted rapidly at a high tem
perature, much iron will be reduced, and deposits of iron be
formed, besides speiss and matt rich in iron. For this reason
a quick smelting may be recommended only for ores poor in
iron and rich in earths. A rapid smelting of pure ores saves.
fuel, but not of impure ores, as then the greater part of the
antimony and arsenic enters the matt, and the working
up of this requires more elaborate processes. In rare cases
hearth furnaces are used, as in India* and Japan.†
With the increase of the width of the furnaces to a certain
extent, the production, the saving of fuel, and the length of
the operations also increase, provided the required uniform
temperature is sustained in the hearth by employing sufficient
tuyeres; cupola furnaces in Sweden‡ have three, and the
furnaces at Bogoslowsk|| four tuyeres. The more difficult
the ore employed is to fuse, the narrower is the hearth. If
the furnaces are too wide, the saving of fuel decreases; but
this may be partially avoided by dividing the furnace by an
inner wall (Sweden)§; Rachette's furnace T has 24 tuyeres.
Ores containing zinc require wide furnaces, which also
facilitate the decomposition of antimoniates, arseniates, and
sulphates by carbonic oxide gas, owing to the slower sinking
of the charges.
The Swedish cupola furnaces, by virtue of their spacious
hearths, allow the accumulation of larger quantities of matt
during a longer time, thus offering the following advantages:-
The matt and slag separate more completely; the copper in
the slag combines better with the sulphide of iron of the matt
by the reaction of the slag and blast upon the matt; the
matt becomes richer in copper, whilst the iron scorifies; and
as the sump is always kept filled with fused matt, a de-
posit of iron is prevented, thus again lengthening the smelting
operations.
* B. u. h. Ztg., 1862, p. 117.
+ Ibid., p. 118.
Ibid., 1859, pp. 89, 71.
|| Bgwkfd., Bd. 20, p. 482.
§ B. u. h. Ztg., 1859, p. 69.
¶ Ibid., 1862, p. 265. DINGL., Polyt. Journ., Bd. 165, Hft.
5.
40
COPPER.
The construction of the furnace hearth greatly influences
the consumption of fuel, the reducing reaction, the formation
of iron deposits, and the more or less sulphuration of the
impure metallic copper which has been separated by the
smelting process, when in contact with the fused matt.
Furnaces constructed as crucible furnaces give good results
as regards economy in fuel and the extent of the production.
They also prevent a separation of copper, as it is kept longer
in contact with the fused matt at a sufficiently high tempera-
ture. But as the reducing reaction is rather strong in these
furnaces, and as most copper ores contain iron, the formation.
of iron deposits on the hearth cannot be avoided. These
deposits cannot be removed when large, and therefore such
furnaces are seldom employed.
Sump furnaces, though not offering the same advantage
with regard to fuel as crucible furnaces of equal height, allow
the removal of deposits which may be formed, and also the
re-absorption of the metallic copper separated; they are,
therefore, chiefly fitted for smelting ores rich in iron.
Channel furnaces with two open eyes possess the same
disadvantage of not allowing the removal of iron deposits in
the hearth, and as their reducing action is greater than
that of sump furnaces they are not adapted for ores rich in
iron. The fused mass running out of the furnace all at
once is allowed to separate outside the furnace at a lower
temperature, and this causes the separation to be less perfect
than in sump and crucible furnaces. The slags being in
contact with the matt in the outside basin are so much
mixed with matt as to require re-smelting. Silicious slags
which chill gradually allow a better separation than basic
slags which chill quickly. These furnaces also facilitate the
formation of coarse copper, as the extracted copper is not
kept sufficiently long at the suitable temperature in contact
with the fused matt outside the furnace.
But when smelting suitable ores, these furnaces possess
some advantages which the sump furnaces have not. They
are cheaper, not requiring the formation and warming of a
fore hearth; the working is easier, not requiring the cumbrous
tapping off; and they allow a larger production, as the fused
THE CONSTRUCTION OF THE FURNACES.
41
masses continually flow off, whilst the slag in the sump
furnaces, when its hearth is filled, becomes forced through
beneath the front wall, so prolonging the process. They
allow longer operations and yield a better product, as in-
jurious substances (antimony, arsenic, zinc) may be oxidised
and volatilised by the reaction of the atmosphere upon the
fused products when running from the furnace.
It is therefore advisable to employ low sump furnaces for
ores containing much iron pyrites, especially if they also
contain antimony and arsenic and are liable to form black
copper. Ores poor in iron (copper slate) are, however, best
smelted in high channel furnaces with two open eyes. These
latter furnaces, more or less high, have some advantage over
sump furnaces for smelting ores somewhat richer in iron.
On comparing the process in cupola furnaces with that in
reverberatory furnaces, we find that less deposits of iron and
less black copper are formed in reverberatory furnaces, but
less antimony and arsenic is expelled, as in cupola furnaces
carbonic oxide gas forms the chief reducing agent, whilst in
reverberatory furnaces, sulphur acts as the reducing agent.
Since any black copper which may be formed in reverberatory
furnaces is kept a longer time in contact with the matt, its
absorption by the latter is facilitated. Zinc is volatilised
less in reverberatory furnaces, than in cupola furnaces, and
the resulting matt is therefore richer in zinc and more diffi-
cult to roast. The slags resulting in reverberatory furnaces
are also somewhat richer in copper than those produced in
cupola furnaces.
Blast cupola furnaces are usually employed, and some-
times air* cupola furnaces, as for instance at Bouc in France.
The blast cupola furnaces are usually furnished with one
tuyere, but sometimes they have more. Hot blast facilitates
the reduction of oxide of iron and the scorification of copper,†
and its use is therefore only advisable when smelting earthy
ores poor in iron and difficult to fuse; it economises fuel
* Bulletin de la Société de l'Industrie Minérale, ii., 415; iii., 541. B. u. h.
Ztg., 1859, p. 242; 1862, p. 63.
MERBACH, Anwend., d. erhitzten Gebläseluft, 1840, p. 191.
42
COPPER.
and allows a quicker and better process. Nordenskjöld*
states that he produced a rich copper matt and poor slags
and obtained a quicker production by blowingt steam into
the furnaces somewhat above the tuyeres. But this result
was not obtained elsewhere, and does not exactly conform with
the theory, though an increase of the production was once
observed when employing steam in an iron blast furnace.
Charcoal and coke are mostly used as fuel, and wood
but seldom.‡ Coke allows a quicker and purer smelting on
account of the higher temperature produced, and the deposits
of iron are smaller as the smelting masses pass the reducing
zone more rapidly, and as coke produces less carbonic oxide
than charcoal does. Charcoal, again, is preferred to coke
at the subsequent matt smelting processes, as the higher
temperature produced by coke reduces more foreign metals
and renders the copper impure. The smelting of copper
ores rich in iron and poor in antimony and arsenic is there-
fore best effected with coke in low sump furnaces, and that
of ores poorer in iron in channel furnaces with two open
eyes more or less high.
At the smelting works in Boston|| Anthracite is used for
smelting pure ores in low sump furnaces.
The normal state of the process and the nature of the
mixture is indicated-
1, By the amount of copper in the matt.
2, By the nature of the slag.
3, By the behaviour of the tuyere ; a nose from 4 to 8 inches
long is usually employed, and the furnace mouth is kept
light.
The suggestions§ of Johns, Duclos, Bell, Davy, Newton,
and others, for improving the roasting and smelting in cupola
furnaces have been proved to be of no practical value.
* SCHEERER'S Metallurgie, ii., 32.
† B. u. h. Ztg., 1856, p. 138.
ERDMANN: Journ. f. ök. u. techn. Chem., xvii., 471.
B. u. h. Ztg., 1859, pp. 310, 325.
§ Bgwkfd., vi., 300, 302, 315; xii., 623. DINGL., Bd. 80, p. 227; Bd. 97,
P. 285.
PRODUCTS OF THE RAW SMELTING PROCESS.
43
The following products may be formed in the raw smelting
process:-
1. Black Copper.-This is produced intentionally when
smelting very pure ores; but when smelting impure ores its
formation indicates a deficient mixture, or a wrong construction
of the furnace.
2. Raw Matt.-Its composition is shown by the following
analyses -
I.
II.
III.
IV.
V.
VI.
VII.
Cu.
23.58 43.62
8:32 42'95 43.81
Fe.
38 42
23 35
62 26
3'10 61.66
27.08 2496 58.50 1189
Zn
1*23
2°30
Ni
3'45
0'57
I*14
116
Co
5.67
Mn
2°33
Pb
I'21
0'87
9'20 I'37
Ag
0'09 0'20
S
32.00 28.70 26 35
28 29
26.57 17.00 23.98
SiO 3
0'51
0°44
o'96
Sb .
1'40
As .
5'30
These crude matts have all been produced in cupola fur-
naces. Nos. 1 and 2 at Mansfeld. According to Ram-
melsberg-
No. I corresponds to the formula 4(FeS,ZnS, NiS), Cu₂S
2
,,
""
FeS,Cu₂S.
The matts from Mansfeld are always of cystalline structure,
sometimes showing regular octahedrons.
No. 3 is raw matt from Fahlun, corresponding, according to
Rammelsberg, to the formula 5(FeS, Cu,S,ZnS, PbS) 4Fe,S.
Nos. 4 and 5 are from Riechelsdorf. No. 4 contains for
every three equivalents FeS about two equivalents Cu₂S,
and perhaps some Fe,S,. No. 5 contains by calculation-
39.09 FeS,
51°06 Cu₂S, and
3'08 metallic copper.
No. 6 raw matt from Freiberg. No. 7 from the Lower Hartz.
These correspond to the formula FeS,2Cu₂S or (FeS, NiS,AgS)
2Cu₂S.
44
COPPER.
At Mansfeld, artificial copper glance* was found in the form
of octohedrons in the cavities of the copper matt of the fore
hearth, and also in forms derived from the octohedron and
cube. At Antonshütte in Saxony, red copper oret was found
in copper matt.
The raw matt is roasted and subsequently treated in
various ways which we shall presently explain.
3. Speiss. This is either desilverised direct or after an
oxidising smelting. At Oker it is roasted and smelted to-
gether with roasted raw matt. It is sometimes concentrated
for the production of cobalt and nickel speiss, and occa-
sionally also worked for the production of black copper.
The following analyses of speiss produced from Fahlerz
show their general composition:
I.
II.
III.
IV.
Cu
12.99
13°52
41*18
48.10
Pb
0'09
0.69
20.69
Fe
12.63
75'74
35'41
I'20
Ni
I'40
0'09
0°32
Co
0°09
0'04
Bi
I'26
2.04
Ag
0*360043
0*026364
0'030143
trace.
Au
0*056883
Sb
60'00
7.36
10'79
21'56
As
7°42
2.66
6'10
0'78
S
2.04
trace
2.60
1.88
Speiss Nos. I to 3 are from Hungary, and have been
analysed by Hauch. No. 4 is from Oeblarn, and has been
analysed by Schenzl.
According to Ferjentsik,‡ speiss poor in noble metals and
rich in antimony may be concentrated by an oxidising
smelting until it contains about 95 per cent of copper.
The volatilised oxide of antimony may be collected in con-
densers.
4. Iron Deposits in the Furnace Hearth.-These are
tecnichally termed ferriferous bears, and in Germany Eisensauen,
and are either thrown aside or worked up for the extraction of
LEONHARDT, Hüttenerzeugnisse, p. 366.
+ LEONHARDT, Jahrbuch, 1861., Hft. i., p. 80.
Oesterr. Zeitschr., 1859, pp. 293, 325.
FERRIFEROUS BEARS.
45
their copper, as at Atvidaberg* and Schmöllnitz.† In Siberia‡
they are treated, when hot, with sulphur to convert them
into matt, from which the valuable metals are extracted.
In Szaszka || they are roasted and smelted with iron pyrites
for the production of matt, and at Schemnitz they are used
as reagents for precipitation in the lead smelting process.
Their composition is shown by the following analyses :-
I.
II.
III.
IV.
C.
S
0'38
0'73
2.06
I'20
1°24
trace.
P.
I'25
As
I'40
Si
0*35
9'03
0.83
Mo
9'97
Fe
76'77
64.82
88.51
76.30
Ni
I'15
Co
3*25
Cu
3'40
32.88
19'90
Mn
0'02
Zn
0'02
Bi
0.85
Vd
0'12
A120₁
3
SiO3
Slag
0'72
1'58
0'43.
T
3°33
No. I is an iron deposit from the Mansfeld copper slate
furnaces, analysed by Stromeyer. § No. 2 from Fahlun,
analysed by Sefström. No. 3 is an iron deposit resulting
at Freiberg from the smelting of old slags, analysed by
Plattner.** No. 4 is produced in Perm, by Coubine.
Besides these ferriferous bears, other deposits are formed
in the furnaces, such as sulphide and oxide of zinc, sul-
phides of lead and copper.
These deposits, together with other furnace residues, are
B. u. h. Ztg., 1859, p. 84.
+ Oesterr. Zeitschr. 1859. No. 42.
+ B. u. h. Ztg., 1853, p. 486.
|| Ann. d. min., 4. sér., x., 555.
§ POGGEND., Annalen, xxviii., 551. ERDM. Journ. f. pr. Ch., ix, 177. ROSE,
Analyt. Chem., 1851, p. 359.
¶ ERDM., J. f. pr. Ch. iii., 300.
**
Kalender für d. sächs. B. u. Hüttenm., 1839, 104.
46
COPPER.
roasted and washed, and then added to the raw smelting.
The resulting smoke is treated either similarly, after being
mixed with lime, or it is extracted by means of acids.
5. Raw Slags.-These are either thrown away or added
at the matt smeltings. Markus* recommends the oxidation.
of the sulphide of copper contained in the slags, and its ex-
traction as sulphate of copper.
6. Furnace Gases.-These are used at Riechelsdorf for
heating the blast; they are sometimes employed for refining
the copper, &c.
Roasting the Raw Matt.
The raw matt forming the chief product of the smelting
of the roasted ores is roasted previous to its being submitted
to further treatment. The chemical reactions which take
place during this roasting are similar to those at the ore
roasting. The object of this roasting and the manner of
its accomplishment vary in different cases, and depend on
the purity of the matt. Matts free from antimony and
arsenic are roasted from five to eight times, according
to the amount of copper and sulphur present, till so much
sulphur is expelled that the remainder is insufficient to
keep all the copper in combination at the subsequent reducing
and purifying smelting; the greater part of the copper will
then be extracted as coarse copper. As it is impossible to
perform the roasting quite perfectly on a large scale, a small
quantity of rich matt will always be formed besides black
copper, and this formation acts favourably by converting
scorified copper into sulphide.
Impure ores roasted in this way yield black copper rich
in antimony and arsenic; for this reason their roasting is
only carried on until a sufficient amount of sulphur remains.
in the ores to form by the subsequent smelting a new matt
containing all the copper in a concentrated form. This matt
will then also contain unvolatilised antimony and arsenic,
and it is again more or less roasted according to its purity,
forming either only matt or black copper at the subsequent
smelting. Some black copper is likewise always formed at
Oesterr. Ztschr., 1857, p. 323.
ROASTING THE RAW MATT.
47
the smelting process for the production of concentrated matt,
owing to the imperfections of the roasting process; and this
is more or less impure according to the period in which it
results.*
The oftener these processes of roasting and smelting are
repeated, the purer, but also the more expensive, will be the
resulting copper; and it is a matter of calculation how often
these intermediate processes can be carried on profitably.
According to Heine,† the advantages of concentrating the
matt are outbalanced by the disadvantages.
Though it is difficult sufficiently to desulphurise by roasting
antimonial and arsenical copper matt containing no more
than about 20 per cent of copper, still it is possible to smelt
them at once without concentrating with an advantage for
the production of black copper, if suitable mounds are em-
ployed for roasting, and if the copper matt during the later
roastings is mixed with concentrated matt resulting from the
black copper smelting. Poor matts can usually be roasted
sufficiently in eight fires, the first four being inexpensive, as
a sufficient amount of sulphur is present. If at the fifth
firing concentrated matt is added, it will serve as fuel. The
temperature at the last fires must be increased, by a larger
foundation of wood and by an admixture of coal, until the
roasting mass cakes. The matt from the first roasting fires
is more apt to form deposits of iron at the concentration
smelting than matt from the later fires, for the reason that
at the former fire much peroxide of iron is chiefly formed by
desulphurising the sulphide of iron, owing to a deficiency of
coal and the presence of sufficient air. This peroxide of
iron is very apt to be reduced to metallic iron in the cupola
furnace. At the later firings a great part of the peroxide of
iron is transformed into protoxide by the action of the coal
and absence of much air; protoxide of iron, being a strong
base, is more easily scorified than reduced.
Bäntch has confuted Heine's opinion.
B. u. h. Ztg., 1859, p. 243.
↑ Bgwkfd., i., 49.
+
Ibid., i., 81.
48
COPPER.
The roasting of the raw matt is sometimes effected in
heaps, covered by a shed. The heaps, and the size of the
fragments of matt vary according to the amount of sulphur
which the matt contains; this amount is not sufficient to
admit of a roasting in the open air. The heaps at
Schmöllnitz are 18 feet broad, 70 feet long, and 1 feet high,
capable of containing about 1,700 cubic feet of matt. The
fragments of matt weigh about ½ lb. each.
I
This roasting is usually carried out in mounds (Mans-
feld, Freiberg, Sweden, &c). The form and dimensions.
of the mounds greatly influence the uniformity of the
roasting. Heinet recommends mounds 6 feet broad, 6 feet
long, and 5 feet high, and a roasting of 10 tons at the
first two firings, but not more than 5 or 6 tons at the later
firings. A larger quantity may be employed in the first
roastings, as the matt then contains sufficient sulphur, and
the fragments are large enough to lie loosely, and allow a
free permeation of the atmospheric air. At the first roasting
a foundation of wood is used with a layer of small coal upon
it; at the two following roastings a larger quantity of small
coal is added, and at the fourth and following firings the
matt is charged with alternate layers of coal. This addition
of coal raises the temperature and decomposes the anti-
moniates and arseniates which may be formed. Heine says
that the following conditions are necessary to obtain a good
roasting :-
The matt is broken into uniform and not too small
pieces, the finely divided matt is separated, the pieces which
have been roasted to a different degree are judiciously distri-
buted at the new firing, and pieces which are finished roasting
picked out. The concentrated matt of the black copper
smelting is distributed over the roasting mass, which is pro-
vided with a covering. The addition of coal is sufficient to
raise the temperature so high that the roasting mass cakes
strongly but does not melt.
At Boston, mounds provided with a grate and arched roof
are employed, allowing a quick and uniform roasting.
*
Bgwkfd., i., 51.
† Oesterr. Ztg., 1859, p. 292.
ROASTING THE RAW MATT.
49
The roasting of copper matt in cupola furnaces has occa-
sionally been attempted,* but with unsatisfactory results, on
account of the easy fusibility of the matt. At some Swedish
smelting workst this method of roasting is said to be carried
on advantageously. Copper matt poor in copper may bc
roasted in kilns if sulphurous acid is intended to be used for
the production of sulphuric acid; care must then be taken
that the admission of air is limited, otherwise the furnace
will cool too much; this roasting is not applicable to rich
matt.
+
+
Bredberg recommended roasting the matt in a finely
divided state in reverberatory furnaces to save the several
firings. His plan was tried in Freiberg, but was more ex-
pensive, and yielded an insufficiently roasted matt. Rever-
beratory furnaces, as a rule, are only employed when the
subsequent desilverising necessitates full control of the
roasting process, as in Ziervogel's process. They are used
at the smelting works on the Rhone§, and the volatilisation
of the antimony is facilitated by an alternate lowering and
raising of the temperature. T
As sulphur is the chief reducing agent when smelting
copper matt in reverberatory furnaces for the production of
black copper, the roasting process can only be carried on to
a certain degree; consequently antimony and arsenic can
only be imperfectly expelled; this may be done by repeated
firings, admixture of coal, &c., when smelting in cupola fur-
naces is intended, as in this process carbonic oxide gas has
a reducing action.
The chief product is roasted matt, consisting of metallic
oxides, salts, and more or less undecomposed sulphides,
according to whether it is to be smelted for the production of
concentrated matt, or for black copper. Well roasted matt is
*
ERDMANN'S Journal f. ök. u. techn. Chem. xvi., 58.
† Bgwkfd., i., 97.
+
ERDMANN'S Journal f. ök. u. techn. Chem. i., 56.
|| Jahrbuch f. d. Sächs. Berg. u. Hüttenm., 1838, p. 48.
§ B. u. h. Ztg., 1859, P. 243.
VOL. II.
Ibid., 1859, P. 251.
E
50
COPPER.
of a dull, bluish-black appearance, caked, soft, and easily
divisible.
Concentration Smelting of the Roasted Raw Matt.-
This process, like the raw smelting, aims at the scorification
of the foreign oxides (chiefly of the protoxide of iron), and the
reduction of the oxide of copper and of the sulphates to
sulphides, which then form a new matt by combining
with the sulphides undecomposed by the roasting process;
the matt also receives into combination any metallic
copper present. Antimoniates and arseniates are decomposed
in the same manner as at the ore smelting. If they are present
in a larger quantity they give rise to the formation of
speiss. When the roasting has been carried on too far,
an impure black copper results from the smelting, and if
such a formation is to be entirely avoided, as at Mansfeld
and Freiberg, where the raw matt is concentrated for de-
silverisation, a smelting in reverberatory furnaces is prefer-
able to one in cupola furnaces.
As the roasted matt is rich in oxides it requires for its
smelting an addition of siliceous fluxes, such as ore slags rich
in silica, and if they are unobtainable, clay, clay-slate, &c., are
substituted. The mixture is made more basic than that of the
ore smelting in order to avoid a scorification of copper by
excess of silica. An excess of protoxide of iron also pro-
tects the copper from scorification. On the other hand, the
basic slags chill more easily and enclose particles of matt,
which are, however, regained by using the slags as a flux in
the ore smelting; though this causes a greater consumption
of fuel, the expense is outweighed by a larger yield of copper.
Slags of the later smelting processes, rich in copper, are also
added to the mixture to extract the copper they contain,
according to the rule of adding the purer fluxes to the later
smeltings, and the impure fluxes to the raw smeltings.
In other respects the smelting process is carried out like
the ore smelting process, and yields the following products:--
1. Black Copper. This is only produced in small quan-
tities.
2. Concentrated Copper Matt.-According to its im-
purities this is either roasted and re-concentrated, thus
CONCENTRATION SMELTING OF THE ROASTED MATT. 51
yielding more black copper, or it is roasted and smelted for
the production of black copper, as in the Lower Hartz and
Schmöllnitz; at Mansfeld and Freiberg it is added to the
desilverisation process. Sometimes capillary copper* occurs
in the matt.
According to Le Play, such a copper from
blue copper matt of Wales contained-
Cu
Fe
Ni
Sand and coal
98.2
0'4
0.6
0°2
The composition of the concentrated matt is shown by
the following analyses :—
I.
II.
III.
Cu
57 27
61.26
72.743
Fe
16.32
13'70
8.638
Co
4'II
Sb
1'012
S
22.17
22.51
17.666
Pb
0*641
No. I is matt from Mansfeld. No. 2, matt from Riechels-
dorf, according to Genth, consists of 3Cu₂S, FeS, and according
to Rammelsberg, of 4(Cu₂S, FeS) Fe₂S. No. 3, matt from
Altenau, analysed by Hahn.t
3. Basic Slags, containing intermixed particles of matt.
These are composed as follows:-
I.
II.
III.
SiO
33.18
29.25
29*099
A1203
II'I2
I'24
4'275
FeO
32°03
63°32
60°513
CaO
17°14
I'475
MgO
2.96
I'30
0'588
Cu₂O
I'90
2.64
2.067
MnO
1*46
ΚΟ
O'18
SbO3
0 245
РЬО
0°431
No. I is from Mansfeld, and is a mono-silicate, analysed by
Wornum. No. 2, from Lautenthal (Upper Hartz), analysed
*
1 Jahrb. des Nass, Ver. f. Naturkunde, 1851, p. 13. BERGGEIST, 1861, p. 12.
PLATTNER'S Röstprocesse, p. 110.
† B. u. h. Ztg., 1861, p. 62.
E 2
52
COPPER.
by Walchner; this slag often crystallises in the shape of
chrysolite (3FeO,SiO3). No. 3, slag from Altenau, analysed
by Fritz Werlisch and Dr. Hahn.
Besides speiss, deposits of iron, &c., may result from this
smelting.
The Roasting of the Concentrated Matt is to effect the
complete volatilisation of sulphur, antimony, and arsenic,
and the oxidation of the copper and the foreign metals.
The scorification of the foreign metals must be effected at
the subsequent reducing and purifying smelting, whilst the
oxidised copper and part of the foreign metals are reduced,
forming black copper. As the roasting process cannot be
carried out quite perfectly, some undecomposed sulphides
will always remain, consequently some matt is formed at
the smelting process. This 'matt is very pure, yielding an
excellent copper, as most of the impurities combine with the
black copper; such a formation of matt prevents the scori-
fication of much copper, as sulphide of iron reacts upon
scorified copper, forming sulphide of copper and silicate of
iron. If the matt contains much antimony and arsenic it
may be roasted less, so as to produce a greater quantity of
pure matt as well as the impure black copper.
The roasting is effected like that of the raw matt, mostly
in sheltered heaps or in mounds, but a greater number of
firings are used (from 8 to 16), and more fuel is required as
the quantity of sulphur in the matt decreases. The greater
amount of fuel employed facilitates the volatilisation of anti-
mony and arsenic.
The roasted matt appears in clustered, porous masses of
a dull dark colour, and frequently contains metallic copper
and sub-oxide of copper by the reaction of the oxidised upon
the sulphuretted copper. It is sometimes lixiviated previous
to its smelting, with water or diluted sulphuric acid to extract
sulphate of copper.
The Smelting of the Roasted Concentrated Matt for
the Production of Black Copper.-The chemical reactions
of this process may be explained as follows:-If the matt,
together with a sufficient quantity of siliceous substances, is
SMELTING THE ROASTED CONCENTRATED MATT.
53
submitted to a reducing smelting at a suitable temperature,
the oxide of copper will gradually be reduced to metallic
copper by the carbonic oxide gas in the upper parts of the
furnace, metallic copper will be formed somewhat lower down
in the furnace by the reaction of oxidised copper upon sul-
phuretted compounds, the peroxide of iron reduced to
protoxide, and the small quantity of sulphates to sulphides.
The admixed siliceous substances combine with the protoxide
of iron in the zone of smelting, and the new slag thus formed
melts into the hearth together with the metallic copper and
the sulphides. Black copper, matt, and slag here separate
according to their respective specific gravities. As it is im-
possible completely to control the reducing action, part of
the foreign oxides is also reduced, and the liberated metals
partially combine with the copper, and some of them, chiefly
the iron, separate and form deposits on the sole of the
hearth. Some unreduced sub-oxide of copper enters the
slag, but is partially extracted by the liquid carbonised
iron, and partly by the sulphide of iron of the new
matt.
In order to avoid scorification of copper as much as pos-
sible, the addition of siliceous fluxes is made in such propor-
tions that either basic or mono-silicated slags result, and the
copper is then protected from scorification by protoxide of
iron. This basic slag has, on the other hand, the disadvan-
tages of rapidly corroding the furnace, of shortening the
smelting operations, and of quickly chilling and enclosing larger
quantities of particles of copper and matt; these slags must
therefore be smelted as an addition to former smeltings.
Pure slags rich in copper, such as copper refining slags, are
usually added to the black copper smelting; and at Boston
oxidised and siliceous ores and furnace ends are also added
for extracting their contained copper, and to make use of
their earths in the formation of slags. If they are added in
too large quantities, slags rich in sub-oxide of copper will
result, unless the roasted matt contains a sufficient quantity
of sulphur. At Atvidaberg part of the gases are made to
escape somewhat beneath the furnace mouth, in order to
limit the reduction of peroxide of iron.
54
COPPER.
The conditions which chiefly influence the process are-
1. The Composition of the Mixture. If the slag is too
siliceous much sub-oxide of copper becomes scorified, and a
too basic slag facilitates the formation of deposits of iron; it
also corrodes the interior of the furnace quicker, and the
separation of the smelted products is less perfect.
2. The Construction of the Furnace.-Channel furnaces
with two open eyes allow a greater production than sump
furnaces; for reasons already stated they also facilitate the
formation of black copper. On the other hand, sump fur-
naces produce purer slags, but they have sometimes the
disadvantage of allowing the black copper to chill in the
tapping hole when the process is carried on too carelessly,
or if the quantity of matt present is not large enough to keep
the copper sufficiently hot.
If deposits of iron are formed in larger quantities, channel
furnaces cannot be employed; their formation is prevented best
by a rapid smelting in low furnaces. In rare cases, as,
for instance, at Perm, a particular construction of the furnace
is adopted in order to facilitate the reducing reaction for the
production of a cupriferous iron. Sometimes (Kupferberg)
the ore smelting and the black copper smelting are carried
on alternately in the same channel furnace.
True crucible furnaces are scarcely used, as they do not
allow the removal of larger deposits of iron, but furnaces are
used at Atvidaberg and Fahlun constructed like crucible
furnaces.
Black copper may be smelted with less loss of metal in
cupola furnaces than in reverberatory furnaces, on account
of the oxidability of the copper, but cupola furnaces allow a
larger production, as the smelting is carried on almost
without an addition of slag, and they also produce a purer
copper and little matt. The waste of copper may be lessened
by intermixing and by covering the smelting mass with fuel.
At Schmöllnitz* the smelting of black copper is now effected
in reverberatory furnaces instead of in cupola furnaces, which
were formerly used.
* Oesterr. Ztschr., 1859, pp. 277, 299.
PRODUCTS OF THE BLACK COPPER SMELTING.
55
3. The Nature of the Fuel.-It has been before mentioned
that charcoal produces a tougher black copper than coke; if
black copper is to be desilverised by the wet way, coke may
be used. At Boston anthracite is used, and causes a very quick
smelting.
The following are the products of the black copper
smelting:-
1. Black Copper.-This is either refined at once or pre-
viously desilverised. Its composition is shown by the fol-
lowing analysis:-
I.
II.
III.
IV.
V.
32205284
Cu.
Fe .
92.83 710 69'5
99'400
138 ΙΙΟ
6.7
0°044
2.79
6'0
6.843
0*26
0'5
Ο ΙΟΟ
2.0
I'05
ΙΟ Ο
8.3
o'988
1'3
0'043
3°5
0'449
0'416
Ca
0°093
Mg
S
1'07 7'0
trace
0°3140
Mn.
Bi.
ΙΟ
Au .
0*1440
0.0008
No. I is black copper from Mansfeld, analysed by Ebbing-
haus. No. 2, from Riechelsdorf, analysed by Willie. No. 3,
black copper from Freiberg, analysed by Lampadius. No. 4,
black copper from South Australia, analysed by Levol.
No. 5, black copper from Altenau (Upper Hartz), produced
from lead matt, after being desilverised by liquation, analysed
by Hahn.
*
2. Matt (Dünnstein).—This is either roasted together
with the raw and concentrated matt, or worked by itself for
the production of black copper. It is sometimes roasted
and lixiviated for the extraction of sulphate of copper, and
occasionally it is added in small quantities at the refining of
the black copper, after being roasted dead. Experience has
* B. u. h. Ztg., 1861, Nos. 7 and S.
56
COPPER.
shown copper produced direct from the ores to be almost unser-
viceable; copper from smelting raw matt, of middling quality;
copper from smelting concentrated matt, of good quality; and
copper produced from the black copper matt (Dünnstein),
excellent, as the matt has been exposed to most firings.
3. Deposits of Iron.-These are worked up as before
described. The richer they are in copper the more easily
they are fused.
4. Black Copper Slags.-These are given back to the
smeltings.
5. Other Deposits, Furnace Ends, and Smoke.-These
are worked up in previous smelting processes.
Refining Black Copper.
As the smallest quantity of foreign admixtures injures the
ductility of the copper, or makes it porous, black copper,
usually containing from 70 to 98 per cent of copper, and
alloyed with iron, lead, zinc, nickel, cobalt, antimony, arsenic,
wolfram, molybdenum, bismuth, sulphur, &c., requires a
most carefully executed purifying process.
The refined copper produced in most German copper
works is delivered to manufacturers in two forms-as,rosette
copper (rohgaar), or in thin round discs adapted for casting
and the making of alloys, but not sufficiently malleable for
hammering and rolling; and as refined or poled copper,
which is sold in the form of ingots or bars, and possesses,
like the best classes of English copper, the property of malle-
ability in a high degree at all temperatures. This latter
copper is technically said to be "at tough pitch" by English
smelters, and "hammergaar" by German smelters.
Rosette copper is produced by a powerful oxidising smelting
(first refining) which causes the separation of those foreign
substances having affinity for oxygen, partly by volatilisation
and partly by scorification, whilst other foreign substances
with less affinity to oxygen, such as gold and silver, remain
with the copper. This copper is more or less contaminated
with suboxide of copper, requiring for its complete purifica-
tion, so as to render it malleable or toughened, a quick
reducing smelting (second refining).
REFINING THE BLACK COPPER.
57
These two refining operations may be carried out with the
following modifications :—
1. First Refining (Oxidising Smelting) of Black Copper.
a. In a small open hearth heated by means of coked fuel.
This process being a sustained contest between oxidation
and reduction, and consuming much fuel, is used chiefly when
working small quantities and tolerably pure black copper
alloyed principally with iron, as it is very simple and may be
interrupted at any time. Other substances, such as antimony,
arsenic, lead, &c., would, if extracted by oxidation, be partly
reduced at the unavoidable contact with coal, and so again
alloy the copper and make the process almost interminable.
b. In a blast reverberatory furnace (Spleissöfen). This
apparatus allows the purification of impure copper in larger
quantities, and with the use of raw fuel, as the fuel does
not come into contact with the copper; but the resulting slags
are richer in copper than those of the small hearth, in
which the scorified copper is again partly reduced by contact
with fuel. Most of the copper contained in the former slags
is extracted by a separate process. At Perm some of the
copper in the slags is extracted at the refining process by an
addition of pyritic ores.
Very impure black copper is refined by dividing the
process into two parts, as a continued process would injure
the furnace and cause more copper to be scorified.
c. At Kongsberg the refining process is effected on the
movable hearth of a reverberatory furnace, using gas for
firing. The test is formed of pulverised slag, and constructed
like those used for refining silver. By lowering the hearth
its contents may be poured into moulds. This fuel-saving
arrangement allows a quick and effective operation, as the fuel
does not come into contact with the copper. The resulting
copper contains suboxide of copper.
2. Second Refining (Reducing Smelting) of Rosette
Copper.—a. This process is conducted in most of the German
copper works with rosette copper on the small hearth.
b. Refining black copper in air reverberatory furnaces.
-The copper is first submitted to an oxidising smelting, pro-
ducing copper containing suboxide of copper, and afterwards
58
COPPER.
rendered malleable by stirring it with wooden poles.
This process, which forms part of the English copper
smelting process, is also used in some German copper works,
Mansfeld, Freiberg, Stadtbergen, &c., for refining black
copper produced in cupola furnaces; the production is larger,
the expenses for wages and fuel are less, and the copper is
purer than that resulting from the process on the hearth;
though the yield of copper is somewhat smaller the advan-
tages of the two are about equal.
Blast reverberatory furnaces (Spleissöfen) are less adapted
to the refining process than air reverberatory furnaces, as
the admission of air, and the temperature required for stirring
with wood, cannot be sufficiently controlled, and they do not
allow a quick removal of the copper, which is indispensable
for preserving it in its pure state. This process will be
furthur explained when describing the English copper
smelting processes.
Refining the Black Copper on Small Hearths.
The construction of these apparatus is shown by
Figs. 1, 2, 3;* Fig. 1 being the section lengthwise, Fig. 2,
FIG. I.

9
α
b
the section across, and Fig. 3 the ground plan; in which a
is the hollow of the hearth; b, a massive wall; c, the mass
of which the hearth is formed; d, d, cast-iron plates
FIG. 2.

h
a
W
100%
www.
* Dr. URE's Dictionary of Arts, &c., i., 884.
REFINING BLACK COPPER ON SMALL HEARTHS.
59
covering the hearth; e, the opening for running off the
liquid slag; f, a small wall; g, an iron curb for keeping the
coals together. The mass, c, consists of a mixture of 7 or
FIG. 3.

9
1,
Ι
S volumes of clay and I volume of small coal with some sand;
the sand is added in order to make the mass porous, to
prevent its cracking, and to facilitate drying. The quantity
of copper to be treated fixes the dimensions of the hollow in
the hearth, which is usually from 8 to 12 inches deep and
from 18 to 24 inches in diameter. The charges of copper
are usually from 3 to 8 cwts., but in Sweden they are from
18 to 30 cwts.
The hearth is first filled with coal, and some burning coal
is thrown before the tuyere; pieces of black copper are now
placed upon the coal about 7 inches distant from the back
wall, the smaller pieces lying on the top of the larger pieces,
and the thickest part being turned towards the tuyere. The
smooth side of the pieces is also turned lowest, so that they
may be readily shifted. If small pieces are too near the
tuyere they are liable to fall into the fused metal and cool it.
After having filled up the space between the copper and
the back wall with coal the blast is put on, and the black
copper is carefully shifted towards the tuyere as it melts off.
The coals before the tuyere are carefully stirred and replaced
by fresh coal when required to prevent the formation of any
hollow spaces.
Care must also be taken not to add coal too
60
COPPER.
often or too much at once, or the mass will cool. Channels
are formed in the coal by means of an iron bar in order to
conduct the flame to the black copper. Coke is seldom used
in this process. The more impure the copper is, the slower
must be the melting, thus exposing the dropping copper
longer to the blast; foreign substances will be oxidised and
the refining of the copper facilitated. Zinc, antimony, arsenic,
and lead become volatilised and give rise to the formation of
deposits on the back wall above the tuyere. Generally, impure
copper melts quicker than pure copper, and rugged pieces
quicker than smooth ones.
As the hearth is not filled by the first charge, an after-
charging is required, otherwise the blast does not sufficiently
react on the metal, and the process is retarded; this after-
charging is varied according to the degree of the fusibility of
the slag. If the slag is easily fusible, containing, for
instance, lead, which flows off by itself with little assistance,
coal and copper are charged alternately, until the hearth is
filled with metal. If the slag is difficult to fuse, containing
more iron than lead, and having the coal of the first charge and
the slag sticking to it, the tuyere, the border of the hearth,
and the slag covering the metal-bath are all to be removed.
For this purpose the latter slag is sometimes cooled with
water, as at Atvidaberg, and some quartz is added to it. The
nose is broken off the tuyere and thrown into the fused
metal; coal is again added, and the blast put in motion,
whilst the upper three or four metal discs of the last refining
process are put on the top of the coals, so keeping coal and
heat together. The slags are thus removed once or twice.
more during the smelting, which lessens the reduction of
oxidised substances, and offers a metallic surface to the
blast; the colour, brightness, and undulation of this surface
indicate the nature fo the copper. A dull surface shows too
low a temperature, and coal must be added, and the blast
continued; but if the metal is bright, green, boiling, and
sparkling, the temperature is too high, and time must be
allowed for cooling before adding more coal; the metal
must be kept covered with coal till the refining is finished,
and the blast blowing through the coal reacts on the metal.
CHEMICAL REACTIONS OF THE REFINING PROCESS. 61
In Kongsberg and Müsen they attempt to hasten the process
by an addition of washed concentrated matt.
The chemical reactions of this process are as follows:-
Part of the impurities, chiefly those metals with a strong
affinity to oxygen, such as antimony, sulphur, arsenic, lead,
and zinc, are volatilised, and the zinc, lead, iron, &c., are
scorified. According to the quality of the black copper, the
slags or hearth ends consist of a fused mass when silicates
from the hearth have entered into the combination, or if
oxide of lead is present in larger quantity, otherwise the
hearth ends consist of a mixture of half fused oxides, chiefly
of iron, nickel, and cobalt. According to its nature, the slag
either runs off by itself or must be removed.
By the
reaction of the blast upon the surface of copper, a small
quantity of foreign metals becomes oxidised together with
copper,* though the affinity of copper for oxygen is less than
that of these foreign metals. The suboxide of copper
formed at the surface gradually mixes with the fused copper,
and, by giving up its oxygen to those metals which have
affinity for it, lessens their specific gravity, and they rise to
the surface and there become scorified. In this manner iron
and lead are most readily removed. Cobalt and nickel,t
having nearly the same affinity for oxygen as copper, can
only be extracted from the copper if a considerable surplus
of suboxide of copper is present. This, however, causes a
scorification of much copper, and the production of copper
rich in suboxide (übergaar), the upper rosettes of which are
frequently so rich in nickel that in Mansfeld they require to
be submitted to a further treatment, as nickel injures the
quality of the copper, and is in itself valuable. The upper
rosettes of copper sometimes contain crystals of oxide of
nickel. Slags containing nickel are difficult to fuse, often
forming a loose, sandy, and but slightly caked mass.
Antimony and arsenic possessing great affinity for copper
can scarcely be completely separated from it, even if
the copper contains a very considerable surplus of sub-
B. u. h. Ztg., 1860, p. 442.
LAMPAD., Fortschr., 1839, p. 139.
1861, p. 133.
Bgwkfd., x., 331. Oesterr. Ztschr.,
62
COPPER.
oxide.
arsenic.
Antimony is still more difficult to separate than
These substances give rise to the following complications.
in the process, chiefly at the commencement. The surface
of the metal bath remains bright, continually emitting
vapours, and scorification cannot take place before this
volatilisation of metals has ceased, as the vapours prevent
oxidation of the impurities of the copper. As soon as the
scorification commences, the metal bath becomes covered
with a hard gritty crust, which renders further oxidation
difficult, and prolongs the process. It is, therefore, advisable
to extract antimony and arsenic at the preceding roasting
and smelting processes, especially as the smallest quantity
of these substances injures the ductility of the copper.
If black copper contains nickel as well as antimony, a
chemical compound, (CuO,NiO)SbO,, called copper mica,
is frequently formed. This then reacts like suboxide of
copper upon the quality of the refined copper, with which
it completely mixes; it forms hexangular laminæ, about one-
twelfth of an inch across, of a golden yellow or copper red
colour, transparent and bright. The following analyses of
copper mica from the Upper Hartz show its composition.
IV.
I.
II.
III.
CuO.
Nio.
46*32
43.38
43°72
67.648
28.26
29*23
39°50
16.101
SbO3
PbO.
24'53 26.57 17.99 18.018
I'69
No. I is copper mica from Lautenthal, analysed by Pfann-
kuch. No. 2, from Andreasberg, by Rammelsberg. No. 3, from
Altenau, by Ramdohr. No. 4, from Lautenthal, by Hahn.
The presence of sulphur in black copper gives rise to
the following singular and important behaviour. When
smelting and refining copper, part of the sulphur goes
off as sulphurous acid, but part remains in the copper,
owing to the great affinity of these substances for each
other, even if copper contains suboxide of copper. If
the copper, containing sulphide and suboxide in a
suitable proportion, is now cooled to a certain degree, the
contents react upon each other, producing sulphurous
acid, which is rapidly evolved, carrying away some copper
MODIFICATIONS OF THE REFINING PROcess.
63
in the form of more or less fine globules. Upon smelting
pure black copper this phenomenon generally occurs at the
end of the refining process, when the copper contains only a
small amount of suboxide, and in such cases condensing
flues are useful. Impure copper, requiring a stronger refining
(thus rendering it rich in suboxide of copper) shows this
phenomenon in a less degree, and occasionally it is absent.
According to Karsten it occurs when the copper contains
about 0.8 per cent of suboxide of copper, but not if it con-
tains 125 per cent. Sulphide and suboxide of copper in
certain proportions seem to form sulphurous acid when
reacting upon each other at a particular low temperature,-
CuS + 2Cu₂O
=
SO₂+5Cu,
or the reaction may take place at a high temperature, in
which case the sulphurous acid is absorbed by or kept in
chemical combination with the copper, and is liberated at a
lower temperature, in the same manner as fused silver and
litharge absorb oxygen; part of the sulphurous acid appears
to remain combined with copper, which is rich in suboxide,
rendering the copper porous.
Explosions occurring whilst granulating copper are
probably caused by the emission of sulphurous acid and the
development of steam. The following modifications are
recommended for accelerating the refining process:-
An addition of about 1 per cent of lead, when treating copper
containing nickel, arsenic, antimony, and much sub-oxide of
copper. The lead acts either by combining with antimony
and arsenic on sinking to the bottom, or by becoming oxi-
dised. Part of the oxide of lead then gives its oxygen to
foreign substances, while another part scorifies together with
other oxides, chiefly copper mica. In many cases lead has
a purifying action upon copper, but it never produces a per-
fect product, as some lead remains with the copper, and
copper containing only o'r per cent is unfit for fine manu-
factures.
Stirring the copper with wooden poles* is effective chiefly
in the case of copper rich in sub-oxide, but it is seldom resorted
*
Bgwkfd., xix., 88. Oesterr. Ztschr., 1856, No. 3.
64
COPPER.
to as it requires much time and causes a great consumption
of fuel.
Bredberg* suggested an addition of quartz to shorten the
process and facilitate the scorification of lead, but the results
of experimentst made at the Lower Hartz were unsatis-
factory.
Fluxes containing potash and lime‡ react very little, but
it seems that at the copper smelting process in India|| these
substances contained in the ash of the fuel, are made to react
through the oxygen they contain.
Fluxes containing chlorine are recommended for the ex-
traction of sulphur, antimony, and arsenic.§
The nature of the fuel essentially influences the process;
coke and hard charcoal, by producing a higher temperature,
facilitate the smelting and retard the refining. The use of
hot blast ¶ also facilitates the smelting, and aids the reduction
of scorified oxides.
The completion, or the approach of the completion, of the
process, is observed-
I. By the Nature of the Flame, which at the com-
mencement of the process is green with a red point at the
end.
2. By the Appearance of the Surface.-The copper is
considered refined if its surface is bright, sea-green, and
has few white spots. It is over refined, containing much
sub-oxide of copper, if the surface is very undulating, bluish, or
sea-green, and with many white spots. It is considered too
raw, i.e., still alloyed with foreign substances, if it has
many dark spots.
It
3. By the Nature of the Slag.-According to the im-
purities of the copper it is more or less fusible, and at the
commencement of the process dark or green coloured.
then takes a brown colour, turning red as the process ap-
proaches completion.
Bgwkfd., v., 321; xiii., 457.
↑ KERL, Rammelsberger Hüttenpr., 1861, p. 125.
LAMPAD., Fortschr., 1839, p. 140.
B. u. h. Ztg., 1862, p. 117.
$ Bgwkfd., i., 129.
[
Ibid., i., 185. MERBACH, Anwend. d. erhitzten Gebläseluft, 1840, pp. 231,
245. KERL, Rammelsberger, Hüttenpr., 1861, p. 123.
PRODUCTS OF THE REFINING PROCESS.
65
The fineness is tested by means of a proof rod of iron thrust
through the tuyere into the melted copper, then drawn out
and plunged into cold water. The refining is finished as
soon as the copper scale appears brownish-red on the outside,
and copper-red within, thin as net-work, and so deficient in
tenacity as to break when strongly bent.
The blast is now stopped, and after being left to cool, the
coals and cinders covering the surface are raked off the
copper. The surface is further cooled by sprinkling water
upon it, and the cakes of congealed metal (discs or rosettes)
are lifted off with tongs, a process called slicing (schleissen),
or shaving (scheibenreissen); this is continued until the with-
drawal of the last convex cake at the bottom of the furnace,
styled the kingpiece. These discs are immediately immersed
in cold water to prevent oxidation of the copper, and the
metal then becomes of a cochineal-red colour, and covered
with a thin film of sub-oxide. Its under surface is studded
over with points and hooks, the result of raising the con-
gealed disc from the liquid metal; these cakes are called
rosette copper. When the metal is pure and free from oxide
these cakes may be obtained very thin, 1-24th of an inch for
example.
Thick yellow coloured cakes indicate too raw a copper,
and deep red cakes a copper rich in oxide. The latter copper
is in some cases of greater value than thin cakes if it con-
tains small amounts of antimony, arsenic, &c. Sometimes
in order to obtain thin cakes lead is fraudulently added to
the copper after the slag has been removed.
Thick cakes are also obtained if too much water is em-
ployed for cooling, and if too little water is used the resulting
cakes are not sufficiently convex, and separate with difficulty
from the metal bath.
Explosions are likely to occur if the cooling with water is
not effected very carefully; therefore in some copper works
in Sweden* the copper is ladled out into moulds.
This purifying process yields the following products :-
1. Purified Copper (Rosette copper).-Owing to a small
VOL. II.
Bgwkfd., v., 325; xiii., 458. B. u. h. Ztg., 1859. p. 83.
F
66
COPPER.
amount of foreign substances, chiefly sub-oxide of copper, it
is brittle and unfit both for rolling and for hammering. A
reducing smelting will toughen it. The greater part of nickel
present* will be found in the first cakes, and at Dillenburg+
one of these cakes was found to contain octohedral, ruby-red
crystals of pure oxide of nickel.
The composition of these different kinds of copper is shown
by the following analyses :-
I.
II.
III.
IV.
V.
VI.
Cu.
98*48 99*45 98.25 83.90
98 97
98*97
Pb
trace
0'14
I'09 0.60
0'07
99*480
0*362
Fe
0*75
0'02
0*13
0'23
Ni
0*26
0*24 ΙΙΟ
0'27
Ag
0'04
0'14 trace
0'13
Ο ΙΟΟ
Sb
0.60
0'19
As
trace
K
0°32
0'07
Ca
Mg
Al
Si
S
Zn
11
1 1 1 1
Ο'ΙΙ
ΟΙΟ
0'04
0'05 O'12
trace
trace
0'02
Au
Bi.
NiO
trace
trace
0'0008
0*0480
13.86
No. 1, rosette copper from Andreasberg, analysed by Bode-
mann. No. 2, from the Lower Hartz, by Biewend. No. 3,
from Mansfeld, by Kobell. No. 4, from Riechelsdorf, first
cake, by Genth. No. 5, the same, last cake. No. 6, from
South Australia, by Levol, brittle owing to the presence of
bismuth.t
2. Slags. They are either added to former smeltings or
worked by themselves for the production of an inferior
copper. If they contain a larger amount of nickel, they are
worked with an addition of arsenical pyrites and heavy spar
Bgwkfd., x., 321; xii., 223.
† Ann. d. Chem., Bd. 53, p. 139. GURLT., pyrog. Minèr., p. 46. Berzelius,
Jahresber, 1857, Hft., i. RAMMELSBERG'S Métallurgie, pp. 210, 248.
+
DINGLER'S Polyt. Journ., cxxxi., 268.
PURIFYING COPPER IN REVERBERATORY FURNACES. 67
for the production of nickel speiss, and copper matt, or of
black copper.
3. Hearth, containing much copper.
Purifying of Black Copper in Blast Reverberatory
Furnaces.
The furnaces generally employed in the Hartz, in the
district of Mansfeld, Saxony, Hungary, &c., are represented
by Figs. 4, 5, and 6.* Fig 4. represents the elevation of the
FIG. 4.

furnace across the line I K of the plan, Fig. 5, which is
taken at the level of the tuyere, n, of Fig. 6; Fig. 6 is a vertical
section on the line L, M, Fig 5. K represents one of two
Y
FIG. 5.

I
下
​reception basins, brasqued with clay and charcoal; n, n, two
tuyeres through which the blast enters; q, door by which
* Dr. URE's Dictionary of Arts, vol. i. p., 886.
F 2
68
COPPER.
the metal to be melted is laid upon the sole of the furnace ;
v, v, two points where the sole is perforated to run off the
melted copper into either of the basins when necessary; x,
door through which the slags of cinders floating upon the
surface of the melted metal are raked out; y, door of the
fire place. The fuel is laid upon a grate above an ash pit,
and below the arch which is contiguous to the dome or cap
of the furnace properly so called. In the section, Fig. 6, the
FIG. 6.

k
Ղ
་་་་
n
following parts may be observed:-1, 2, 3, the masonry of
the foundation; 4, vapour channels or conduits for the escape
of moisture; 5, bed of clay; 6, brasque composed of clay
and charcoal, which forms the concavity of the hearth.
At Tajova,* the arched roof is less convex, causing a saving
of fuel and a shortening of the process.
The furnace is charged with from 20 to 70 cwts. of black
copper, and the cakes are so arranged in the furnace as to
The copper
leave a suitable space between for the flame.
becomes red hot after a firing of from 3 to 6 hours, and then
the blast is put in motion; pieces of wood are from time to
time placed between the copper and the tuyeres. The copper
is melted in about 10 hours, and the caked crust formed upon
it, consisting of oxides difficult to fuse, is removed, and a
new slag is formed by the reaction of the blast, whilst less
fire is given. This slag is again removed, and these operations
are repeated till at last the slag, which in the beginning is
easily fusible and dark coloured, becomes pasty and red from
the scorified suboxide of copper.
* Oesterr. Ztschr., 1857, p. 130.
PURIFYING COPPER IN REVERBERATORY FURNACES.
69
A red slag indicates the completion of the process and the
fineness of the copper, and is now tested by a proof rod of
iron as before described. If the copper is sufficiently pure
it is allowed to run out of the furnace into two outside basins,
where it is formed into cakes, as in the small hearth.
Besides the refined copper a great quantity of slags rich in
copper results from this process, and they are either treated
as the slags resulting from the process in the small hearth,
or they are purified from copper in the refining process itself.
At Perm, this is effected in the following manner:-After
having melted 3 tons 4 cwts. of copper for 12 hours the slags
are allowed to remain, and 8 cwts. of pure pyritic ores, con-
taining about 5 per cent of copper, are spread out upon the
slag whilst the blast is interrupted, and they are mixed with the
slags as soon as they have become pasty from cooling. The
working door is now closed, and a stronger heat is given for
about three hours. The slag is then removed and a new slag
formed by directing the blast upon the bath of metal. This
slag is removed until its colour indicates approaching fine-
ness of the copper, when the blast is stopped, and a very
strong heat given, causing the development of sulphurous
acid together with a strong bubbling of the fused mass.
soon as this ceases, a sample of copper is taken out by means
of a ladle coated with loam, and the fineness of the copper is
ascertained from the colour of its surface and of the fracture.
The resulting copper contains some suboxide, and is poured
into an outside basin, from which it is rapidly ladled into
cast-iron moulds.
As
The sulphide of iron contained in the added pyritic ores
sulphurises the copper in the slags, and the siliceous gangue
combines with the oxides produced by the smelting, forming
easily fusible slags. The formed sulphide of copper mixes
with the copper, and most of its contained sulphur is ex-
tracted towards the end of the process when an excess of
sub-oxide of copper is present, forming sulphurous acid with
the oxygen of the sub-oxide. It is important that the pyrites
should react upon the slag for a sufficiently long time.
RIVOT, Métallurgie du Cuivre, 1859, p. 102.
70
COPPER.
Very impure black copper is purified in two operations,
otherwise the furnaces would suffer too much, and the loss
of copper by scorification would be too great. The first opera-
tion, by which the copper is brought only to a certain degree
of purity, is usually carried on in the reverberatory furnaces.
The copper is then tapped off into an outside basin and
formed into cakes by cooling with water in the usual manner.
The loss of copper is smallest when impure copper is sub-
mitted to an oxidising smelting in reverberatory furnaces till
larger quantities of copper begin to scorify, and when the
purifying is finished in the smaller hearth. The first opera-
tion is considerably shortened by conducting the blast under
the grate.
Plumbiferous copper resulting from the liquation process.
may be treated in various ways. It may be melted on the
hearth of a reverberatory furnace and exposed to the oxi-
dising action of the blast; or it may be heated without under-
going fusion with the admission of atmospheric air, when
only an oxidation of lead and copper on the surface will take
place. The oxidised copper is again reduced by the particles
of lead with which it is in contact, thus interfering with the
equilibrium, to recover which the lead moves from the in-
terior of the mass towards the surface, until it is all oxidised,
forming a slag. The smelting of the copper effects a stronger
oxidation, but the greater part of the silver in the black copper
remains with the copper, a small amount only entering the
slags. The operation of merely heating the copper allows a
partial extraction of the silver, as it follows the lead, concen-
trating in the crust which is formed by suddenly cooling the
black copper.
Copper containing much antimony and
arsenic is treated in Hungary so as to form a considerable
amount of sub-oxide of copper; the blast is then stopped, and
a certain quantity of charcoal thrown upon the surface of the
copper to react upon it until all the suboxide is reduced; the
remaining coal is now removed, and the operations of forming
sub-oxide and reducing it are repeated in proportion to the
impurity of the black copper.
*
Jahrbuch, d. K. K. Reichsanstalt, 1853, No. 4, p. 626.
REFINING OF THE COPPER IN SMALL HEARTHS.
71
This method, somewhat modified, is adopted at Schmöll-
nitz* for the treatment of antimony copper speiss. This
speiss is converted in a reverberatory furnace at a slow
firing, into a pasty state, the fire is then increased and the
oxidising blast applied, and oxide of antimony is volatilised
in thick vapour, which may be conducted into a flue and
collected for the production of antimony. When vapour
ceases to be evolved the blast is stopped, the slags (chiefly
consisting of antimoniates of copper, iron, and antimony,
and some antimonial iron) are removed, and the blast is re-
applied without raising the temperature, by which operation
a strong volatilisation of antimony will again take place;
when this ceases, the metal bath is covered with a crust of
antimoniate of antimony, which may be decomposed by
mixing and stirring it with saw- or coal-dust, while a stronger
fire is given, and the blast removed. These operations are
repeated until the copper is sufficiently fine.
In order to eliminate the antimony as quickly as possible,
the speiss must be treated at a low temperature, as at too
high a temperature the antimoniate of antimony melts,
forming a fused covering upon the metal bath, and preventing
further extraction of antimony.
Second Refining of Copper in a Small Hearth.
Copper resulting from the first refining always contains
suboxide of copper, and is consequently brittle, and cracks
when treated with a hammer or roller. Karsten states that
copper containing 1'1 per cent of suboxide cracks at the
edges when worked at the common temperature (cold short-
ness) and when containing 1'5 per cent it cracks if worked
at a higher temperature (red shortness); suboxide of copper
may be reduced by a quick reducing smelting, yielding a
ductile copper. If in this process the copper is kept too
long in contact with the reducing agent, the copper is
supposed to combine with carbon, and become more brittle.
than before. Other foreign substances, chiefly antimony,
arsenic, and sulphur, if not removed by the first refining,
* Oesterr. Ztschr., 1859, p. 325, 329.
72
COPPER.
cannot be extracted by the second refining, except with
peculiar reagents, which lessen the yield of copper.
Copper is one of the most sensitive metals with regard to
impurities, as the smallest quantity often impairs its
ductility.
In rare cases only are both refining processes combined
in one.
The hearth employed for the first refining process is used
for the second, with the difference that the tuyere is put
several inches above the edge of the hearth, and is directed
at an inclination of from 6° to 10°; by these means the
blast blows mostly into the coals, producing a reducing
atmosphere of carbonic oxide. At the first refining the
tuyere has an inclination of 20° and more, and is put close
above the surface of the metal bath. The diameter of the
hearth varies according to the charge which, at the Lower
Hartz, for instance, amounts to about 5 cwts., and at Grün-
thal to 20 cwts.
After having warmed the hearth and filled it high up with
charcoal, the copper is added opposite the tuyere, and against
the border of the hearth, where it is not touched by the
blast; it is covered with coal and the blast put in slow
motion, which is gradually increased until all the coal is
burning. Care must be taken that there is plenty of coal.
before the tuyere for producing carbonic oxide gas; that the
pieces of copper are kept upon coals, in the combustion zone;
and that, melting uniformly, they flow through the coal; the
solid masses of copper must also be kept covered with coal.
As soon as the copper is melted, a sample is taken either
by a proof rod of iron or by a ladle.
If the proof rod of iron has been used the scale of
copper is quickly removed, and it is ascertained whether
it may be bent or beaten together at the ordinary and
at higher temperatures without the edges cracking. If
this is the case with several samples quickly tested, the
blast is stopped, the coals are removed, and the liquid
copper is poured into cast iron moulds coated with loam. If
the scale of copper is yellow coloured, showing a granular
texture, the copper has been too long in contact with coal
THE RISING OF COPPER.
73
and has become red-short; the coals are then partially
removed, and the blast is conducted into the hearth with a
somewhat greater inclination of the tuyere, till a new
sample shows the copper to be ductile; fresh copper is
sometimes added.
After having removed coal and scum from the copper a
sample is taken by means of a ladle. When cooled it is
broken, and the appearance of the fracture* shows the state
of the process.
Fine copper shows an uneven fibrous
fracture, a silky lustre, and rose-red colour, and is very
ductile. Carboniferous copper has a coarse granular crystal-
line fracture, and a yellow colour.
At the Lower Hartz and Grünthal, hot blast is used, and
has the advantage of helping the smelting and economising
fuel.
Some sorts of copper rise after having been poured into
the moulds, the ingots thus become more or less porous
with a convex surface, whilst good copper has a solid
fracture, and chills with a smooth surface. This vesi-
cular copper may be quite as ductile as solid copper, but
the products manufactured from it have flaws; it is, there-
fore, re-melted and again submitted to the refining process.
A blistered ingot breaks more easily than a solid one, and has
not the silky lustre, but the blisters show a fine red colour‡
when light falls into them reflecting towards the observer.
The phenomenon of the rising of copper is explained in
several ways, but almost all agree that it is a purely mechani-
cal effect, and depends on the rapidity of the chilling of the
copper. The interior of the ingot is supposed to contract
from the chilling of the surface, forming blisters; this may
be avoided by casting the copper at a certain moderate
temperature in quickly chilling moulds of metal, and not in
moulds of loam.||
Copper has the suitable temperature if a drop of water
* Oesterr. Ztschr., 1862, p. 5. Bgwkfd., xix., p. 773.
+ KERL, Rammelsberger Hüttenprocesse, 1861, p. 114.
B. u. h. Ztg., 1856, p. 3, 4, 5.
|| Bgwkfd., xii., 537. DINGL., Polyt., Journ., Bd., 154, p. 193. B. u. h.
Ztg., 1860, p. 320.
74
COPPER.
put upon its surface remains in the middle of it in the form
of a globule; the suitability of the temperature may also be
judged by the colour, movement, &c., of the copper.
According to Stölzel* this phenomenon of the rising of
copper is not to be confounded with spitting, which is
caused by an absorption of oxygen, and may be avoided
when re-melting pure copper by preventing the admission of
atmospheric air at the smelting process as well as at the
casting of the copper; it must, therefore, be melted under
a cover of coal, and poured into closed oiled moulds.
Coverings containing oxygen, such as borax, carbonate of soda,
and glass, are said to render pure copper porous; common
salt does not produce this effect, but renders it brittle.
The formation of sulphurous acid appears to be the chief
cause of the spitting as well as the rising of the copper,
when we take into consideration the behaviour of black
copper when being refined in reverberatory furnaces, and
the action of carbon as observed by Russell and
Matthiessen. The spitting takes place before the rising, and
happens if at the period of oxidation a certain quantity of
suboxide of copper has been formed in comparison to the
sulphide of copper present. The last particles of sulphur
are pertinaciously retained by the copper, and only con-
verted into sulphurous acid by a great excess of suboxide;
the sulphurous acid is then absorbed by the copper con-
taining suboxide, and the rising (formation of blisters)
commences together with this absorption. Sulphurous acid
is not absorbed by pure copper, but by that which contains
some suboxide of copper. In the reverberatory furnace pro-
cess the suboxide is reduced by stirring the copper with
wooden poles, whereby the copper loses its capability of
absorbing sulphurous acid, which is consequently evolved,
being facilitated by the boiling of the copper, caused by the
stirring; the resulting copper is not porous. The escape of
sulphurous acid may be observed by the smell.
The rising of the copper in the moulds also seems to ori-
ginate either from the formation or retention of sulphurous.
* B, u. h. Ztg., 1860, p. 320.
REMEDIES AGAINST THE RISING OF COPPER.
75
acid, if at the same time sub-oxide of copper is present, and
it may be facilitated by ladling the copper at too high a tem-
perature, as has been proved by experiments made in several
copper works. Sulphur may be contained either in the
copper or in the fuel.
This supposition is confirmed by the advantageous results
of the application of the following remedies:
1. Stirring the Copper with Wooden Poles.
2. An Addition of Lead.-This is oxidised; the oxide of
lead formed combines with the sub-oxide of copper, and
liberates the absorbed sulphurous acid.
I
3. An Addition of Chloride of Mercury.-At Perm* this
reagent is added to the refined copper before it is ladled out,
chiefly to copper intended for the manufacture of copper
plates for percussion caps, and is said to render the copper
more solid and ductile. The sublimate (about 1 lb. to 80 lbs.
of copper) is either thrown upon the surface, when it vola-
tilises, or it is stirred into the copper by a wooden pole. The
purification of the copper which takes place may be caused
either by the chlorine of the sublimate combining with anti-
mony, arsenic, and sulphur, or by a mechanical expulsion of
absorbed sulphurous acid, as on sinking to the bottom the
sublimate takes the form of gas agitating the mass.
4. Casting the Copper with Atmospheric Air Excluded
as much as possible.-According to the investigations
of A. Dick, copper containing sulphur forms a porous ingot
when poured in the usual manner into open moulds, but the
same copper forms perfectly solid ingots if the covered mould
is filled with coal gas, by throwing some finely pulverised coal
into it, and if the mould is filled as quickly as possible. If
air is admitted, sub-oxide of copper will be formed, which,
according to Dick, comes into contact with sulphide of copper
during the movement of the metal whilst chilling, and at this
moment sulphurous acid is formed, causing the rising of the
copper. Chemically pure electrotype copper does not rise even
when air is admitted, unless, indeed, sulphur is added in the
Bgwkfd., xix., 573. Oesterr. Ztschr., 1862, p. 5.
† B. u. h. Ztg., 1856, p. 345. PERCY'S Metallurgy, 1861, i., p. 275.
76
COPPER.
fuel. When required very solid, rolls are produced by casting
under pressure in iron moulds; or a minute proportion of
phosphorus is added to the metal (Parkes's process).
Refining Black Copper by Poling.
This refining by means of stirring with wooden poles cannot
be effected as advantageously in small hearths as in air rever-
beratory furnaces, as the metallic mass is too large in propor-
tion to its surface, and not sufficiently under control. The
combinations formed of hydrogen with arsenic, antimony, and
sulphur also partially decompose when passing the glowing
coals, which is less the case in reverberatory furnaces.
Although requiring more time and consuming more fuel,
this method is adopted in Hungary,* Ural, Mansfeld,† Brix-
legg, and Achenrain.
+
+
According to Rössner, the fused and descorified metal is
covered with coal and exposed for several minutes to the blast,
which reacts through the coal forming sub-oxide of copper.
Sufficient of the sub-oxide of copper is present when a sample
of the copper is very brittle, and shows a darkish brick-red
colour, a dull fracture and blisters. The copper bencath the
cover of coal is now stirred with a pole of birch wood, or
some other wood containing resin, until the copper has the
appearance of being overpoled. The blast is then made.
to react until a sample shows the copper to be fine. The
coals are removed and the copper quickly ladled out; the
stirring with poles must be repeated oftener (three or four
times) if the copper is more impure.
The black copper of Mansfeld is nearly pure, and is there-
fore treated with blast for 10 or 12 minutes, after a sample
taken by means of the proof rod of iron has shown the fineness
required for the first refining process. The slag is skimmed
off once or twice, and the copper is afterwards stirred with
wooden poles under a cover of fresh coal until quite fine.
The resulting ingots are hammered to make them more solid
and to remove the scale.
* Oesterr. Ztschr., 1855, p. 377. B. u. h. Ztg., 1855, p. 385.
† B. u. h. Ztg., 1861, p. 468.
Bgwkfd., xix., 88.
REFINED (TOUGH) COPPER.
77
The following products result from the refining or tough-
ening process:
Refined Copper, Tough Copper.-It is rarely crystallised.
Its composition is as follows:-
522
I.
II.
III.
IV.
V.
99'61 99*944 99'60-99'70 99*85 99*60-99*70
0*10- O'15
trace
0'02
Ag
Sn
trace
0*27
trace
0*056
O'10- O'15 ΟΙΟ
0*10- O'15
Sb
Ο ΟΙ
0'04- 0.06
No. I from Norway, analysed by Genth. No. 2 from Dil-
lenburg, by Genth. No. 3, English best selected copper. No.4,
selected copper, but not sold as best, owing to its amount of
antimony. No. 5, English best cake copper.
According to Percy,* the appearance of the fracture does
not always decide the quality of the copper, as impure copper
may have a good appearance; the same copper may have
different fractures according to the manner of its being cast.
The fracture of copper poured into the moulds when very
hot shows a collection of large and small more or less per-
fect crystals; this is not the case if the copper is cooled
previous to casting. These differences can only be observed
in large ingots. Copper ladled at too high a temperature
has also the disadvantage that when poured into pots for the
formation of several pieces or cakes, the under surface of the
pieces will be melted and stick to the upper portions. Too
high a temperature also facilitates the rising of the copper,
which remains liquid for a long time when containing the
smallest amount of sulphur, and gives rise to the formation
on its surface of sub-oxide, which may sink and come more
easily into contact with the sulphide of copper than would be
the case with a copper of lower temperature. The move-
ment from the surface towards the interior decreases as the
copper becomes colder; the materials used for the moulds
(copper or iron) is not likely to influence the appearance of
the fracture.
The colour of the surface depends on the temperature of
* PERCY, Metallurgy, i., 368.
78
COPPER.
the water employed for cooling the copper. When cold water
is used the colour is orange red; warm water gives rise to a
rose-red tint similar to that of Japanese copper, which is cast
in moulds coated with canvas, and standing in warm water.*
Whether copper sheathing is more or less strongly attacked
by sea-water depends partly on its chemical composition, but
chiefly on its physical properties; for instance, on the tem-
perature at which it was cast, its porosity, &c.t
Marchand and Scherert found the specific gravity of the
copper to vary according to whether it was cast under a cover
of fluor spar, borax, glass, or common salt
common salt; whilst, ac-
cording to Percy,|| the average specific gravity was 8.899 for
all pure coppers whichever flux had been employed. The
specific gravity of copper subjected to hydraulic pressure
amounted to 8'931, and that of copper wire to 8'9488. The
porosity of every kind of copper almost precludes a correct
determination of its specific gravity. O'Neill§ states that it
decreases upon hammering the copper, owing to the heat
thereby engendered.
Impurities in copper exert a considerable influence upon
the electric conducting power of copper wire, as shown by
the following observations of Holymann and Matthiessen T
(silver wire cold drawn = 100).
Pure copper wire
HARD DRAWN.
Copper containing sub-oxide of copper
Conducting Observed
Power. Tempera-
ture.
93'08
18.9°C.
•
73°32
19'5
2.50 per cent of phosphorus
7°24
17'5
,,
0'95
23°24
22'I
0'13
67.67
20'0
,
5'40
arsenic.
6.18
16.8
2.80
""
13°14
19'1
trace
""
19
57.80
19'7
3'20
zinc
""
56'98
10'3
*B. u. h. Ztg., 1862, pp. 118, 347.
↑ PERCY, Metallurgy, i., 505.
ERDM., J. f. pr. Ch., 1842, p. 193.
|| PERCY, Metallurgy, i., 288.
B. u. h. Ztg., 1862, p. 304.
¶ PERCY, Metallurgy, i., 288.
REFINED (TOUGH) COPPER.
79
HARD DRAWN.
Copper containing 1'60 per cent of zinc.
trace
وو
""
Conducting Observed
Power. Tempera-
ture.
76.35 15.8
85°05 19°0
I'06
iron.
毋
​""
""
26'95
13'1
,,
0*48
34'56
II.2
""
4'90
tin .
19'47
14'4
2.52
""
32.64
17'1
""
1'33
48.52
16.8
2'45
silver
79°38
19'7
I'22
86'91
20'7
3'50
""
gold.
65*36
18.1
0'31
antimony
>>
0°29
lead
sle
64'5
12'0
Australian Burra-burra, heated
88.86
14'0
Russian Demidoff copper, heated
•
59'34
12.7
Best English copper, heated
81.35
14'2
American copper from the Lake Superior,
heated
92'57
15'0
According to Abel and Field, most commercial copper
contains arsenic and silver, often bismuth, less frequently
tin and antimony, and very rarely lead, except when inten-
tionally added in the process. Riley states that phosphorus
may be found in it; Delarne, Müller, and Abel found
sulphur and gold; and Müller, Genth, and Maximilian,
Duke of Leuchtenberg, found nickel.
As small amounts of arsenic and tin in copper are very
likely to be overlooked in the usual analytical methods,
Abel and Field* have endeavoured to find more exact ana-
lytical methods for the determination. of antimony,
arsenic, bismuth, lead, and iron.
Copper usually loses in compactness, ductility, and
appearance of fracture, by a small admixture of foreign
substances. Lead, antimony, arsenic, and iron are said to
render copper red- and cold-short; zinc, tin, bismuth, carbon,
and sulphur, red-short; copper mica and suboxide of copper,
cold-short. Copper is usually classified into three sorts,
with regard to its purity and its applicability. The first
quality, sometimes called crown copper, is required for the
Quar. Journ. Chem. Soc., xiv., p. 290.
80
COPPER.
manufacture of brass plate and brass wire; the second
quality even may be rolled, but the third sort is only
employed for brass castings and other alloys. The following
observations, which seem partly to contradict each other,
have been made concerning the influence of foreign
substances upon the qualities of copper :—
Suboxide of Copper.-Copper containing suboxide shows
a scaly granular, and sometimes radiating fracture, a brick-
red or dark-red colour and dull lustre ; when forged it shows
isolated fibres, and often cracks at the edges; it fuses more
easily than pure copper, but is somewhat pasty; it expands less,
and at the common temperature it is less hard and ductile.
Karsten states that copper containing only II per cent of
suboxide cracks in the cold, and rolled hot when con-
taining 12 per cent. In "dry" refined copper Dick* found
Impure copper
as much as 17'74 per cent of suboxide.
containing lead, zinc, antimony, and arsenic, &c., is red-, or
cold-short, or both at the same time, and has an inferior
colour, a scaly granular texture, and a dull appearance. If
it contains these substances, together with a little suboxide.
of copper, even 2 per cent, its ductility is increased. It
is still doubtful whether these impurities are contained in
the copper in the metallic or the oxidised state; if oxidised
they are probably combined with the suboxide of copper, as
in copper mica. According to Dick,† the best kinds of copper
in the market contain some suboxide of copper and other
metals in small quantities, thus:-Russian Demidoff copper
contains traces of arsenic, iron, nickel, and antimony;
Australian Burra-Burra copper, traces of iron; copper from
Lake Superior, traces of iron and silver; best English
copper, traces of iron, nickel, and antimony; Spanish
Riotinto copper, about 2 per cent of arsenic, with traces of
lead, iron, and nickel.
Carbon. It is generally supposed that copper free from
suboxide absorbs some carbon in combination when in con-
tact with coal for a longer time, and, according to Karsten,
so small an amount as o‘05 per cent of carbon renders copper
* Phil. Mag., June, 1856, (containing also the method of analysis).
PERCY, Metallurgy, i., 269, 289.
IMPURITIES CONTAINED IN COPPER.
81
red-short. Carboniferous copper is malleable at the common
temperature, but very brittle when heated; it has a coarse
granular fracture, a yellow colour, and a strong but not silky
lustre. When forged it becomes uniformly fibrous, and
shows in its interior many channel-like or kidney-shaped
holes with a light yellow colour; it flows almost like pure
copper, except that it has a brighter surface. A. Dick* takes a
different view; he has not ascertained with certainty the
presence of carbon in over-poled copper, but as he always
found small quantities of other metals, such as lead,
antimony, &c., contained in the best tough copper, as well
as an amount of suboxide of copper up to 3°37 per cent, he
attributes the brittleness of over-refined copper to those
impurities, and presumes that carbon did not enter into
combination with copper, but had only an indirect reaction, by
reducing the suboxide of copper, which is always present;
thus neutralising the disadvantageous influence of the other
impurities, which are perhaps reduced from oxides, and
allowed to combine with the copper, injuring its ductility.
This is further confirmed by the observations which have
shown that the best refined copper produced in copper-works,
if heated with reducing agents, such as hydrogen, carbonic
oxide, carbon, or fused with common salt, loses some of its
weight by the reduction of the suboxide, which is not the
case with copper produced by the galvano-plastic process.
The latter also remains perfectly ductile when kept longer
in contact with coal whilst in a liquid state. Further experi-
ments, most carefully made on account of the difficulty of
determining very small amounts of carbon occurring
together with other impurities, are required before a satis-
factory explanation can be given. Specimens of the kind of
copper experimented on are shown in the Museum of
Practical Geology, London.†
Lead. The presence of o'r to o°3 per cent of this metal
renders copper cold- and red-short, and if containing o'r per
cent it is no longer fit for the manufacture of fine brass
plate and wire; on the other hand, a small amount of lead
* B. u. h. Ztg., 1856, pp. 330, 337, 338. PERCY, Metallurgy, i., 269.
† B. u. h. Ztg., 1861, p. 54.
VOL. II.
G
82
COPPER.
seems to increase its facility for rolling, and is therefore
sometimes added to copper of good quality.
Iron. In small quantity this metal is not so deleterious.
as lead, but larger quantities render the copper hard and
brittle. The best copper may contain o'10 to 0.15 per cent
of iron, and as much as 5 per cent is sometimes found in
ancient copper coin.*
Antimony. A trace of antimony is injurious to copper.
Experiments made in England show that the presence of
one-thousandth only of antimony renders copper unfit for
the production of brass plates and wire, though not for
rolling purposes. Copper containing 30 ounces per ton
appears ragged at the edges when rolled.†
Arsenic. This metal behaves like antimony, but is more
easily extracted.
+
+
Bismuth. According to Levol, one-thousandth of this
metal injures the quality of the copper, chiefly by lessening
its ductility. Field|| discovered bismuth in many sulphu-
retted copper ores, chiefly in those of Cornwall.
Nickel. According to Genth,§ copper containing o'2 or
0.3 per cent of nickel is less adapted for the manufacture of
brass than for German silver. Copper containing protoxide
of nickel is somewhat brittle. V. Leithner has determined
the amount of cobalt and nickel contained in the copper ores
and smelting products of Tergove.
Aluminium.-According to Tissier,** 1 per cent in copper
prevents the oxidation of the fluid copper.
Sulphurtt renders copper red-short; o'6 per cent of zinc
and 0.25 per cent of tin produce the same effect.
Silicon forms combinations with copper which are
more or less brittle and fusible, according to the amount of
PERCY, Metallurgy, 1861, p. 504.
+ Ibid, p. 504.
LEVOL, Répertoire de Chim., appliq., tom. i., p. 256. Bgwkfd., xvii., 641.
Polyt. Centr., 1862, p. 410. DINGLER'S Polyt. Journ., Bd. 163, p. 356.
ERDMANN, J. f. pr. Ch., Bd. 42, p. 189.
§ Bgwkfd., x., 331; xii.. 223;
Oesterr. Ztschr., 1861, p. 133.
** B. u. h. Ztg., 1862, p. 384.
tt ERDMANN, J. f. pr. Ch., x., 237.
B. u. h. Ztg., 1856, p. 338. Polyt. Centr., 1858, p. 1512. PERCY, Metal-
lurgy, i., 282.
++
IMPURITIES CONTAINED IN COPPER.
83
silicon, and are sometimes white. Copper with 1.82 per cent
of silicon has the colour of bronze, and may be rolled and
hammered when cold, but becomes brittle when hot. If con-
taining 52 per cent the alloy is yellowish white, and very
brittle. Upon heating coal, silica, and copper, CuSi is
formed, and it may thus result in the refining process; yet no
trace of silicon is found in copper that has been over-refined.
Phosphorus* increases the fusibility and hardness of
the copper; if containing 1 per cent, copper may still be
rolled at the common but not at a high temperature. A
larger amount of phosphorus renders copper brittle in a
cold state also. The colour of the combination varies
according to the amount of phosphorus, 1 per cent renders
it bronze-coloured, and II per cent steel-grey.
Ruolzt produced alloys of silver, copper, zinc, and nickel
with to 2 per cent of phosphorus: The latter increases
the fusibility, whiteness, and homogeneity, but not the
ductility. If intending to roll such an alloy it must
previously be heated to redness for several hours in a
closed crucible, together with charcoal, thus extracting the
phosphorus.
Sodium has a purifying action upon copper, according
to Tissier. It combines directly with sulphur, phos-
phorus, and arsenic; it produces, with antimony, bismuth,
and other analogous metals, very oxidisable alloys, and
reduces suboxide of copper, the soda oxidising carbon
and silicon. Upon this is perhaps based the behaviour
observed by D'Arcet and Berthier, that fluxes containing
potash have a purifying reaction upon copper in the refining
process.
Copper does not affect the health of the workmen, or only
very slightly.§
Slags and Copper-scale also result from the refining
PERCY, Metallurgy, i., 280.
+ Mining Journ., 1862, p. 164.
B. u. h. Ztg., 1862, p. 118.
|| KERL, Rammelsberg. Hüttenpr., 1861, p. 125.
§ B. u. h. Ztg., 1848, p. 777. BROCKMANN, Metallurg. Krankheiten des
Oberharzes.
G 2
84
COPPER.
process. The slags are usually washed and added to the
refining process or at some other stage of the smelting process.
The copper-scale, consisting of oxidised copper, is either
worked up in copper smelting processes or used for the manu-
facture of blue vitriol.
Thomson* recommends the purification of copper con-
taining iron, antimony, arsenic, &c., by smelting one
hundred parts with ten parts of copper scale and ten parts
of powdered bottle glass. This method was tried at Altenau
for purifying copper mica, but as several smeltings were
required to produce a good copper the process was very
expensive.
Illustrations of the German Method employed for
Smelting Copper Ores free from Lead, Silver, and
Gold.
I. Smelting in Sump Furnaces.
a. Atvidaberg's Copper Works, near Linköping, in Sweden.†
-The orest employed here are classified into soft ores
(Blötmalm), and hard ores (Hardmalm). The soft ores
occur in the Bersbo mine; they consist of a mixture of zinc.
blende, copper pyrites, iron pyrites, magnetic pyrites,
magnetic iron ore, mica, quartz, and a little granite;
occasionally also some calc spar, and, very rarely, some
traces of galena. Zinc blende forms on an average one-third
of the whole mass. Hard ores also occurring in the Bersbo
mine contain more quartz and silicates than sulphides;
these ores are mixtures of copper pyrites, blende, a little
iron pyrites, and much quartz, mica, felspar, ehrenbergite, and
granite. Hard ores occurring in the Mormor, Hagen, and
Malmvik mines, are a mixture of copper pyrites with a little
iron pyrites, and sometimes magnetic pyrites, with a great
deal of quartz or silicates, but almost without zinc blende;
at Hagen, variegated copper ore, and at Malmvik, cobalt ore,
Bgwkfd., ii., 31. DINGLER'S Polyt. Journ., Bd. 73, p. 283.
+ Bgwkfd., xiii., 401. B. u. h. Ztg., 1850, p. 193; 1859, p. 69. PERCY,
Metallurgy, i., 395, 502.
忄
​V. COTTA, Erzlagerstätten, ii., 325. TURLEY, Schweden's Erzlagerstätten, in.
B. u. h. Ztg., 1861, p. 382.
THE RAW SMELTING PROCESS.
85
in a crystallised state are sometimes found disseminated in
copper pyrites. The ore contains on an average 5 per cent
of copper.
The smelting process at Atvidaberg comprises the fol-
lowing operations:-
1. Roasting the Ores.-Soft ores are twice roasted; hard
ores from the Bersbo mine once, on account of the blende
they contain; skumnas (page 32) are also once roasted, but
hard ores from the other mines are not roasted at all. They
are usually roasted in open heaps; a level area of about
30 square feet is covered about 10 inches high with wood;
upon this are put, to about 11 feet in vertical height, alternate
layers of ore and small coal, and the top as well as the sides
are covered with a layer of ore dust about 5 inches thick.
After the heap has burned for about five weeks it is roasted
in a second fire, the large pieces being previously broken to
the size of the fist.
Skumnas are more difficult to fuse, and are roasted in
larger heaps in order better to keep up the temperature. A
stronger wood foundation and more coal are used; these
heaps burn four weeks.
Ι
Mounds 28 feet long, 10 feet high, and 14 feet broad, are
sometimes used for roasting. A foundation of wood I foot high
is laid; upon this 4 feet of soft ores in pieces weighing about
25 lbs. each, then 4 inches of small coal, again about 4 feet
of ore, and so on, till the charged mound contains four layers
of ore and three layers of small coal; the surface and front
are covered 2 inches thick with small ore.
A mound con-
taining 100 tons of ore burns about six weeks, and consumes
about 224 cubic feet of wood and about 200 cubic feet of
small coal. At the second firing the wood foundation is
1 feet thick and the layers of coal 6 inches thick.
second roasting requires about four weeks, and the ore is
considered well roasted if a raw matt containing from 20 to
30 per cent of copper is produced from it by the raw smelting.
In 1859, 85 heaps were roasted at the first firing and 59 heaps
at the second, consuming 1,568 tons of charcoal and 157,000
cubic feet of wood.
This
2. The Raw Smelting.-Ores containing an excess of
86
COPPER.
silica are mixed with basic substances, such as roasted
skumnas, black copper slags, and sometimes also lime, in such
proportions as to yield a bi-silicate slag with about 45 per
cent of silica. In 1859, one charge consisted on an average.
of o⚫6 cwt. of twice roasted soft ores, o‘6 cwt. of once roasted
hard ores, 0*2 cwt. of rich ore, 0'4 cwt. of impure slags, and
about 03 cwt. of black copper slags (total, 2 to 2°1 cwts
of mixture), which is smelted with o'3 cwt. of coke or o'6 cwt.
of charcoal. One part of English coke is as effective as
2 parts of charcoal; 4 parts of ore mixture may be smelted by
means of I part of charcoal, but 5 parts of ore mixture are
smelted with one part of fuel consisting of a mixture of coke
and charcoal.
The smelting furnaces employed are of three different
constructions.
a. Bredberg's furnaces, provided with four tuyeres, repre-
sented by Figs. 7, 8, and 9. a is the sump formed of clay
FIG. 7.
α
Lexa
h
FIG. 8.


α
10
15
20 FT
and coal dust; b, front wall; c, tuyeres; d, foundation, with
channels for the escape of the moisture; e, tapping hole;
ƒ, rough walling; g, space between inside and rough walling;
SMELTING FURNACES.
87
FIG. 9.

h, inside walling.
The furnaces used at present are some-
what different in their dimensions from those given in the
drawings.
B. Carlberg's furnaces, represented by Figs. 10, 11, and
12, are built to economise fuel. a shows refractory talc
FIG. 10.

Oc
α
e
slate; b, bricks; c, iron; d, copper; e, tymp of iron; f, leather;
g, granite. The dotted parts indicate a filling of sand.
y. Furnaces with four tuyeres constructed by Carlberg are
provided with a division wall I foot thick and commencing
10 feet above the bottom of the furnace. They are intended
to facilitate a uniform sinking of the charges. According to
Ahrend, Carlberg's furnaces allow a saving of fuel compared
with Bredberg's furnaces, but Malmquist thinks the re-
verse to be the case. The process is quicker in Bredberg's
PERCY, Metallurgy, i., p. 398.
*
88
COPPER.
FIG. II.

དྲ་བ་
C
α
furnaces, and they yield most favourable results at Kongsberg.*
The dimensions of Bredberg's furnace, with division wall, and
of Carlberg's furnaces are as follows :—
Carlberg's
Furnace.
Bredberg's
Furnace.
Feet. Inch.
Feet. Inch.
Height from the hearth sole to the
furnace mouth
Width of the hearth sole
24 O
18
3
8
3
8
at the level of the tuyeres
4
3
10
of the furnace mouth
6
0
3
6
??
Depth of the hearth sole
5
6
5
4
of the furnace mouth
I
>>
9
I IO
B. u. h. Ztg., 1855, p. 257.
THE RAW SMELTING PROCESS.
89
FIG. 12.

d
C
a
The nozzles have a slight inclination, and are formed of
an iron plate 6 feet long, 8 or 9 inches wide, and 3 of an inch.
thick. The copper tuyeres of in. diameter are connected
with the blast tube by a leather pipe. The smelting process is
carried on with a pressure of blast of o°8 inch of mercury,
and a nose from 4 to 6 inches long. If the nose gets dark,
holes are pierced in it; if it grows too much, an inter-
mediate charge of fuel is given. A furnace with three tuyeres
smelts in 24 hours from 80 to 90 charges, and a furnace with
four tuyeres, at first from 90 to 100 charges, and later from
100 to IIO. About 60 per cent of slag, containing or per
cent of copper, is made to run into a sand bed at the side
of the fore-hearth and is thrown away. The slag taken out of
90
COPPER.
the hearth, and containing from to 1 per cent of copper,
is returned to the smelting. As soon as the hearth is filled
with matt containing from 20 to 30 per cent of copper, which
will take place every 48 to 72 hours, it is tapped off into
sand moulds, and broken into pieces of the size of the fist,
after cooling for two days. The hearth and fore-hearth are
then cleared, and the tapping hole is closed with a wooden
plug. From about 2 to 4 tons of raw matt result, amounting
to from 17 to 20 per cent of the weight of the charge. The
character of the slag facilitates the separation of the matt.
When the process has lasted two months, the zinciferous
deposits are removed every month, for which purpose the
furnace charge is allowed to sink 6 or 7 feet, and the blast is
interrupted; the smelting operations last 8 or 10 months.
The following products are obtained :-
Raw Matt, containing from 25 to 30 per cent of copper;
it is roasted.
Raw Slag, bi-silicate, grey or yellowish grey, vitreous,
containing on an average per cent of copper.
114
Ferriferous Bears, which may be easily divided, especially
if they are exposed for a longer time to the air. They con-
tain much sulphide of iron and zinc and very little sulphur,
and from 6 to 12 per cent of copper. They are added to the
raw smeltings frequently in pieces weighing several cwts.
Furnace Ends.-If coarse, they are added direct to the
raw smelting; if finer, they are previously washed.
Zinciferous Deposits are thrown away.
In 1859 were smelted about 7,463 tons of twice roasted
soft ores, 8,824 tons of hard ores, 2,201 tons of rich
ores, 5,997 tons of impure slags, and 3,769 tons of black
copper slags, giving a total of 28,255 tons, at a consump-
tion of about 3,959 tons of coke and 1,611 tons of charcoal,
and producing 5,954 tons of raw matt, or in 24 hours 69 cwts.,
equal to 24°32 per cent of the charge. o665 cwt. of coke
and 0'270 cwt. of charcoal were consumed for the production
of I cwt. of raw matt, and I cwt. of ore mixture required for
smelting o*140 cwt. of coke and o'056 cwt. of charcoal.
One furnace is worked by two smelters and two assistant
THE BLACK COPPER SMELTING PROCESS.
91
workmen, and they are paid about 4d. and 3d. respectively for
every 10 cwts. of produced raw matt.
3. Roasting the Raw Matt.-This is effected in mounds
II feet long, 4 or 5 feet broad, 5 feet high, and with a capacity
of 5 tons. At the first roasting a foundation of wood is
laid 8 inches high, and the matt charged upon it without any
addition of fuel. At the second roasting 15 cubic feet of
coal are added to the wood foundation; at the third roasting
50 cubic feet of coal; at the fourth 80 cubic feet; at the fifth
100 cubic feet are given in addition to the wood: at the sixth
roasting 100 cubic feet are spread out upon the wood founda-
tion, and 20 cubic feet in two layers between the matt. The
whole roasting is finished in seven or eight weeks by twelve
workmen, receiving altogether Is. Id. for every 380 lbs. of
refined copper.
In 1859, 6,020 tons of raw matt were
roasted at a consumption of 1,612 tons of charcoal and
182,000 cubic feet of wood.
4. Black Copper Smelting Process.-The roasted raw
matt being rich in bases requires siliceous fluxes in sufficient
quantity, chiefly 10 to 20 per cent of pure ore slag, in order
to form a mono-silicate slag. One charge consists, for
instance, of 2 cwts. of roasted raw matt, o°2 to 0'4 of
cupriferous residues, o'2 to 0'4 of ore slags, o'r cwt. of
quartz, and o'6 cwt. of charcoal.
Bredberg's furnace, represented by Figs. 13 and 14, is
frequently used for smelting the roasted matt. a, is the
FIG. 13.
FIG. 14.


92
COPPER.
hearth; b, fore wall; c, tuyeres; d, channels in the foundation.
Carlberg's furnaces are also used, and in 1859, a furnace
with three tuyeres and a division wall was erected; it con-
sumes less fuel by 19 per cent, and the production is increased
by 23 per cent, while the gases are caught 8 feet above the
hearth and conducted to the boilers for heating purposes; the
reduction of oxide of iron is thus diminished.
The dimensions of the furnace are as follows :—
Height from sole to furnace mouth
>>
ft. in.
ft.
in.
17
3
division wall .
the tuyeres
6
6
O
4
Dimensions of the hearth sole.
""
at the level of the tuyeres
6 feet above the tuyeres
at the furnace mouth.
Thickness of the division wall.
•
2
2
2 IO
~ ~ ~ +2
ХХ
222
6
4
6 x
6
6 x I
6
O
6
The sole stone is coated with a layer of gestübbe (a mixture
of small coal and loam) and loam; the breast is walled with
brick-work up to 6 inches above the wall stone, and the
remaining opening is almost closed with sand.
The smelting is done with a nose 4 or 6 inches long, and
the slag is tapped off by means of a channel formed in the
sand, when the smelting mass reaches nearly to the tuyeres.
The slag runs off into a gutter provided with trough-like
hollows formed in the sand. Matt and slag are tapped off
later together, to allow the pure black copper to run out.
The matt collects chiefly in the first two hollows, and
flows out of the solidified but still red-hot slag pieces when
turned round and broken. The black copper is tapped off
every two or three days, and is made to run into a gutter
36 feet long and provided with nine division walls, each lying
2 inches higher than the following one. The copper is kept
covered with coal that it may run into all the divisions; in
order to divide the single cubes, the division walls are broken.
out while the copper is still red hot. Each operation in the
furnaces lasts from six to eight months.
The smelting products are--
Black Copper, about 20 or 30 per cent of the weight of
THE BLACK COPPER SMELTING PROCESS.
93
the mixture; it is refined. An analysis made at the Mining
School at Fahlun, showed this black copper contained-
Cu
Fe
Zn
Co and Ni.
Ni.
Sn
Pb
94'39
2.04
I'55
0'63
0'07
0'19
ΟΙΙ
0.80
Ag
S
Matt (Dünnstein), with from 55 to 72 per cent of copper.
When containing a lower amount of copper it is reddish blue,
and steel grey when containing a higher; it is added in the
third firing of the raw matt roasting.
Slag, bisilicate, added to the ore smelting.
Ferriferous Bears.-They are red in colour, being chiefly
iron with a large amount of copper, and are very difficult to
divide. They are often worked in large pieces in a hearth
similar to those used for refining copper for the production
of black copper. The hearth is furnished with two tuyeres,
1 inch wide on the mouth, lying 1 foot from each other,
and having an inclination of 26°; they reach 8 inches
into the hearth, and lie 6 inches above the hearth on the
back wall. From 90 to 100 cwts. are worked in 60 hours, at
a consumption of 514 cubic feet of coal. The resulting
black copper is refined, the formed matt roasted together
with raw matt, and the slags are added to the ore smelting
operation.
In 1859 were worked nearly 6,020 tons of roasted raw matt,
730 tons of cupriferous residues of other processes, and
733 tons of ore slags; total, 7,483 tons, at a consumption of
2,546 tons of charcoal; the result was 1,344 tons of black
copper, or in 24 hours 34°47 cwts., amounting to 19°19 per
cent of the roasted raw matt and residues. I cwt. of black
copper consumed 189 cwt. of coal, and I cwt. of mixture
was melted at a consumption of 0.32 cwt. of coal.
I
Each furnace is attended to by two smelters and two assistant
workmen in shifts of eight hours, each of them receiving
4d. and 3d. respectively for every 10 cwts of refined copper
produced in the works.
94
COPPER.
5. Refining the Black Copper.-The purity of the black
copper allows it to be refined in larger quantities in large
hearths; they are from 2 to 2 feet in diameter and from 15
to 18 inches deep, and constructed of sand and clay; the
copper tuyere is 1 inch wide, with an inclination of 45°,
and reaches 4 inches into the hearth. The blast has a
pressure of 30 or 40 lines of mercury.
On beginning the process, the hearth is filled with coal;
pieces of black copper are put on the sides of the hearth
and upon these three blocks of copper, one above the other.
They are then covered with coal and smelted. The ferrugi-
nous slag, which is blue in colour at the beginning, is several
times removed whilst adding quartz, and as soon as a sample
taken by means of the proof rod of iron shows a still some-
what impure copper of a brass-yellow colour, it is ladled out
into moulds by a ladle which is suspended by a chain, whilst
the copper in the hearth is kept covered with small coal.
About 28 cwts. of black copper are refined in 4 or 4 hours,
causing a loss of about 10 per cent, and a yield of about 25
cwts. of refined copper.
The results are-
Refined Copper of good quality, but inferior to the Russian
copper. An analysis made at the Mining School at Fahlun
gave the following result :-
Cu
Fe
99*460
Ο ΟΙΙ
Co and Ni
Sn and Pb
•
Ag
Au
S.
ΟΙΙΟ
trace
0*065
0'0015
O'017
Hearth Ends, which are returned to the smeltings.
In 1859, 26885 2 cwts. of black copper were refined, pro-
ducing 22984 27 cwts. of refined copper, and consuming 11088
cwts. of coal; I cwt of refined copper being produced at a
consumption of 0.48 cwt. of coal.
On an average 4'64 per cent of refined copper results from
the ores and slags, 5'82 per cent from the ore, 19'09 per cent
from the raw matt, and 85°5 per cent from the black copper.
THE FAHLUN SMELTING PROCESS.
95
The loss of copper amounts to about a quarter per cent of
the weight of the raw materials.
In 1861 an English reverberatory furnace, provided with
charcoal gas firing, was erected for refining copper.
The workmen receive a little more wages than the smelters
of black copper.
The smelting expenses for 4 cwts. refined copper are as
as follows:
Materials to produce 4 cwts. of Refined Copper.
For ore mining and transport.
roasting
residues, slags, re-melting, &c.,
""
"
matt, transport, division, &c.,
black copper, melting, refining, matt
roasting.
coke.
charcoal for smelting and refining.
wood for ore roasting
£ s.
25*36
0'94
2.79
-6.25
3.73
I 8.60
I 4'75
I'21
,,
wood for matt roasting
I'44
£5 15°11
The total cost, inclusive of superintendence, &c., amounts
to £9 9*72s. per 4 cwts.
b. The Fahlun Smelting Process.*—This process differs little
from that adopted at Atvidaberg, and similar ores are em-
ployed. Soft ores containing 1 or 2 per cent of copper are
richer in iron pyrites, poorer in zinc blende, and oftener con-
tain galena, which is then carefully separated; besides iron
and copper pyrites, chiefly quartz, talc, slate, and other
silicates, the hard ores with 3 or 4 per cent of copper some-
times contain magnetic iron ore and calcspar, and rarely zinc
blende and galena.
Hard ores are not roasted, and soft ores are roasted in
mounds for three or four weeks, in quantities of about 60 tons,
at a consumption of about 175 cubic feet of wood.
The smelting is effected in furnaces with three tuyeres,
ERDM., J. f. ök. u. techn. Chem., iii., 285; iv., 310; xii., 207, 318. Ann.
d. min., 3 sér., xvi., 643.
96
COPPER.
constructed like those used in Atvidaberg; the ore mixture is
composed of one part of hard ore, two parts roasted soft ore,
from 10 to 30 per cent of black copper slag, with some raw
pure copper pyrites which is spread over the mixture. Every
two days about 3 tons 5 cwts. of raw matt containing from
10 to 16 per cent of copper, are tapped off. This matt is
roasted four or five times, in quantities of about 7 tons, con-
suming about 95 cubic feet of wood and 220 cubic feet of coal.
The resulting bisilicate slag is generally composed as follows:-
Sio,
FeO
44
per cent
46
4
Al2O3
MgO
Cao.
I'2-7
3 -4
The mixture being more easily fusible than that of Atvida-
berg, allows a saving of fuel in the smelting process; the
smelting operations last eight or ten weeks.
The roasted raw matt is mixed with quartz or sula slag, and
smelted in furnaces similar to those used in Atvidaberg.
The matt and black copper are tapped off and run into a
cast-iron mould 8 feet long, I foot high, 12 inches wide at
the bottom, and 15 inches wide at the top; the matt and slag
which stick to it are removed from the surface of the copper,
and the copper is divided into three parts by pressing into it
an iron cutter.
In 1844, 85 tons 5 cwts. of black copper were produced
from 755 tons of raw matt. The consumption of coal per
374 lbs. of produced black copper amounted-
At the ore roasting to
""
ore smelting.
matt roasting
I'90 cubic feet of coal
669*04
57.82
black copper smelting 154'45
""
""
""
The black copper slags are returned to the raw melting;
the black copper containing from 70 to 90 per cent of copper is,
at Avesta, refined in similar hearths to those represented by
Figs. 15 and 16. The space, a, formed by the side walls, b,
is filled with beaten down brasque in which the hearth, c, is
cut out; d are flues for the escape of moisture; e is an iron
plate. The massive wall, f, contains the tuyere, h, and the
THE FAHLUN SMELTING PROCESS.
97
vault, g, leading to the tuyere. A chimney piece is fixed
some distance above the hearth, c.
Ъ
FIG. 15.
У
€
e
FIG. 16.


C
E
Z
onc
F:
When using the hearth for toughening rosette copper the
tuyere is placed somewhat higher and less inclined.
Dr. Molin has of late considerably modified the Fahlun
process, and remodelled it on the basis of that adopted in
Oker, in order to extract the silver and gold contained in the
ores. The latter are roasted once and smelted for the pro-
duction of raw matt, which then is roasted and smelted for
the production of auriferous and argentiferous black copper
and copper matt or regulus. This regulus upon roasting
"dead" yields an excellent black copper, well fit for the
manufacture of brass. The black copper containing o‘1 per
cent of silver and gold is treated with dilute sulphuric acid,
and the resulting copper vitriol is mixed with coal and added
to the roasting of copper matt. The residue resulted from
the treatment with sulphuric acid containing gold, silver,
copper, and lead is roasted and then treated first with dilute
sulphuric acid and hereafter with concentrated acid at higher
temperature, thus extracting silver and copper, which form
sulphates. The silver contained in the sulphate may be pre-
cipitated either by copper or common salt. The auriferous
residue of the last treatment with concentrated acid is treated
with aqua regia, and the insoluble parts returned to the
black copper melting; the gold contained in the solution is
precipitated by iron vitriol.
The processes at Garpenberg and Näfvequarn* in Sweden
are similar to the Fahlun process.
WINKLER, Erfahrungen über die Bildung der Schlacken, 1827, p. 40.
VOL. II.
H
98
COPPER.
In 1857, the production of copper* in Sweden amounted to
about 1,703 tons.
c. Smelting Process at Röraas in Norway.t-Copper pyrites‡
associated with iron pyrites and a little zinc blende occurs in
chlorite slates. It contains on an average 5 per cent of
copper, and is treated by the following operations :-
1. Ore Roasting. The roasting is effected in heaps con-
taining about 240 tons of ore; the ore schlich is charged at
the top, and the sulphur collected in the same manner as in
the Lower Hartz. 300 cwts. of ore consume about 100 cubic
feet of wood, and cost 6s. 6d. for wages. The duration of
the roasting varies between seven and twelve weeks.
2. Ore Smelting.-The ore is mixed with copper slags in
FIG. 17.
FIG. 18.

8
12
16 FT
* SCHWEDEN'S Metall-production: B. u. h. Ztg., 1858, p. 8.
+ RUSSEGGER'S Reisen, iv., 571. DUCHANOY, in Ann. d. min., 2 livr., de
1854, p. 214. (Oesterr. Ztschr., 1856, p. 212). HERTER, in B. u. h. Ztg., 1855,
p. 256. PERCY, Metallurgy, i., 411. EGGERTZ, in Jernkontoretz Ann. for
1849, p. 275.
‡ DUROCHER, in B. u. h. Ztg., 1855, p. 29. DUCHANOY, in Oesterr. Ztschr.,
1856, p. 212. HERTER, in Bgwkfd., xix., No. 9. COTTA'S Erzlagerstätten,
ii., 522.
SMELTING PROCESS AT RÖRAAS.
99
such proportions as to form mono-silicate slags at the sub-
sequent smelting, which is effected in sump furnaces, as
represented by Figs. 17 and 18; crucible furnaces are also
sometimes employed. About 100 cwts. of mixture are smelted
in 24 hours; the matt is tapped off twice during this time and
made to run into sand moulds. 320 cwts. of roasted ore
yield about 100 cwts. of raw matt, and consume on an average
977'5 cubic feet of coal, costing £I IS. for wages. The
smelting operations last six days.
The resulting products are—
Raw Matt, with 16 to 18 per cent of copper. According
to Eggertz, it is composed of
Cu.
Fe.
S
Residue
22.03
52°14
25°15
3'00
This matt is roasted in lots of about 450 cwts. in mounds
with six to seven firings, so that 300 cwts. of matt consume
888 cubic feet of wood, costing £1 14s. for wages.
Slags, composed, according to Eggertz and Stalsberg, as
follows:-
SiO 3
A1203
CaO
MgO
FeO
Cu
I.
II.
28.48
29.17
9'58
8.89
1*18
2'95
II°55
5'06
50°99
52.78
0'38
trace
3. Black Copper Smelting.-The roasted raw matt mixed
with rather more than 10 per cent of slags, is smelted in
crucible furnaces having a section at the level of the tuyere
of 2 feet by 3. The black copper produced contains scarcely
90 per cent of copper, and the formation of matt is avoided
as much as possible; but this causes a scorification of more
copper and the formation of iron deposits. According to
Eggertz, the resulting slag is composed of―
SiO3
Al₂O
FeO
CaO
3
Cu
30.85
4'00
66*25
0'47
0.63
H 2
100
COPPER.
The slag is partly thrown away or, when containing.
globules of copper, is added to the raw smelting. 300 cwts. of
raw matt, including mixture, require about 1,580 cubic feet
of coal, and cost £2 2s. for labour.
4. Refining the Black Copper.-Six or seven cwts. are
charged, and the refining is carried to a high degree, so that
thick cakes result, which are sold under the name of Dront-
heim copper.
To produce 320 cwts. of black copper 2,418 cubic feet of
coal are consumed, and £4 3s. expended for labour; the
yield of copper amounts to from 3'5 to 6'4 per cent.
For instance, I cwt. of ore yields—
31 lbs. of matt,
6'25
5.61
""
black copper, and
refined copper,
costing 10d. for wood, coal, and labour.
Duchanoy calculates the expenses of the production of
100 kilos. at from 123 to 175 francs. About 106 cubic feet of coal
are consumed in the production of 1 cwt. of refined copper,
and the annual production amounts to from 5,800 to 6,500 cwts.
d. At Szaszka* in Banat, a mixture of sulphuretted and
ochreous copper ores, containing on an average 19 per cent
of copper, is smelted with an addition of iron pyrites, black
copper slags, and calcareous mica and poor copper ores with
0*75 per cent of copper; charcoal is used as fuel. At the
raw smelting about 18 per cent of raw matt, containing from
8 to 14 per cent of copper, results, which is then roasted in
open heaps in quantities of 300 cwts., and smelted in low
furnaces with an addition of quartz, raw slag, and furnace
ends for the production of concentrated matt containing from
25 to 40 per cent of copper. After being roasted, the concen-
trated matt is again smelted in cupola furnaces, yielding
black copper and some matt; the black copper is refined in
small hearths.
The cupriferous iron deposits resulting from the concen-
tration smelting are removed from the furnace whilst still
hot, and roasted four times in open heaps in crushed
quantities of from 800 to 1,200 cwts; from 12 to 20 per cent
* Ann. d. min., livr. 6, de 1846, p. 55. COTTA'S Erzlagerstätten, ii., 287.
SMELTING PROCESS AT SCHMÖLLNITZ.
ΙΟΙ
of iron pyrites is added at each roasting. When smelting
the roasted iron deposits, a matt (klosslech) results, which is
roasted and added at the concentrating smelting.
Somewhat different are the processes employed at Csik-
lowa, Dognarka, and Moldawa, in the Banat* mountains,
and at the Buckowina.t
e. Rohhütte, at Altgebirg in Hungary. The ores consist of
copper pyrites with a little malachite, occurring in Greywacke,
and are mixed with limestone up to 80 per cent, yielding a
mixture difficult to fuse; their composition is as follows:-
Sio,
Cu
Fe
Sb, As
S
50*64
13'00
12.12
0*76
A1,03
CaO
MgO
16.19
2·67
178
0'63
This mixture allows only the small production of 1 ton in
12 hours, 5 tons of the mixture consuming about 1,170 cubic
feet of coal. The limestone employed contains about 40 per
cent of carbonate of magnesia; if pure limestone is used
instead, a very pasty slag, consisting of bi- and tri-silicates,
will be formed. The usual raw slags do not contain more
than 8 or 10 per cent of protoxide of iron, besides a con-
siderable amount of silica, lime, and magnesia. An addition
of a large quantity of mono-silicate slags of peroxide of iron
allows the limestone to be reduced to 20 per cent, and pro-
duces from 30 to 35 cwts. in 24 hours, at a consumption of
less than 170 per cent of coal.
Black copper, matt, and speiss are sent to the copper works
at Tajova.||
f. Process at Schmöllnitz in Upper Hungary.—About
50,000 cwts. of copper pyrites§ occurring in clay slate (gelf,
*
Jahrb. d. k. k. Geolog. Reichsanstalt, 1853, No. 4, p. SOS. Oesterr.
Ztschr., 1862, p. 277.
† Jahrb., &c., 1854, No. 1, p. 222.
+
‡ Oesterr. Ztschr., 1861, p. 132. V. COTTA's Erzlagerstätten, ii., 303. V.
COTTA's Gangstudien, iv., I.
|| Oesterr. Ztschr., 1861, p. 121.
§ V. COTTA's Erzlagerstätten, ii., 306. V. COTTA's Gangstudien, iv., 1.
102
COPPER.
or schiefer-erze), containing about 4 per cent of copper, are
worked annually. About 7,500 cwts. of copper ores, con-
taining about 8 per cent of copper, are bought from private
copper works, and about 11,000 cwts. of cement copper
schlich, containing about 60 per cent of copper. Ores from
the Royal mines containing less than 2 per cent of copper,
and ores from private mines containing less than 4 per cent,
are not accepted.
The rich ores are smelted in a raw state, but the poor ores
are previously roasted in lots of from 2,000 to 4,000 cwts.,
with a wood foundation of 0.25 cubic feet of wood per cwt.,
and an addition of o‘04 cubic feet of coal per cwt.
About 80 per cent of roasted ores and schlich, with about
4 per cent of copper, and 20 per cent of raw ores, containing
about 9 per cent of copper, are mixed with about 16 per cent
of raw slags and 3 per cent of quartz. About 135 cwts. of
this mixture are smelted in 12 hours, yielding about 37 per
cent of raw matt containing 20 per cent of copper, at a con-
sumption of 29 per cent of fuel.
The matt is roasted with four or five fires, which occupies
about four weeks, at a consumption of o˚46 cubic foot of coal
and 0.64 cubic foot of wood per cwt.
The smelting furnaces have the following dimensions:-
total height, 20 feet; height from the furnace mouth to the
tuyeres, 14 feet; thence to the bottom, 6 feet. The round
furnace shaft is 48 inches in diameter from the bottom to the
tuyeres, and is contracted gradually up to the furnace mouth,
where it measures 30 inches. The tuyeres have an incli-
nation of about 8° when melting roasted ores, and of 4° when
melting raw ores.
100 per cent of the raw matt is first roasted and then
smelted in admixture with about 20 per cent of quartz, pro-
ducing about 50 per cent of concentrated matt with 30 or
40 per cent of copper, and consuming about 30 per cent of
coal. The concentrated matt requires nine or ten firings for
the roasting, which is effected in eight weeks; I cwt. of
roasting mass consumes about 2.75 cubic feet of wood and
2.10 cubic feet of coal.
SMELTING PROCESS AT AGORDO.
103
This smelting is performed in cupola furnaces, similar to
those employed at the raw smelting, their tuyeres having
an inclination of 10°.
The mixture for the black copper smelting is composed as
follows:
70 to 75 per cent of concentrated matt, containing 30 to 40
per cent of copper.
9 to 10 per cent of raw matt, from the copper smelting,
containing 60 to 65 per cent of copper.
12 to 16 per cent of cement schlich, containing 59 to 65 per
cent of copper.
0'5 to 1 per cent of residues, containing 17 to 20 per cent of
copper.
15 to 18 per cent of quartz.
About 130 cwts. are smelted in 12 hours, producing from
30 to 35 per cent of black copper, containing from 93.5 to 95
per cent of copper; and from 10 to 13 per cent of matt, with
from 60 to 65 per cent of copper, at a consumption of from
33 to 35 per cent of coal.
g. Smelting Process at Agordo in the Venetian Alps.*-The
ores, whicht consist chiefly of iron pyrites more or less mixed
with copper pyrites, arsenical pyrites, zinc blende, and galena,
contain on an average 2 per cent of copper, also some
cobalt and tin, and about 2.5 per cent of antimony and arsenic.
According to Lürzer, the ore contains on an average 2 per
cent of copper, 43 per cent of iron, 50 per cent of sulphur,
and 5 per cent of earths; according to Rivot, 160 of copper,
42'00 of zinc and iron, 50 of sulphur, 1'50 of arsenic, and
I of silica. The ores are divided into poor ores with less
than 2 per cent, good ores containing from 2 to 4 per cent, and
rich ores with from 4 to 8 per cent or more of copper. They
are worked in the proportion of 5163: 46'16: 2:22. The first
two sorts of ore are submitted to a kernel roasting in open
heaps, and sometimes in mounds, when the sulphur is col-
lected. The crust is separated from the kernels and lixiviated,
KRAUS'S Jahrb., 1848, p. 40; 1852, pp. 223, 231. B. u. h. Ztg., 1863, p. 440.
LURZER, in TUNNER'S Jahrb., 1853, p. 340; 1854, p. 242. HATON, in Ann.
d. min., 1855, viii., 407. RIVOT, Metallurg. de cuivre, 1859, p. 332. B. u. h.
Ztg., 1859, p. 385. PERCY, Metallurgy, i., 439. Oesterr. Ztschr., 1856, p. 338;
1860, pp. 328, 336, 350, 356, 389; 1861, p. 324. B. u. h. Ztg., 1861, p. 223.
† COTTA, Erzlagerstätten, ii., 334. Oesterr. Ztschr., 1860, No. 20.
104
COPPER.
the cement copper precipitated and smelted together with
the rich ores and kernels for the production of matt.
Roasting in mounds gives a better yield of sulphur, and
economises fuel and time. The mounds employed are repre-
sented by Figs. 19, 20, and 21. The bottom of the mound
FIG. 19.

>
•
100 50 0
3
•
L
•
ㅁ
​do
2
d
12 MET
is formed of a number of inclined planes, b, b. These planes
in the newer mounds are provided with a number of gutters,
FIG. 20.

f
J
J
£
12 MET
100 50 0
UM DA
9
which communicate at their deepest points with the chief
gutter, and serve to conduct the fused sulphur. d, d are
FIG. 21.

e
ooo gro
>
O
O
n
O O
men
channels for running the sulphur into the basins, e, c. The
channels, c, c, admit air into the mound; the apertures, g, g,
emit the sulphur vapour into the chambers, f, f.
Both roasting processes yield about 14 per cent of kernels,
SMELTING PROCESS AT AGORDO.
105
with about 45 per cent of copper and 76 per cent of crusts,
containing on an average 1325 per cent of copper; the
yield of sulphur amounts to o'2 and 1°4 per cent respectively,
and the expense for roasting one ton of ore to 3'914d. and
4*398d. respectively. The greater expense of roasting in
mounds is caused by the outlay for apparatus, and for re-
building the front wall after charging the mounds, but this is
balanced by the increased yield of sulphur. The sulphur is
refined by a re-melting in iron boilers. The oxidised crust is
separated from the kernel by a hammer in such manner that
it contains not more than 7 or 8 per cent of copper, while
the kernel sometimes contains from 20 to 40 per cent of
copper.
The expense of separating is Is. 4'15d. per ton. According
to Forbes, the silver concentrates in the crust, and causes
loss.
The mixture for raw smelting is composed as follows:-
0*120 tons of rich ores, containing 6·053 per cent of copper.
0'753
O'127
0*065
0°200
O'190
")
kernels
4'50
cement copper,, 40'26
smoke and furnace ends.
rich slags.
red sandstone.
""
وو
""
I'455 tons.
The cement schlich is dressed on percussion tables, so as to
form three sorts, containing respectively from 60 to 70,
from 7 to 9, and from 2 to 4 per cent of copper, and to
extract as much as possible of the arseniate of iron present.
Higher and lower cupola furnaces are used for smelting
the ores; the lower furnace being 43 metres high. The
smelting is effected with a nose about o'17 metres long; the
slags are made to run into an outside basin, and after their
removal from this basin the matt is run into it and
formed into cakes. If, owing to the presence of sulphide
of zinc, the matt is very compact, it is made to run in thin
layers on the floor of the smelting house. The pure slag,
containing not more than 04 per cent of copper, is thrown
PERCY, Metallurgy, i., 447.
106
COPPER.
away, and impure slags are added to the black copper smelting.
I ton of ore yields about 0'073 ton of matt, containing about
25 per cent of copper and from 26 to 28 per cent of sulphur.
17 tons of ore mixture are smelted in 24 hours, while the
smelting operations last about 20 days. I ton of rich ore,
kernels and cement copper, consumes 0'313 tons of charcoal,
at a cost of 17s. 3d.
For raw matt roasting, the crushed matt is roasted in
mounds five times; in the last firings the roasting mass is
provided with a layer of fuel. I ton of matt consumes o'085
ton of coal, o'019 ton of turf, and o'195 ton of wood, costing
IOS. 6d.
Black Copper Smelting.-A smelting mixture con-
sists of:-
Raw matt
Concentrated matt
Smoke and residues
Rich slags
Ore slags.
Sandstone
I'000 tons
0'230
0'045
0*065
0°225
0*220
1'785
Half cupola furnaces are used for smelting black copper,
thus preventing a reduction of iron and the formation of a
ferruginous black copper, which can only with difficulty
be refined; on the other hand, the consumption of fuel is
larger; the operations last ten or twelve days, and 9 or 10
tons of mixture are smelted in 24 hours, consuming about
4 tons of charcoal and producing about 25 per cent of black
copper matt (Dünnstein), and 26 per cent of black copper,
calculated upon the raw matt charged. The slag containing
about 40 per cent of silica is added to the matt smelting
process; the black copper containing about 93 per cent of
copper is refined; the black copper matt, containing about.
63 per cent of copper, is three times roasted in mounds. I ton
of roasted raw matt consumes o‘695 ton of coal, and costs
£1 18s. for smelting. o'51 part of coal is consumed in the
production of I part of matt, and 0.284 part in smelting one
part of mixture.
The first refining (roasting of the black copper) is effected
SMELTING IN CHANNEL FURNACES.
107
in hearths consisting of 3 parts sandstone and I char-
coal. One hearth is capable of holding about 8 cwts. of black
copper, which are refined in 12 hours; at the commencement
the inclination of the tuyere is about 14°, and is increased
later on.
100 parts of black copper yield 78 parts of refined
copper, and about 55 parts of hearth ends and rich slags,
which are given to the matt smelting. I ton of black copper
consumes o*632 ton of coal, and costs £1 15s. for labour.
The second refining (toughening of the refined copper) is
performed with charges of about 8 cwts., yielding 94°7 per
cent of tough copper. The hearth ends are added to the
black copper smelting. I ton of tough copper consumes
0˚516 ton of coal, and the expenses are £1 9s. 3d.
The total cost of I ton of rosette copper amounts to
£99 15s. Other reports state the smelting cost of copper,
sulphur, and vitriol per cwt. to be, respectively, £2 10s. 7d.,
9'7d., and 2*3d., while the market prices are, respectively,
£5 14s. 8d., 12s. 4'5d., and 2s. 9d. The average annual
production during the last 30 years is 3,681 cwts. of copper,
767 cwts. of sulphur, and 11,163 cwts. of iron vitriol.
The loss of copper amounts to 7 or 8, or at the highest
10, per cent; this result may be considered most satisfactory,
as the ores contain on an average only 1 per cent. At
Riotinto the loss amounts to 50 per cent, whilst the ores
worked there contain 3 per cent.*
Smelting in Channel Furnaces with Two Open Eyes.
a. Process at Sterne and St. Josephsberg, near Linz, on the
Rhine.t-The following orest are worked at the Sterne
copper works:-Poor oxidised copper ores containing to
3 per cent, and poor sulphuretted ores containing up to 12
per cent, and similar ores containing upwards of 12 per cent,
also copper matt from different lead works on the Rhine.
The roasting of the ores is effected in double reverberatory
* B.u.h. Ztg., 1862, p. 301.
† FETIS, in Revue Univers., iv., 158, 433.
BERGGEIST, 1858, No. 45.
B. u. h. Ztg., 1859, pp. 107, 223, 438; 1860, pp. 27, 191.
+
‡ COTTA, Erzlagerstätten, ii., 148.
108
COPPER.
furnaces. The very pure and well-roasted ores are smelted
in channel furnaces 4 feet high above the tuyere, together
with iron refinery cinders and limestone, sometimes in the
proportion of 30: 8: 8; now and then 20 per cent only of
limestone is added with a variable quantity of Dünnstein,
furnace ends, and (formerly) cement copper. About 30 cwts.
of roasted ore are smelted in 24 hours, producing 35˚5 per cent
of black copper, and consuming 16 or 17 cwts, or on an
average 82.09 per cent, of coke.
The refining is effected in charges of 2 cwts. of black
copper, yielding 70 per cent of refined copper.
b. Smelting Copper Pyrites at the Upper Hartz.*-The
copper pyrites associated with quartz, calcspar, iron pyrites,
some blende, and a little galena, contains on an average
18 per cent of copper.
The pyrites in small pieces is roasted in sheltered heaps,
containing about 11 tons, for at twelve weeks, at a consump-
tion of about 250 cubic feet of wood. Pyrites in the form.
of schlich is roasted in quantities of 4 tons for about nine
days, after being previously mixed with lime and pressed into
moulds.
For the purpose of smelting, the roasted ore is mixed
with about an equal quantity of slags resulting from the
matt smelting, and the smelting process is carried on in
channel furnaces, 11 feet high and I foot 5 inches by 3 feet
6 inches wide at the level of the tuyere, and 2 feet 7 inches
at the furnace mouth.
72 cwts. of mixture are smelted in 12 hours, and 100 cwts.
of pyrites, containing 18 per cent of copper, produce 53 cwts.
of raw matt containing about 36 per cent of copper, and
slags containing from to 1 per cent, consuming 47'5 cwts.
of coke and 108 lbs. of charcoal; the process is carried on
for 24 days without interruption.
Owing to the great amount of antimony contained in the
ores the matt requires to be smelted five times, each time
after being previously roasted, before its final conversion
into black copper. At the commencement the matt is
KERL, Oberharzer. Hüttenprocesse, 1860, pp. 311, 636.
SMELTING PROCESS AT KUPFERBERG.
109
roasted six or eight times, and at last from nine to twelve
times. The matt smelting is also effected in channel
furnaces with two open eyes, and about 40 cwts. of matt
mixed with about 36 cwts. of ore slags, are smelted in 24
hours. 100 cwts. of raw matt yield 33 cwts. of black
copper containing from 94 to 98 per cent of copper, and
I cwt. of matt containing about 75 per cent of copper,
besides slags with 1 or 2 per cent of copper.
The black copper is purified in charges of 3 or 4 cwts. in
about 3 hours, yielding about 90 per cent of rosette copper,
and consuming 10 cubic feet of charcoal to I cwt. of rosette.
copper. The refinery slags, with an addition of 120 cwts.
of matt slags, are melted for the production of black
copper, which is roasted in reverberatory furnaces and then
purified in small hearths.
The total loss of copper in the preceding processes
amounts to about 10 per cent, 6.88 per cent of which takes
place in the ore smelting, 178 per cent in the matt smeltings,
and 134 per cent in the roasting and purifying process.
About 840 cwts. of copper are produced annually from
copper pyrites, and 550 cwts. of copper from cupriferous
matt derived from the lead smelting.
c. Smelting Process at Kupferberg in Upper Silesia.-The
ores* worked are partly copper, iron, and arsenical pyrites,
and variegated copper ore, with a little copper glance,
fahlerz, cobalt and nickel ores, and associated with amian-
thus, chlorite, serpentine, magnetic iron ore, calc spar, fluor
spar, brown spar, and zinc blende, containing from 4 to 20
per cent of copper (on an average 5 per cent) after having
been dressed. Partially oxidised ores are worked, such as
malachite, azurite, red copper ore, &c., which have the same
gangue as the sulphuretted ores; the amount of their con-
tained copper is very variable; cupriferous and argentiferous
residues from arsenic and vitriol manufactories are also
here submitted to the smelting process.
When smelted by themselves, sulphuretted ores are
previously roasted, the schlich being first mixed with lime
and formed into bricks; if oxidised ores are worked at the
*
COTTA, Erglagerstätten, ii.. 218.
IIO
COPPER.
same time, then both sorts of ore are mixed together. A
smelting mixture in common use consists of 75 cwts. of
oxidised ores and 25 cwts. of sulphuretted ores, containing
on an average 5 per cent of copper, 10 cwts. of fluor spar,
and 50 cwts. of black copper slags.
At the raw smelting, 2 cwts, of mixture are smelted with
1'42 cubic feet of coke, and in 24 hours 48 cwts. of mixture,
containing 30 cwts. of ore, are worked; the resulting raw
matt (about 5 cwts., containing from 30 to 35 per cent of
copper), is tapped off two or three times; a very siliceous
vitreous black slag is also produced. The matt is sometimes.
so rich in silver that it is sold to establishments where silver
is extracted.
The raw matt is roasted in lots of 40 cwts. eight or nine
separate times, and 100 cwts. of mixture, containing 35 cwts.
of raw matt slags, are smelted in charges of 4 cwts. (or of
5'4 cwts. of slags and fluxes), each charge requiring 1°42
cubic feet of coke. The effect of coke at the matt smelting
is 2.7 greater than at the ore smelting, other circumstances
being the same; the slag resulting from the ore smelting is
siliceous, and that from the matt smelting is neutral. From
80 to 85 cwts. of mixture are melted in one day; during that
time it is tapped off eight times, and from 15 to 31 cwts. of
black copper result, besides from 4 to 10 cwts. of matt,
which is returned to the smelting. The resulting slags
contain from 1 to 2 per cent of copper, and are added to the
ore smelting.
Black copper is purified in small hearths in charges of 3 or
4 cwts., and 18 cwts. of black copper yield 16 cwts of
rosette copper and slags, which are worked up in the ore
smelting. The copper contains auriferous silver.
d. At Dillenburg in Nassau (Isabellenhütte), copper pyrites*
containing from 20 to 25 per cent of copper is roasted once,
consuming 42 cubic feet of wood and 128 cubic feet of charcoal
to every 5 tons of ore. The roasting mass is smelted in
channel furnaces, 50 or 60 cwts. of ore consuming in 24 hours
* BURAT, in Ann. d. min., 4 sér., xiii., 351. COTTA, Erzlagerstätten ii., 151.
Bgwkfd., xiii., 33. B. u. h. Ztg., 1856, p. 158.
SMELTING COPPER SLATE.
III
48 cubic feet of coke, and producing 43°3 per cent of raw matt,
which is roasted three times and concentrated in a similar
furnace. At the concentrating process 29 per cent of black
copper and 37 per cent of concentrated matt results. This
matt is again roasted three times and then smelted for the
production of black copper, about 77 per cent of which is
produced together with some matt. 100 cwts. of ore and
intermediate products require for smelting 28 cubic feet of
coke and 5'31 cubic feet of charcoal. At the purifying of black
copper in small hearths, about 12 cwts. of rosette copper
result in 12 hours, consuming 132 cubic feet of charcoal.
The total yield of copper from the ores is 24 per cent.
e. Smelting of Copper Slate at Riechelsdorf and Friedrichs-
hütte.*—The smelting process is carried on with copper
slatet containing from 2 to 6 per cent of copper, and poorer
sand ores which, besides lime and clay in the proportion fit
for the formation of slag, contain chiefly variegated copper ore,
iron pyrites, red copper ore, copper glance, malachite,
azurite, native copper, fallow ore, molybdenum glance, zinc,
blende, galena, cobalt and nickel ores, and other minerals.
The copper slate, containing on an average from 3 to 4
per cent of copper, is roasted for three or four weeks in
open heaps containing about 3700 cwts. ; the heaps are
provided with a foundation of brush-wood. The outside ores.
are removed and re-roasted in small heaps with an addition
of small charcoal, either by themselves or together with fresh
ores. The sand ores are roasted in separate heaps.
Both sorts of ore are smelted together with an addition
of fluor spar, in channel furnaces, 18 feet high, as repre-
sented by Figs. 22, 23, and 24. The blast produced by the
bellows either enters from the reservoir, a, by the tuyere, b,
1½ inch in width, direct into the furnace, or it is previously
heated to 100° or 150° C. by being conducted by means of the
tube, c, into the lenticular iron reservoir, d, which is heated
* GENTH, in ERDM. J. f. pr. Chem., Nos. 3 and 4, de 1846; xl., ii., 189.
Bgwkfd., x., 305, 321, 337; xii., 223. B. u. h. Ztg., 1846, p. 617. RAMMELSBERrg,
Metallurgie, p. 236.
+ COTTA, Erzlagerstätten, ii., 77, 679. Berggeist, 1859, No. 74. B. u. h.
Ztg., 1860, p. 190.
II2
COPPER.
by the furnace gases, and then back by the tube, e, into f,
whence it enters the furnace by the tuyere under a pressure
FIG. 22.
FIG. 23.

d
P
P
Г
g
h
d
о
on
m
k
of about 20 inches of water; during this manipulation, ƒ, is
closed against the reservoir, a, by a slide. The furnace mouth
FIG. 24.

d
a
SMELTING COPPER SLATE.
113
is closed by a lid, g, and the furnace gases are collected at
h, conducted by two tubes, k, which run down the front wall
and terminate at m, where they heat the reservoir, d, the
flame escaping through the opening, n. The plates, p, serve
to protect the workmen from the heat; they move on hinges
at q, and may be turned upside down by counterweights, r.
The fused masses flow from the sole stone alternately out
of the eyes, n, into two outside basins one foot in diameter,
which are formed of two-thirds coal dust and one-third
sifted loam. 10 tons of ore are melted in 24 hours, producing
8 cwts. of raw matt, charging every hour two parts in weight
of mixture to one part of light coke. The resulting product
is similar to those obtained at Mansfeld, namely raw matt
(see analyses on page 43). According to Genth, the following
is the composition of raw slags :-
SiO 3
A1,0,
CaO
MgO
FeO
Cu,O
ZnO
KO
NaO
MOO
FeS
Cu₂S
MOS₂
I.
II.
4447 51'44
12'96 19'32
21'20 17.80
7.00
I'40
7.85
5.88
I*23
0*30 0*89
2*90 1'78
0·87
0.65
0°38
I'04
0*67
0'20
The slags are amorphous, mostly black, sometimes full of
blisters; their specific gravity is from 2.683 to 2.834, and
they are mixtures of mono- and bi-silicates. The cavities
are frequently lined with augite.* Besides deposits of iron,
zinciferous deposits and smoke result from this smelting.
Bunsen investigated the furnace gases obtained when
working with coke and hot blast, and the gases resulting
when employing hot blast and coke with 1-5th of charcoal,
VOL. II.
LEONHARDT, Hüttenerzeugnisse, p. 42.
I
114
COPPER.
also those resulting when using charcoal alone and hot blast,
and finally charcoal and cold blast.
1. Gases from the
Upper part of the
Furnace.
II. Gases from the
Lower part of the
Furnace.
II.
I.
II.
I.
Pressure of blast.
18"
19″"
19″″
19""
Temperature of the same 135.
10°
1432
ΙΟ
Temperature of the gases 300
285°
Collected below the furnace
mouths
6'
5'
12'
12'
Nitrogen
Carbonic oxide.
68.45
67'97
70'52
64.66
13.62
19'07
2'79
II 05
Carbonic acid
11.81 7°41
21°03
20'II
Carburetted hydrogen
2.63
3'77
I'47
0'53
Hydrogen
I'94
0'92
3'17
3'44
Sulphurous acid
I'55 0.86
1'04
0'21
The gases from the upper parts represent the medium
composition, containing the proportion of oxygen to nitrogen
nearly resembling that existing in atmospheric air, whilst
the gases from the lower parts of the furnace do not contain
them in the same proportion, as the fuel there becomes de-
composed with the evolution of volatile products by the heat.
The weight of the fuel (1, 2), burned in the same time
with equal quantities of air, and under the same circum-
stances, is respectively 100 and 109 parts. When using
charcoal and hot blast in iron blast furnaces it amounts to
147 parts.
The loss of fuel caused by the escape of gases amounts to
39°2 and 43.8 per cent respectively of carbon. In iron blast
furnaces this loss amounts to 75 per cent, on account of the
larger formation of carbonic oxide gas, caused by the greater
height of the furnace.
4
To this loss of temperature must be added that which is
absorbed by the gases themselves, and which amounts at
Friedrichshütte, where a mixture of coke and of char-
coal is employed, to 8.8 per cent; this causes a total loss of
50 per cent of fuel, which is increased to 58 per cent when
using charcoal and cold blast.
The gases from the raw smelting border on the limit of
combustibility, as they contain from 20 to 24 per cent of
combustible gases (CO, H, CH2). When using charcoal and
SMELTING OF COPPER SLATE.
115
hot blast in iron blast furnaces they amount to 30 per cent.
The heating power of the furnace gases stands in an inverse
proportion to the quantity of fuel consumed, as coal exerts the
greatest effect if converted at the moment of combustion into
carbonic acid; the reverse is the case with regard to carbonic
oxide gas.
The gases produced by employing a mixture of one-fifth of
charcoal and four-fifths of coke and hot blast produce when
burned a temperature of about 912° C; the gases formed by char-
coal and cold blast, a temperature of 1,c97° C. Their appli-
cability is similar to that of the gases of iron blast furnaces.
The raw matt is roasted nine times in nine months, in
mounds 16 feet long, 9 feet broad, and 7 feet high, and in
the following manner:-Wood in lumps and in the form of
brushwood, and small coal, forms the foundation, upon which
are laid 100 cwts. of raw matt, then a layer of coal, and again
100 cwts. of raw matt; the last fires require more fuel than
the first; the matt resulting from the black copper smelting
is added at the fifth firing; 100 cwts. of raw matt consume
162 cubic feet of wood, 120 bundles of brushwood, and 496
cubic feet of charcoal.
The roasted raw matt is smelted in admixture with ore
slags with good charcoal and cold blast in a small sump
furnace; 36 cwts. of matt are smelted in 24 hours, yielding
black copper, matt, and slag.
Two or three cwts. of black copper are purified in two or
two and a-half hours with charcoal and coke, in a hearth
about 9 inches deep and 26 inches wide, the tuyere having
an inclination of 30°. The resulting purified copper contains
nickel, and its thinner upper cakes are employed for the
manufacture of brass, and the thicker lower ones are tough-
ened. The hearth ends, rich in nickel, are smelted for the
production of a nickeliferous copper, the composition of
which is shown in the following analyses :-
Cu
Ni
I.
II.
III.
76.8 83.25 96*98
Fe
S
ค
13.6 12.82 2'99
4'0 3°40
0'20
5°I I'19
ΟΙΟ
I 2
116
COPPER.
No. I is copper produced from refinery slags of Riechels-
dorf, by Wille. No. 2, the same from Friedrichshütte,
uppermost cake, by Wille. No. 3, the same, lower cake.
At Thalitter, in the Grand Duchy of Hesse, copper slate
containing from 1 to 3 per cent is smelted direct in charcoal
furnaces for the production of black copper, which is purified
in small hearths; the annual production is about 230 cwts.
f. At Nischnetagilsk the usual mixture is about 30 cwts.
of copper ore,† containing from 25 to 60 per cent of copper ;
27 cwts. of copper schlich, containing from 1 to 25 per cent
of copper; 9 cwts. of copper pyrites, containing from 1 to 25
per cent of copper; and 21 cwts. of raw slag. It is smelted in
channel furnaces 12 feet high, receiving from 300 to 400 cubic
feet of air per minute. About 4 tons of mixture are smelted
in 24 hours, consuming 250 cubic feet of wood, and producing
about 13 cwts. of raw matt, containing 48 per cent of copper.
The raw matt is roasted in three fires, then mixed with
one-half of its weight of raw slag, and smelted for the produc-
tion of black copper and matt; 90 cwts. of mixture consume
about 250 cubic feet of charcoal; the charcoal is advanta-
geously replaced by wood. The black copper is refined in rever-
beratory furnaces. The use of superheated steam in the copper
ore smelting, and the employment of Rachette's furnace
(page 39) yielded favourable results.
Smelting in Crucible Furnaces.
Low furnaces of this kind are only used in rare cases for
copper smelting in India, Japan, &c.|| The resulting
matt is then taken out of the furnace hearths either in cakes
or in one piece, when it is broken, roasted, and smelted for the
production of black copper; the black copper is refined in
peculiar hearths. The Japanese method of casting copper
has been explained on page 78.
* ERDMANN'S Journ. f. ök. u. tech. Chem., xvii., 471. B. u. h. Ztg., 1862,
pp. 265, 384.
+ COTTA's Erzlagerstätten, ii., 543.
† B. u. h. Ztg., 1862, p. 117.
|| Ibid., p. 118. Bgwkfd., vi., 371; xvi., 17.
COPPER EXTRACTION AT THE DESILVERISATION OF ORES. 117
Illustrations of Smelting Argentiferous and Auriferous
Copper Ores by the German Method.
1. Extraction of Copper at the Desilverisation of Ores
and Matts.
a. At the district of Naygbanya, in Hungary.*-The gold,
silver, copper, and lead orest of this district, associated with
gangues, such as quartz, heavy spar, calc spar, felspar,
and occasionally gypsum, blende, and manganese, are
desilverised (by means of lead) at the smelting works of
Fernezely, Kapnik, Laposbanya, Olahlaposbanya, and
Sztrimbul; copper matt results as the final product of
these processes; it is treated, at Felsöbanya particularly, by
the following operations for the extraction of its copper.
1. The copper matt, containing from 35 to 40 per cent of
copper and o˚34 to 0°42 ounces of silver in 100 lbs., besides
lead, iron, sulphur, arsenic, and antimony in variable quan-
tities, is first roasted in quantities of 500 cwts. in alternate
layers with wood, for about a fortnight; more wood is
added at the following eight or nine firings. 75 cubic feet
of wood and 5 cubic feet of coal are required for roasting
1,000 lbs. of matt in ten fires.
The roasted matt is smelted in cupola furnaces 6 feet
high, in admixture with refinery slags in the proportion of
160 to 18; from 54 to 62 cwts. of mixture are smelted in 24
hours, 1,000 lbs. of matt consuming 48'64 cubic feet of coal.
The resulting concentrated matt contains from 60 to 65
per cent of copper, and is roasted in nine fires and added to
the black copper smelting. Besides the concentrated matt,
black copper, containing from So to 85 per cent of copper,
results (about 40 or 50 per cent of the quantity of matt
1
KERSTEN, in ERDM. J. f. pr. Chem., i., 193, (1834.) AUDIBERT, in Ann. d.
min., I livr. de 1845. Bgwkfd., x., 209. RIvor and DUCHANOY, in Ann. d.
min., 5 ser., 1 livr. de 1853, p. 63. Jahrb. d. k. k. Geolog. Reichsanst., 1853,
No. 3, p. 568. B. u. h. Ztg., 1853, p. 591; 1854, p. 25. Oesterr. Ztschr., 1854,
P. 81.
↑ Corra, Erzlagerstätten, ii., 291.
ť Jahrb. d. k. k. Geolog. Reichsanst., 1851, No. 3, p. 157.
1852, p. 145.
B. u. h. Ztg.,
118
COPPER.
smelted), and poor slags with per cent of copper, which
are thrown away.
The first refining (roasting or purifying) of the black
copper is effected in reverberatory furnaces; the hearth is
formed of loam well beaten down and covered with a
layer of quartz sand. 13 cwts. of copper are charged,
resting upon a foundation of straw, and the copper is slowly
melted in 48 hours, in order to avoid cracks in the furnace
hearth; next, 16, 18, and 20 cwts. of copper are charged in
succession. As soon as the copper is melted, the blast is
put on and made to react intermittently upon the surface of
the metallic bath, thus causing the formation, at the com-
mencement of the process, of a great deal of half-fused
oxidised substances; as
as soon as these substances cease
to be formed, and the surface of the metal bath appears
clear and bright, indicating the fineness of the copper, the
copper is tapped off into three moulds, and there formed into.
cakes, after having previously reduced the suboxide of
copper formed, by covering the surface of the copper with
charcoal. If the copper contains a considerable amount of
antimony and arsenic, charcoal is several times spread over
it previous to the tapping off, and the blast is made to react
upon it for a time after removing the charcoal; the tapping
off and forming into cakes (rosettes) takes about three or
four hours. From 78 to 80 per cent of cake copper is pro-
duced by this process, besides slag containing at least 20 per
cent of copper and 15 per cent of lead, causing a consumption
of 216 cubic feet of wood to every 1,000 lbs. of black copper.
Rosette copper is refined in a small hearth, in quan-
tities of 537 lbs., in three hours, causing a loss of 3 per
cent of copper, and consuming 72 cubic feet of charcoal to
1,000 lbs. of rosette copper. The fused copper is ladled out
and cast in cast iron moulds. The loss of copper at the
working of the cupriferous matt does not exceed 6 per cent.
The slag formed in the reverberatory furnace* is smelted in
admixture with from 35 to 45 per cent of raw pyrites not con-
taining gold; about 45 cwts. of mixture in 24 hours con-
suming 43 cubic feet of coal to 1,000 lbs. of slag. Some
KRAUS, Jahrb., 1854, p. 85. ·
SMELTING PROCESS AT MÜSEN.
119
argentiferous lead results from this process (about 40 per
cent of that contained in the scum), and matt containing from
35 to 45 per cent of copper, which is smelted for the pro-
duction of black copper after having been ten or twelve
times roasted.
The copper works at Felsöbanya treat annually 2,000
cwts. of matt, and produce from 800 to 820 cwts of copper,
partly in cakes or in ingots, and partly forged into bars.
The annual production of metal in the royal works at
Nagybanya amounts to about 300 lbs. of gold, 7,700 lbs. of
silver, 1,450 cwts. of copper, 40 cwts. of lead, 4,200 cwts. of
litharge, and 30 cwts. of sulphur; the private works pro-
duce 180 lbs. of gold and 100 lbs. of silver.
b. At the Gustav-Adolph Silver Works at Fahlun.*—The
copper lead matt resulting from the lead smelting, and con-
taining from 12 to 16 per cent of copper, and 13 ounces of
silver in 100 lbs. is roasted three times and smelted in
The resulting
admixture with quartz, litharge, and hearth.
alloy is submitted in a reverberatory furnace to a combined
process of liquation and oxidation for the extraction of its
lead and silver; the remaining impure copper is refined
in a small hearth.
c. At Müsen† copper matt which has been desilverised
by means of lead, is submitted to the following processes:
2
I. Concentration of the Matt.-The matt is twice roasted,
consuming 46 cubic feet of wood and 9 cwts. of charcoal, at
a total cost of £1 9s. per cwt. IO Cwts. of the roasted matt
are mixed either with 5 or 6 cwts. of once roasted copper ores
poor in silver and 15 cwts. of lead slags, or with from to
I cwt. of refinery scum and 12 cwts. of lead slags, and smelted
in more or less low cupola furnaces. In the very low
furnaces from 40 to 50 cwts. are smelted in 24 hours, and in
furnaces somewhat higher from 70 to So cwts. of matt, ore,
and scum, exclusive of the addition of slag, consuming
respectively from 42 to 50 or from 50 to 60 cubic feet of coke,
so that in the lower furnaces 7 lbs. of mixture (including
*
Bgwkfd., xi., 601.
Preuss. Ztschr., 1862, Bd. 10.
Lief. 3, p. 203,
120
COPPER.
fluxes) are smelted by 1 lb. of coke, and in the somewhat
higher furnaces 9 lbs. The resulting products are—
a. Crude Lead, containing about two ounces of silver in
the cwt., or plumbiferous black copper at the end of the process
in the lower cupola furnace, when the roasting has been too
strong; therefore, the hearth sole of the lower cupola furnace
is constructed with an inclination of 15°, thick pieces of
wood are placed upon it, and upon them several pieces of
the impure copper. Upon burning the wood, lead containing
about 2 ounces of silver in the cwt. will melt out, while the
remaining metal, termed kiehnstöcke, containing about 65
per cent of copper, 5 per cent of lead, and an ounce of
silver, is added to the black copper smelting.
b. Concentrated Matt, containing about 55 per cent of
copper, 5 per cent of lead, and 3 of an ounce of silver.
c. Slag, containing about I per cent of copper.
2. Black Copper Smelting.
The concentrated matt is roasted about nine times, with
respectively 27'5, 30, 32'5, 35, 37*5, 40, 42*5, 45, and 50, making
together 340 cubic feet of charcoal and coke. The kiehnstöcke
are roasted by themselves in the first fire, being placed in one
corner of the heap. 100 cwts. of matt consume 144 cubic
feet of wood, 117 cubic feet of charcoal, and 106 cubic feet
of coke, costing nearly £5 8s. The roasted matt is smelted
in a lower cupola furnace constructed with two open eyes, with
an addition at the commencement of the process of slags from
the concentration smelting, and afterwards of slags from the
same process. Either from 40 to 50, or from 60 to 70 cwts.
of matt are smelted in 24 hours, consuming 44 cubic feet of
charcoal and about 90 cubic feet of coke; 1 lb. of coke
smelts 5 lbs. of mixture, allowing 7 cwts. of slag to 10 cwts.
of matt. Black copper and black copper matt are formed
into cakes; the black copper contains about 70 per cent of
copper and of an ounce of silver, and the black copper
matt 60 per cent of copper and an ounce of silver; the
slag contains I per cent of copper. The black copper matt
(Dünnstein), is either returned to the furnace, or is roasted
dead and added at the refinery process.
SMELTING PROCESS AT MUSEN.
121
3. Refining the Black Copper.
When new, the hearths are 10 inches in diameter and
5 inches deep, and are capable of containing 60 or 70 lbs. of
copper, whilst a hearth which has been several times used
will contain from 150 to 180 lbs. The copper tuyere has an
inclination of 6°; it is 12 inches long and its mouth is
1 inch in diameter. The hearth is formed of three parts of
12
refractory sand and one part gestübbe. Part of the copper
and a small addition of roasted black copper matt are smelted
with coke; the slag formed is then removed together with the
coke, and in order to evolve sulphur and to oxidise foreign
metals, a low blast is made to react upon the copper so long
as it remains sufficiently fluid. Fresh copper and coke are
next charged, and the process repeated until the hearth is
filled, and a sample shows its fineness; from 150 or 180 lbs.
are refined in about four hours. The copper is ladled into
separate hearths prepared for the purpose.
From 5 to 6 cwts. of black copper, besides some matt, are
refined on an average on one hearth in 24 hours, consuming
about 29 cubic feet of coke and about 3 cubic feet of charcoal
for warming the hearth. The rosette copper contains about
one ounce of silver. One hearth lasts 35 hours without
repairs, and produces about 17 cwts. of rosette copper.
These processes, adopted at Müsen and Rothenbach, have,
since 1859, been considerably improved at the Loher Smelting
Works. Part of the ore is now roasted in reverberatory
furnaces instead of in heaps, with the following advantages:-
reduction of the roasting period to 19-20ths of its former length,
thus saving a considerable amount of interest of capital, and
metal; independence of the weather; the more uniform and
perfect roasting without evolving noxious vapours. The
smelting process comprises the following operations :—
1. Ore Roasting.-The ores, consisting of grains of
2 millimetres, and of schlich, are mixed with some cinders
and roasted in a Ramsbecker furnace, 64 feet long out-
side, 14 feet broad, and 6 feet high. The hearth lies.
3 feet above the floor of the smelting house, and is provided
near the fire-bridge with a flat excavation 5 feet long, I feet
122
COPPER.
high, which is beaten out with refractory sand. The exca-
vation and the grate are separated into two parts by a vertical
division wall. The flat arched roof is raised 1 foot 8 inches
above the hearth sole, and each long side of the furnace has
cighteen working doors. The flames of both grates unite
upon the hearth, pass across it long-ways, the hot gases and
vapour are conducted for 18 feet to and fro beneath the hearth
for depositing contained ore and escaping dust, and then enter
a condensation channel 130 feet long, which communicates
with a chimney 50 feet high. This channel is intended to
receive the smoke of all the smelting and cupelling furnaces.
About 6 tons of lead ores are roasted in 24 hours, consuming
35 cubic feet of coal. The ores are charged every hour in
small lots by the last working doors, and are spread out upon
the hearth 4 or 5 inches deep; they are gradually moved
towards the fire bridge, and every six hours the somewhat
caked masses are removed from the deepening of the hearth
out of the furnace; therefore the ores remain in the furnace
from 18 to 24 hours. When roasting ores at a higher tem-
perature so that they become scorified, a considerable loss of
metal, chiefly of copper, takes place. The fire is stirred.
every 10 or 15 minutes, and an equal temperature sustained
during the whole process. One furnace is attended by five
smelters and one assistant workman; each smelter receives.
Is. 6d. for one shift of twelve hours, and the assistant Is. 3d.
The loss at the roasting amounts to 8 or 9 per cent; some of
this loss is caused by the moisture of the ores, and o'275 per
cent by the ore dust, containing 10 per cent of lead, and
3-10ths of an ounce in the cwt. This method of roasting
costs 3'3d. per cwt. of ore, whilst roasting in heaps costs
3'1d.
2. Smelting Lead Ores.-Mixtures of 10 cwts. of lead
ores 2 cwts. of copper ores (poor in copper and argentiferous),
o'5 cwt. of fallow ores, 15 cwts. of iron refinery cinders,
o'5 cwt. of old iron, and 12 cwts. of lead slags, are smelted
in cupola furnaces, 6 feet high above the tuyeres, and having
each 3 tuyeres.
54 cwts. of ore exclusive of fluxes and
furnace ends of the former smelting are smelted daily,
yielding 1 cwt. of matt to every 3 cwts. of resulting crude
SMELTING PROCESS AT LOHE.
123
lead. From 10 to 11 lbs. of mixture are smelted with 1 lb.
of coke, formerly, only 9 lbs. of ore could be smelted when
roasting in heaps.
3. Lead Matt Smelting.-Owing to the more perfect
roasting of the ore the matt only requires to be roasted and
smelted once, or at most twice. The smelting operation is
performed in the ore furnaces, and about 45 cwts. of mixture
are smelted with 31 cubic feet of coke in 24 hours. If the
resulting matt is still too rich in lead and poor in copper, it
is re-smelted in admixture with roasted ores rich in silver,
after having been previously twice roasted.
4. Desilverisation of the Copper Matt.-For its desilver-
isation, copper matt containing 1 or 1 ounces of silver
requires to be twice melted, so that it may contain not more
than about an ounce. A mixture consists of 10 cwts. of
copper matt, containing 137 ounce of silver and 20'6 per
cent of lead, o˚5 cwt. of ores rich in copper, 4 cwts, of hearth
and litharge, and 10'5 cwts. of lead slags. 46 cwts. of
mixture are smelted in 24 hours, at a consumption of 31 cubic
feet of coke at the first desilverisation, and at the second,
4172 cwts. of mixture with 28 cubic feet of coke.
5. Concentration of the Desilverised Copper Matt.-
This matt is first roasted moderately in open heaps, and then
smelted without any addition of copper ores, which always
contain silver, but simply with an admixture of lead slags.
If the resulting slags are too basic in their nature some scum
of the refining process is added. This smelting may also be
effected in the furnaces used for the ore smelting, in which
54'05 cwts. of matt are smelted in 24 hours, at a consumption
of 31 cubic feet of coke.
6. Black Copper Smelting.-The concentrated matt is
finely divided and is roasted as perfectly as possible in the
same manner as the lead ores in reverberatory furnaces, from
90 to 100 cwts. being treated in 24 hours at a consumption
of 39 cubic feet of coal. The roasted matt is smelted in
channel furnaces with two open eyes, 77°29 cwts. in 24 hours
producing black copper and matt, and consuming 67 cubic feet
of coke. The black copper is refined in the usual way, and the
matt is roasted five or six times in open heaps and smelted
124
COPPER.
in the same kind of furnaces, producing a very pure black
copper.
II. Extraction of Copper at the Desilverisation of Matt.
At Oeblarn,* in Upper Styria, a mixture of copper pyrites,
fallow ore, and red silver ore, containing 1 per cent of copper,
from o'006 to 0'007 of silver, and some gold, is roasted in
open heaps, and the resulting sulphur, realgar, and sulphate
of copper are collected. The roasted ores are then smelted
with an addition of clay slate and slags in cupola furnaces
8 or 9 feet high, producing raw matt and furnace ends. The
raw matt is smelted for its desilverisation in admixture with
litharge, hearth, Villacher lead, and some furnace ends,
yielding a lead or copper matt, speiss, and lead containing
sufficient silver to be worth cupelling. An analysis of the
resulting speiss is given on page 44.
The lead or copper
matt is roasted and smelted with an addition of clay slate
and slags, producing copper matt and a caked mass difficult
to fuse, which is called hartwerk, and is rich in silver. The
matt is ten or twelve times roasted and smelted for the pro-
duction of black copper, which is finally refined in blast
reverberatory furnaces.
III. Extraction of Copper at the Desilverisation of
Black Copper.
a. Smelting Works of Oker in the Lower Hartz.-The ores
and products treated here for the extraction of copper are the
pure copper ores of the Rammelsberg, the copper-lead matts.
of the lead smeltings, and the cupriferous residues.
1. Treatment of the Copper Ores.-These consist of an
intimate admixture of copper pyrites, iron pyrites, zinc
blende, arsenical pyrites, antimonial ores, &c., and some
small quantities of earths; they contain from 1-16th to th
ounce of silver per cwt., and 1-7,300,000 of gold.
They are roasted in kilns, and the sulphurous acid thus
formed is used for the production of sulphuric acid.
In this
TUNNER'S Jahrb., 1847, p. 74; 1853, P. 348. Jahrb., d. k. k. Geolog.
Reichsanstalt, 1850, No. 2, p. 343. COTTA, Erzlagerstätten, ii., 313.
+ KNOCKE, in B. u. h. Ztg., 1859, p. 362. KERL, Rammelsberger Hütten-
processe, 186r, Anh. pp. I7, 25.
SMELTING PROCESS AT OKER (LOWER HARTZ). 125
manner 15 per cent of the sulphur is made profitable, whilst,
when roasting in open heaps, which was formerly practised,
only from to I per cent of sulphur was gained. After the
roasting in kilns, the ores are roasted twice in sheltered
heaps.
The roasted ores contain from 6 to 8 per cent of copper,
and are smelted in cupola furnaces, mixed in the following
proportions:-
329 cwts. of roasted ores, about 4'25 cwts. of clay slate.
containing small veins of copper ore;
2.18 cwts. of burned clay slate, about 4'84 cwts. of slags of
the matt smeltings;
besides furnace ends and ore slags, according to require-
ment.
The furnaces are represented by Figs. 25, 26, and 27.
FIG. 25.
(?

C
a
W
77
foundation; b, flues for the escape of moisture; c, sole stone;
d, back wall; e, pillar; f, counters; h, chimney; i, funnel;
k, layer for the tuyere; 1, tuyere; m, tuyere arch; n, fore-
hearth; o, outside basin for receiving metal and matt ; þ, sole
126
COPPER.
FIG. 26.

if
ན
C
α
Ъ
α
FIG. 27.

G
of the furnace, formed of a mixture of small coal and loam ;
r, front wall; t, sump for collecting the slag. The smelting
is effected by keeping a nose 5 or 6 inches long, and giving
250 cubic feet of blast per minute. The mixture is smelted
SMELTING PROCESS AT OKER (LOWER HARTZ).
127
in three hours, consuming 2'933 cubic feet of charcoal and
114 cubic feet of coke, and producing as chief product a sort
of speiss called königskupfer, containing 89 per cent of
copper and from 2 to 3 ounces of silver. Its formation is
caused chiefly by the presence of arsenic and antimony in
the ores. An analysis by Bodemann shows its composition
to be as follows:-
Cu.
Pb .
Fe
Ni
Co
Ag
Sb
81.87
10.26
2.75
traces
traces
0°22
As
S
2.55
Ι'ΟΙ
o'60
It is submitted in quantities of 60 cwts. to an oxidising
smelting, and then granulated, which process yields 42 or 43
cwts. of granulated copper. The furnaces used are repre-
sented by Figs. 28 and 29; their construction is thus:-
a, foundation; b, flues for the escape of moisture; c, rough
walling; d, opening for the slags; e, working opening ; ƒ, iron
FIG. 28.

10
ان
7~
T
n
Z
go og
ga
h
B
Ъ
Ъ
t,、,。
S
10
gutter; g, tuyeres; h, fire-place on which bundles of wood
are burned; k, fire-bridge; m, opening in the arched roof for
128
COPPER.
Ꮴ
F
y
f
FIG. 29.

ՂԱՆ
-D
0
cooling the furnace; n, cover to it; o, iron pillar founded in
the sole of smelting house, and to which the grappling irons
are fixed; p, slag hearth; q, hearth formed of bricks ;
r, smelting sole formed of loam and a mixture of small coal
and loam; s, interior walling; t, grate; u, blast tube;
v, granulating tub with an iron cover and an opening (y) for
admitting the copper from the gutter, z; x, water reservoir;
w, gutter for conducting the water to the tub.
The granulated copper is treated with dilute sulphuric
acid, as explained in chapter ii., Silver (vol. i., p. 399).
The resulting copper vitriol* is a commercial product; the
argentiferous and auriferous residues are smelted together
with litharge for the production of crude lead. This mode
of desilverisation gives a larger yield of silver and gold, and is
simpler in its manipulations, and cheaper than the former
desilverisation of black copper by means of lead. One disad-
vantage of this method is that the copper is obtained in the
state of blue vitriol, which does not always find a good market.
In such cases the vitriol may be smelted after being heated
KERL, in B. u. h. Ztg., 1860, p. 66.
TREATMENT OF INTERMEDIATE PRODUCTS.
129
with coal, or better, with saw dust, or it is added to the ore
roasting with an addition of coal, as at Fahlun. Rochel*
decomposes the copper vitriol by heat, for the extraction of
its sulphuric acid.
The following products also result:-
3
From 24 to 24 cwts. of raw matt, containing 48 to 52 per
cent of copper and to 1 ounce of silver. It is roasted from
two to four times; its working we will describe later on;
4
24 cubic feet of raw slags, containing from to 3 per cent
of copper, which are thrown away if pure; and
Some smoke, containing 2 or 3 per cent of copper, which
is roasted in reverberatory furnaces, and treated with dilute
sulphuric acid for the production of copper vitriol.
The roasted raw matt is smelted in admixture with 1-12th
of clay slate in the ore smelting furnace, producing-
Black copper containing 92 per cent of copper;
Copper matt containing 60 per cent of copper.
The purification of the black copper is effected in small
hearths in quantities of from 3 to 3 cwts., which takes
about three hours; from 10 to 15 cubic feet of charcoal are
consumed for I cwt. of copper. The copper is made dry and
again reduced to fineness by an addition of lead, 2 lbs. to
I cwt. of copper.
IV. Treatment of the Cupriferous Intermediate Products
and Residues.
a. The Copper Lead Matt (vol. i., p. 123) from the lead
smelting, containing about 60 per cent of copper and from
I to 1 of an ounce of silver per cwt., is treated as the
copper raw matt for the production of black copper. The
black copper is submitted to an oxidising smelting, and refined
in small hearths, together with the black copper of the copper
smelting; or it is treated with sulphuric acid.
b. Copper Lead Matt (vol. i., p. 185).—It is oxidised in a
reverberatory furnace and granulated; if the resulting copper
contains 5 or 6 ounces of silver, the silver is extracted by
VOL. II.
* Oesterr. Ztschr., 1857, p 283.
K
130
COPPER.
means of sulphuric acid. The slags resulting from the oxi-
dising smelting are re-smelted for the production of lead and
speiss; the speiss is then submitted to an oxidising smelting
and treated with sulphuric acid.
c. Refinery Slags resulting from the refinery process in
small hearths; they are smelted for the production of black
copper and matt. The black copper is submitted to an
oxidising smelting and afterwards to a refining in air rever-
beratory furnaces; the roasted matt is smelted to produce
black copper, which is then treated in reverberatory furnaces.
in the same manner. In 1861, 4599 cwts. of refined (tough)
copper were produced at Oker.*
Treatment of Copper Lead Matt at the Upper Hartz.*---
Such a matt from Altenau, Lautenthal, and Andreasberg,
containing from 22 to 25 per cent of copper, 6 to 14 per
cent of lead, and 0‘03 to 0‘'05 per cent of silver, is roasted
from six to nine times, in the same manner as the raw matt
resulting from the smelting of copper pyrites. When roasted
it is smelted, and yields black copper containing o'08 to 0‘30
per cent of silver and 70 to 98 per cent of copper, and some
matt (Dünnstein) containing o'03 per cent of silver and
70 to 75 per cent of copper. 100 cwts. of matt produce
about 15 cwts. of black copper and o‘6 cwt. of matt.
The copper lead matt at Andreasberg is rich in antimony
and arsenic, and is submitted, in quantities of 40 cwts., to an
oxidising smelting in blast reverberatory furnaces, the first
time for eight or ten hours, and the second for twelve or
sixteen hours, in quantities of 50 cwts. The resulting copper
matt, containing from 20 to 25 per cent of copper and o‘06 to
o'09 per cent of silver, is smelted for the production of black
copper containing about o'16 per cent of silver, after having
been roasted from fifteen to eighteen times.
The black copper is desilverised by means of lead and the
liquation process, the remaining copper (Kiehnstöcke) is
submitted, in quantities of about 43 cwts., to an oxidising
smelting in reverberatory furnaces. At this process from
three to four hours' firing is given without employing the
KERL, Oberharzer, Hüttenprocess, 1860, pp. 519, 640, 674, 698.
SMELTING PROCESS AT CSIKLOVA.
CESS
IZI
blast, which is then put in operation and made to react upon
the fused copper for six or eight hours, whilst the formed
slag is skimmed off. 5 tons of Kiehnstöcke produce about
55 cwts. of metal and 45 cwts. of slag. The metal is refined
in a small hearth, and the slag is treated, together with
copper refinery slags, for the production of copper mica.
The annual production of this copper at the Upper Hartz
is about 500 or 550 cwts.
At Csiklova* in Banat, copper ores, consisting of copper
pyrites, fallow ore, iron pyrites, arsenical pyrites, blende,
calc spar, quartz, and containing from 3 to 4 per cent of
copper and a small amount of silver and gold, are smelted
together with basic rich slags, producing raw matt. This
raw matt is twice roasted in mounds and afterwards smelted
in low cupola furnaces with an addition of siliceous slags,
when it yields a concentration matt containing 35 or 36 per
cent of copper. This latter matt is three times roasted and
re-smelted in low cupola furnaces with an addition of slags
from the former process, yielding an arsenical and antimonial
black copper containing about 70 per cent of copper, and
some black copper matt (Dünnstein) containing from 60 to
65 per cent of copper. This matt is twice roasted and then
added to the black copper smelting. The black copper,
yielding o'68 ton of rosette copper, and 1741 kilos. of silver
to the ton, is desilverised by the amalgamation process. The
roasting at this process is effected with common salt, causing
an evolution of antimony and arsenic. Whilst still in a
humid state, the cupriferous residues of the amalgamation
process are mixed with 25 per cent of pulverised iron pyrites
and 6 per cent of powdered charcoal and formed into bricks,
which are then smelted in high narrow cupola furnaces, in
admixture with some siliceous slags. The resulting products
are black copper containing about 85 per cent of copper, and
matt with about 64 per cent of copper. The matt is roasted
in mounds twelve times, and smelted in admixture with
* CHANCOURTOIS, in Ann. des mines, 4 sér., tom. x., livre 6, de 1846, p. 577-
RIVOT and DUCHANOY in Jahrb., d. K. K. Geolog. Reichsanstalt, 1853, No. 4.
p. 807. RUSSEGGER, in KARSTEN'S Archiv. 1, R. ix,, 405. KRAUS, Jahrbuch,
1852, p. 128. RIVOT, Métallurgie du Cuivre, 1859, p. 513.
2 K
132
COPPER.
pyrites, slags, and residues of the refinery process. The
resulting black copper, which is somewhat purer, is purified
in small hearths; a little matt also results, which is worked
up if collected in sufficient quantity. The former black
copper is purified first in blast reverberatory furnaces, and
afterwards in small hearths, yielding a tough copper of
medium quality.
I ton of ore, containing o'033 ton of copper and o'0835 kilo.
of silver, yields 0.0305 ton of copper and 0.077 kilo. of auri-
ferous silver, consuming 1*33 tons of fuel, at a total cost for
labour, smelting materials, and tools, of 19s.
TREATMENT OF SULPHURETTED ORES, &c., IN
REVERBERATORY FURNACES.
The process of extracting copper in reverberatory furnaces
has been largely developed in England, and its chief points
are based upon the principles forming the basis of the
extraction of copper in cupola furnaces. It comprises the
following operations :-roasting the ores; reducing and
purifying smelting the roasted ore for the production of
matt; roasting and concentrating the matt several times,
according to whether more or less foreign substances, chiefly
antimony and arsenic, are present; treatment of the
roasted concentrated matt for the production of black copper;
and finally, refining the black copper. At the reducing
and purifying smeltings of the roasted ores and matt in
cupola furnaces, the reducing agents are chiefly coal and
carbonic oxide gas from the reducing re-agents; but in
smelting in reverberatory furnaces it is the sulphur of the
sulphides, which have remained undecomposed in the roasting
processes, that chiefly reduces, by decomposing the oxides and
salts, becoming itself converted into sulphurous acid.
There is also a difference in the refining, as by conducting
this process in reverberatory furnaces the copper becomes at
once toughened.
When smelting copper ores in reverberatory furnaces, the
resulting slags, which are thrown away, are usually richer in
copper than those resulting in cupola furnaces, and the
ENGLISH COPPER SMELTING PROCESS.
133
concentration of copper in the matt in reverberatory furnaces
aiso takes place in a considerably less degree; this is caused
by the more imperfect roasting of large quantities in a limited
time, wherefore more concentration smeltings are required.
Slight differences in the form and size of the apparatus,
chiefly of the roasting furnaces, exist in the different
copper works, and also in the processes themselves, according
to the quality of the ore, with regard to its contained metal,
its purity, its oxidised or sulphuretted state, the quality
of the products, &c.
The English copper smelting process comprises at least
six operations, viz. :-
1. Calcination of the Ores.
2. Fusion of the Calcined Ore for the production of coarse
metal.
3. Calcining the Raw Matt (coarse metal).
4. Fusing the Roasted Raw Matt for producing fine metal.
5. Roasting the Fine Metal for the production of black
blister copper (pimpled metal).
6. Refining or Toughening the Copper.
Impure ores may render necessary a repetition of the matt
concentration smeltings, and more operations may be re-
quired in producing copper of the first quality, in separate
smeltings of slags, &c. These circumstances, according to
Le Play,* necessitate the following ten operations :—
1. Calcination of the Sulphuretted Ores.
2. Production of Raw Matt from the Roasted Ores.
3. Calcination of the Raw Matt.
4. Production of the Common White, or the Concen-
trated Matt, by fusing the roasted raw matt with rich ores.
5. Production of Blue Concentrated Matt (blue metal)
by fusing roasted raw matt with roasted ores containing a
medium amount of copper.
6. Production of White and Red Matt (or metal) from
slags Nos. 4, 7, and 8.
* LE PLAY, Description des Procédés Métallurgiques Employés dans le pays
de Galles pour la fabrication du Cuivre, Paris, 1848. (Deutch, von Hartmann,
1851.
134
COPPER.
7. Roasting Smelting the Blue Matt No. 5, and produc-
tion of white extra matt.
8. Roasting Smelting the White Extra Matt and pro-
ducing concentration matt.
9. Roasting Smelting the Common White Matt, the
Concentrated Matt, and the Cupriferous Hearth Bottoms
for the production of black copper.
10. Refining the Black Copper.
The copper smelting processes in reverberatory furnaces
are performed on the Continent much more simply than in
England, comprising only a roasting and raw smelting of
the ores, one or two concentrations of the raw matt, a
roasting and smelting of the concentrated matt for the pro-
duction of black copper, and the refining of the black copper.
In the English copper works, an addition of copper
ores of different purity, and containing unequal amounts of
metal, is made to the roasted raw matt, and thus a copper
of different quality is the result. This is also the case at the
separate smelting of slags.
During a number of years suggestions have been made for
the improvement and simplification of the complicated English
copper smelting process. These gave a partially good result,
but did not wholly admit of a practical application. The pro-
posals were with a view to improve the roasting processes,
and to cause a better yield in the smelting processes by using
very different fluxes. Gurlt* has studied these proposals,
and carefully superintended the execution of them in the
copper works. He has stated his conviction that all the
operations required for working copper ores in reverberatory
furnaces may be reduced to not more than two roasting pro-
cesses at the highest, and three smelting processes, viz. :-
1. Calcining the Ores in a Pulverised State in Parkes's
double roasting furnace, provided with an apparatus for
turning the roasting mass, and with an addition of a chloride,
such as common salt or chloride of calcium, for the formation
of volatile chlorides.
* GURLT, Bemerkungen über die Fortschritte des Kupferhüttenprocesses
in England. B. u. h. Ztg., 1852, p. 265, No. 16.
CALCINING THE SULPHURETTED ORES.
135
2. Smelting the Roasted Ores for the production of raw
matt, forming at the same time a siliceous slag.
3. Roasting Smelting the Raw Matt with or without the
addition of a chloride, by hot blast for the production of black
copper.
4. Refining the Black Copper.
In cases when it is intended to extract a pure copper from
stanniferous ores, a separation smelting of the raw matt will
be required. This is usually combined with the smelting of
rich slags, and is similar to a process we shall describe
later on.
These propositions have not, however, been adopted here;
the processes in use retain their original complexity, and the
roasting apparatus only has been somewhat modified. The
chief reason appears to be that these modifications would
interfere with the necessary connection between the quality
of the produced copper and the quality of the ores standing
for sale.
In the following pages it is proposed to give a statement of
the theory of the English roasting and smelting processes, a
description of the apparatus, and the newest improvements.
It will perhaps be as well here to enumerate the names of
the most recent authors who have written on the English
copper smelting process, in addition to Le Play and Gurlt,
whose writings we have already mentioned. These are-
Warington Smyth, in his lectures at the Royal School of
Mines, London, J. Arthur Phillips,* Hyde Clark, Napier,:
Percy, and Rivot.§
Calcining the Sulphuretted Ores.
This process is required for the reasons stated on page 18,
and is performed either in special reverberatory furnaces
or in furnaces used at the same time for smelting. These
* A Manual of Metallurgy, by J. A. PHILLIPS: London, 1852, p. 350.
+ Mining Journal, 1858, No. 1215. Allgem., B. u. h. Ztg., 1859, pp. 6, 11,
49, 179, 229.
Philosophical Magazine, 1852, vol. 4, p. 45; vol. 5, p. 30.
Metallurgy, by JOHN PERCY: London, 1851, vol. 1, pp. 289–382, 454, 494.
B. u. h. Ztg., 1852, pp. 316, 346.
§ RIVOT, Metallurgie du Cuivre Paris, 1859, pp. 117, 194.
HARTMANN: Leipzig, 1860, pp. 73, 116.
Deutch von C.
136
COPPER.
furnaces are usually provided with a direct firing, as in Eng-
land, Freiberg, Dillenburg; at Freiberg gas reverberatory
furnaces are also used; and in Wales they are provided with
clinker grates,* which are chiefly adapted for the use of dusty
fuel. Clinker grates are formed artificially by means of the
ash of the fuel which rests about o'6 metre high upon iron
bars, and upon this the fuel is placed o'6 to 0.7 metre high. The
air enters by channels formed in the ash, where it is warmed,
and then comes into contact with the fuel, which evolves
combustible gases, which are richer in carbonic oxide gas, the
more quickly the air is admitted. The coal used at the
Hafod smelting works, near Swansea, consists of a mixture
of equal parts of caking coal of Mynydd-Newydd (a), and of
anthracite coal of Tyrcenol (b), and Pentrefelin (c). They
form a caking ash, and, according to Percy, are composed as
follows:-
a.
b.
C.
Carbon.
73.87
76.81
78.49
Hydrogen.
3'73
3°42
3'73
Oxygen and nitrogen
8.02
5.65
4'15
Silica
5'05
4.68
4.24
Alumina
3'75
3'74 3°29
Peroxide of iron.
0.88
ΟΙΟ
•
Lime
0.83
0.60
0'27
Magnesia
0*28
0*28
0°19
Potash.
0'36
0*39
0.16
Soda.
0'09
0'12
0'07
Sulphuric Acid
0'23
0'54
0.69
Iron pyrites
(iron
(sulphur.
•
1*36
1'71
2.16
I'55 I'99
2:56
The air required for burning the coal gases enters the
furnace by the separate openings. Having a greater specific
gravity than the hot gases it fills the space above the roasting
mass, exerting a strongly oxidising action upon it, whilst the
hot combustible gases above come into contact with the
atmospheric air at their lower part and burn, thus creating
the uniform and moderate temperature required for the
roasting process.
The roasting furnaces vary chiefly in the dimensions of
Allgem., B. u. h. Ztg., 1859, p. 50.
CALCINING THE SUPHURETTED ORES.
137
the hearth; these dimensions are fixed chiefly by the propor-
tion of the roasting to the other processes.
The smaller furnaces are capable of containing 3 or 4 tons
of ore, and the larger ones, described by Percy, 7 tons.
FIG. 30.

G
HAMM
α
a
JJANGKRUDTESTA DYLJ|
2
4
• Met.
Figs. 30 and 31 represent one of the smaller furnaces; the
drawings are sufficiently explicit to be understood without
further description.
FIG. 31.

α
1
3
5
6 Met.
Figs. 32 and 33 show certain modifications of this kind
of furnace ;* a is the stoke-hole; b, the grate; c, the fire-bridge;
d, the chimney; e, e, working openings; ƒ, f, iron hoppers; g, g,
openings in the roof; h, the hearth sole with the holes i, i. The
hearth has a suitable oval shape, and is covered with a flat
arch. Its length is 16 feet, breadth 133, mean height 2 feet.
Fig. 34 shows a different shape of the hearth.
* Dr. URE's Dictionary of Arts, &c., vol. i., p. 881.
138
COPPER.
The following statements may be made with regard to the
chemical reactions which take place in the roasting process.
The sulphur must not be expelled completely by the roasting,
FIG. 32.

but as much must remain as is required at the subsequent
raw smelting for the formation of sulphide of copper, and
of sufficient sulphide of iron for the amount of iron in the
FIG. 33.

f
d
f
B
k
9
Te
h
α
b
latter to be about equal in weight to the copper present.
Antimonial and arsenical ores require a strong roasting with
an addition of pure pyrites, if it is possible to add a sufficient
FIG. 34.

quantity of pyrites to the mixture for the raw smelting.
When this is not possible the roasting must not be carried
CALCINING THE SULPHURETTED ORES.
139
en so far. The sulphides are converted by the roasting pro-
cess partly into oxides, sulphates, antimoniates, and ar-
seniates, and partly into lower sulphides, while anhydrous
sulphuric acid is evolved, which, by absorption of the water
contained in the atmosphere, forms a visible white vapour.
The roasting process is to be judged by the more or less
caking of the ores after raking, the temperature of the fur-
nace, the nature of the escaping flame, the time, &c. Le
Play and Napier* have investigated the roasting process by
analyses of the roasting products.
Vivian's older experiments of condensing the roasting
smoke by conducting it through long moist channels, gave
unfavourable results, and when using that smoke for the
production of sulphuric acid, the draught of the furnace was
impeded; this would not have been the case if some blast
had been conducted under the grate. The smoke has hitherto
been conducted into high chimneys, injuriously affecting the
neighbouring vegetation, but it is now for the most part con-
densed and used in the formation of sulphuric acid.
According to J. Cameron, ore of the composition-
Cu.
Fe.
S
SiO3
8
24
23
45
100
is modified by the roasting process as follows:-
18 parts Cu
2
""
S
(8Cu
{SCu) Cu₂S
2S
Sulphide of copper
IOS lost or replaced by
Sulphide of iron
21S
(24Fe
34
protoxide of iron
II S
Silica
45
12 Fe sulphide of iron
45
If besides the sulphuretted ores a sufficient quantity of
oxidised ores are available, the sulphuretted ores need not be
roasted; this is the case at the copper works on the Elbe.
* Phil. Mag., 1852, vol. 4, p. 45; vol. 5, p. 20. PERCY, Metallurgy, i., 333.
140
COPPER.
The following proposals or experiments have been made
for the improvement of the present roasting process:-
I. Parkes's method.-Gurlt* has proved that when roasting
large charges of 7 tons in common reverberatory furnaces
only one-fifth of the charge becomes modified by the roasting,
consequently the copper can only be slightly concentrated at
the subsequent raw smelting. The roasting is much more
perfect when the ores are calcined in a pulverised state in
Parkes's double furnaces,† which are provided with an
apparatus for raking the ore, and when the ore is mixed with
5 per cent of substances containing chlorine (chloride of
sodium, calcium, barium, or muriatic acid and small coal),
so as to effect the volatilisation of chlorides of antimony and
arsenic. These furnaces were formerly employed at the
Pembrey copper works in Carmarthenshire; one is repre-
sented by Figs. 35 and 36.
FIG. 35.
FIG. 36.


T
で
​9
Z
q
k.
Ъ
Ъ
f
P
k
h
d
d
୯
n
C
a is the lower circular hearth; b, the upper hearth; c, vault
upon which the hearth rests; d, vaults connected with the
lower hearth; e, openings communicating with the lower
hearth; f, arched roof of the lower hearth; g, arched roof of
the upper hearth; h, square openings between the upper
and lower hearths; k, working doors; l, openings for charging
* B. u. h. Ztg., 1852, p. 267.
+ Ibid., 1852, p. 304. DINGLER'S Polyt. Journ., cxx., 190.
IMPROVEMENTS IN ROASTING COPPER ORES.
141
the ore; m, a hollow, vertical, movable iron axis with two
arms to which rakes are fixed; n, fire-place; o, fire-bridge;
p, opening for communication between upper and lower
hearth; q, flue; r, chimney.
2. Longmaid carefully roasts pyritic copper ores with an
addition of common salt, at the same time admitting air
which has been dried by conducting it through burned lime.
The chlorine hereby evolved is used for the production of
chloride of lime.
3. Mitchell, Alderson, and Warrinert propose to carry
on the roasting process as carefully as possible. They control
the process by repeatedly testing samples by means of
lixiviation.
4. Sussex aims at the extraction of part of the sulphur
as sulphide of carbon, by an addition of coal, and partly by
an addition of alkaline salts. Hill endeavours to facilitate
the roasting by an admission of oxygen which has been
evolved from manganese, and so to burn the smoke more
completely. These methods, as well as the proposed appli-
cation of a galvanic current, are not likely to succeed.
5. Budd and Morgan§ recommend the use of hot air at the
roasting and smelting, and for this end they propose the use
of particularly constructed furnaces. Davy had before re-
commended the use of heated air in the roasting of copper
⚫ores. However, according to Gurlt,¶ there are great diffi-
culties in the way of this plan. Cumenge** employed steam
successfully when roasting antimonial and arsenical ores.
At Messrs. Vivians' smelting works the calcining++ furnace
represented by Figs. 37, 38, 39, and 40, is in use.
The furnace rests upon a vault, c, into which the ore is
raked down after being calcined; it is built of bricks, and
**
TUNNER'S Jahrb., 1852, p. 157.
† B. u. h. Ztg., 1852, p. 286.
Bgwkfd., xii., 525.
|| B. u. h. Ztg., 1852, p. 287.
§ Bgwkfd., x., 115.
¶ B. u. h. Ztg., 1852, pp. 289, 301.
Ibid., 1853, p. 33. Jahrbuch der k. k. Geol. Reichenstalt, 1854, p. 421.
†† Dr. URE's Dictionary of Arts, vol. i. p., 874.
142
COPPER.
B
FIG. 37.

I
E
F
E
B
A
C
FIG. 38.
bound by iron bars, as shown in the elevation, Fig. 37. The
hearth, B, B, Figs 38 and 39, is placed upon a level with the
lower horizontal binding bar, and has nearly the form of an
ellipse truncated at the two extremities of its greater axis.
FIG. 40.
If
FIG. 39.

B
В
He
A
The hearth is horizontal, bedded with fire bricks set edge
ways, so that it may be removed and repaired without
disturbing the arch upon which it reposes. Holes are left
in the sole before each door, c, c, through which the roasted
ore is allowed to fall into the subjacent vault. The hearth,
B, B, varies from 17 to 19 feet in length, and from 14 to 16
feet in breadth; the fire place, A, Fig. 39, is from 4 to 5
feet long, and 3 feet wide. The bridge or low wall, b, Fig.
40, which separates the fire-place from the hearth, is 2 feet
thick; it is hollow, as shown in the figure, and communi-
cates with the atmosphere at both ends, in order to conduct
a supply of fresh air to the hearth of the furnace. This
judicious contrivance is for the purpose of introducing a
IMPROVEMENTS IN ROASTING COPPER ORES.
143
continuous current of air upon the metal, in order to
facilitate its oxidation, and was originally invented by Mr.
Sheffield, who disposed of his patent right to Messrs. Vivian.
Of late years great attention has been paid to the roasting
of copper ores, as this operation is the chief cause of the
formation of injurious fumes, commonly called “ copper
smoke," emitted by copper works. The parts of the smoke
which are chiefly injurious to animal and vegetable life
consist, according to analysis, principally of sulphurous acid,
and only to a very small degree of sulphuric acid. By the
experiments of Mr. Turner it has been ascertained that an
amount of sulphurous acid, even as small as 1-10,000th part
of the atmosphere, proves fatal to vegetable life. Now
copper smoke is charged with about 1-10th of this substance,
and the neighbourhood of copper works in this country,
(chiefly in South Wales and Lancashire) show the highly
injurious effects which it produces. But besides being
injurious to the property which adjoins the works, the
copper smoke causes, at the same time, a considerable waste
of a valuable substance. About 20 years ago the actual loss
of sulphur, arising from the copper works in the immediate
neighbourhood of Swansea only, was estimated by the
French metallurgist, M. Le Play, at £200,000 a year at least.
Dr. Gurlt estimates this loss connected with British copper
smelting at £400,000 a year, as since Le Play's statement
the annual production of copper in Great Britain has vastly
increased. Mr. Peter Spence has put forth, however, a still
higher estimate of the sulphur in the copper smoke. The
quantity of copper ores smelted at Swansea is about 5,000
tons, and the proportion of sulphur in them averages from
24 to 28 per cent. "This," says Mr. Spence, "is equivalent
to 3,300 tons. of brown oil of vitriol, and this weight I would
undertake to
to produce therefrom. The present value.
(October, 1865,) of this weekly quantity is £9,900, which is
at the rate of £514,800, or more than half-a-million sterling,
per annum. Mr. Spence proceeds: "This quantity of acid
would meet the requirements of our staple chemical manu-
facturers, or nearly so."
The obnoxious parts of the smoke are generated in the
144
COPPER.
calcining and roasting furnaces, whilst the smoke derived
from the smelting and refining furnaces is comparatively
innocuous. It would, therefore, suffice for every practical
purpose if the smoke of the first-named furnaces were
subjected to a suitable process of condensation and utilisa-
tion. This fact has long ago been acknowledged, and to the
late Mr. Vivian, of Swansea, we are indebted for a series of
ingenious, though unsuccessful, experiments, which he pub-
lished in the " Philosophical Magazine."
Dr. Gurlt reduces the methods by which the separation
of the sulphur can be effected to three principles :—1. By
condensing the sulphurous acid as such. 2. By reducing it
to free sulphur. 3. By oxidising it to sulphuric acid.
Upon the first principle many experiments were based.
The sulphurous acid gas has been condensed by bringing it
into close contact with cold water in so-called rain chambers,
or other suitable apparatus. However perfect the absorp-
tion of the gas at these trials might have been, they laboured
under the serious inconvenience that the acidulated water
thus obtained cannot be utilised, and, if allowed to run
away, it poisons the streams and water-courses into which it
passes; whilst this method, on the other hand, does not pre-
vent a portion of the sulphurous acid from evaporating, and
thus reproducing the evil.
Concerning the second principle, Dr. Gurlt states the
following:-The second principle embraces the reduction of
the sulphurous acid to free sulphur, which is not only an
easily saleable substance but would always be manufactured
in preference to sulphuric acid, as it combines a far greater
value with a greater facility of transport, which, á priori,
are recommendations in themselves. The reduction of
sulphurous acid can be effected either through the agency
of sulphuretted hydrogen or of carbon. The first agent
decomposes, under favourable circumstances, the sulphurous.
acid in such a manner that SO₂+2HS form 3S+2HO, i.c.,
sulphur and water; while, through the second agent,
SO₂+C are formed into S and CO2, i.e., sulphur and
carbonic acid. In both instances sulphur in its native,
although very finely divided, state, known as flowers of
2
M. GURLT'S METHOD OF ROASTING ORES.
145
sulphur, is separated, the first requiring only a temperature
slightly above the boiling point of water, the second a good
red heat. For this reason reduction by means of sulphu-
retted hydrogen would be preferable, supposing this gas could
be produced at a very cheap cost and on a large scale. Up
to the present, however, all known methods for its produc-
tion are so expensive, including the cheapest mode by
decomposing carburetted hydrogen with vapours of sulphur
at a red heat, that the use of this agent for the separation
of sulphur from copper smoke must be abandoned for the
present.
There still remains the use of carbon, which being the
cheapest and most easily applied agent, promises by far the
best results, as it is only necessary to bring the vapours of
this acid in close contact with carbon, by passing it through
a layer of carbonaceous substances at a red heat, such
as anthracite, coal, or coke, and to condense and collect
the reduced vapours of sulphur in suitable apparatus. As
the copper smoke which is derived from the calcining and
roasting furnaces always contains free atmospheric oxygen,
which was found by Messrs. Faraday and Phillips, at the
Hafod Copper Works, to vary from 9 to 9 per cent (a
quantity sufficient to keep up combustion of fuel in a pro-
perly constructed fire-grate), this smoke carries in itself the
element for the production of the required red heat. The con-
ditions for the reduction will be fulfilled if the copper-smoke
is only made to pass through a sloping or step grate,
which is continuously provided with fuel of even inferior
quality, such as coke, coal, or anthracite slack, or even the
waste fuel from the smelting furnaces, the ashes being
removed in any convenient way. The gases, which after
passing the fire-grate, carry with them the vapours of the
reduced sulphur, will then have to be cooled by passing
through a system of iron pipes, which are conveniently
arranged and cooled with water, in order to condense the
sulphur to its solid state. Finally, these cooled gases have
to be forced by mechanical means, such as a fan, through a
so-called air filter, which may be a system of horizontal
and vertical bags or sacks made of permeable cloth, in which
VOL. II.
L
146
COPPER.
the flowers of sulphur are deposited, whilst the cool perma-
nent gases, freed from all injurious substances, escape in a
perfectly harmless state into the air. The flowers of sulphur
are from time to time removed from the collecting air filter,
and at once form a marketable product. From this sketch
of the process it will be seen that it operates upon the copper
smoke as it comes at present from the calcining and roasting
furnaces, without interfering in any way with the existing
arrangements of the works, and that it can entirely dispense
with tall chimneys, for the required draught may be easily
produced by mechanical means. If compared with the
process of sulphuric acid making, it will further prove far less
expensive in its first introduction and its permanent working,
as it only requires waste fuel to feed the grate with."
Dr. Gurlt has constructed and patented (1855) an
apparatus, which he expects will secure successful appli-
cation in any existing copper works. These apparatus, the
consumer, the cooler, and the collector, are of so simple and
inexpensive, but at the same time efficient, a construction, that
the adoption of the smoke reducing process promises to
secure a large benefit through the utilisation of the sulphur;
whilst the working expenses are limited to the consumption
of almost valueless fuel.
Mr. Peter Spençe tried for many months a process
similar in principle to that of Dr. Gurlt, and patented it
January 25th, 1855. In the specification he proposes bringing
sulphurous acid gas into contact with carbonaceous matters,
such as coke, charcoal, peat charcoal, or other deoxidising
substances, kept at a red heat, but not subjected to com-
bustion by admission of air, &c. As the result of many
trials, continued for a considerable period, Mr. Spence found
great practical difficulties in the conversion of the sulphurous
acid into sulphur by means of carbon; and in addition to
this, the sulphur produced was so inferior in purity and
appearance that it could never be brought into the market
in competition with Sicilian sulphur, but could only have.
been introduced in competition with pyrites for the use
of vitriol makers. Mr. Spence succeeded, indeed, in con-
verting about 20 per cent of the sulphurous acid, on which
MR. BELL'S METHOD OF ROASTING ORES.
147
he operated, into sulphur, but the colour was similar to
that of a compound of two parts lamp-black and one yellow
ochre.
We are not aware whether Dr. Gurlt's process avoids all
the practical difficulties experienced by Mr. Spence, or
whether his process has been practically applied in any
copper works, but it appears to be most ingenious and
applicable to other manufactures in which sulphurous acid is
evolved.
The third principle, namely the oxidisation of the sulphur-
ous acid is carried out on a large scale with great success by
the methods of M. Gerstenhöffer and Mr. Spence. Before
proceeding to them we will describe a method patented by
Messrs. Thomas Bell and Thomas L. G. Bell, on the 14th
of September, 1865.
Mr. Thomas Bell describes this method as follows:
The greatest part of the sulphurous acid gas is given off
from the calcining furnaces, that from the melting, roasting,
and refining furnaces being small compared with it. We
propose to deal with these separately, so as to obtain the
greater part of the sulphurous acid in as concentrated a form
as possible, and in the best state for condensation as sul-
phuric acid; for this purpose we use our calcining furnace.
This furnace consists of a drum or cylinder of cast-iron
lined with thin fire-bricks. Two of these are set under a
brick arch, and heated by one small fire below (after the
manner of gas retorts). These drums rest upon friction-
wheels, so as to be higher at one end than the other. At the
upper end is fixed a hopper with spindle to regulate the feed
of ore, and also the pipe for exit of the sulphurous acid.
The lower end passes through out of the fire arch into a
small brick chamber, into which the calcined ore falls.
These drums are made to revolve slowly (say) once in three
minutes. The ore is thus continually turned over, and travels
to the lower ends; allowing it to be twelve hours calcining,
one of these furnaces will do 100 tons per week. The power
required is equal to about 3-horse power. With regard to
the wear of the tubes or drums by heat, furnaces of some-
what similar description for re-burning animal charcoal are
L 2
148
COPPER.
worked at a higher temperature than is required for calcining
copper ores, and yet frequently last six years without repairs.
The advantages of this furnace are-
1. That it completely separates the sulphurous acid from
the coal smoke.
2. That from there being no opening of doors to rake or
stir, the sulphurous acid is given off uniformly, and mixed
with no more air than necessary.
3. That it is less expensive than the ordinary mode of cal-
cination, both in the first cost of the furnace and the working
expenses for coal and labour; the quantity of coal being
only something like one-third of what is consumed by other
calciners, and labour being dispensed with altogether.
4. That it is more effective, inasmuch as it will calcine.
every description of ore, no matter how large or how small
the quantity of sulphur in it, and the degree of calcination
can be exactly regulated.
In consequence of the continuous exposure of fresh surface
to the action of the air and heat, the ore may easily be cal-
cined perfectly "sweet," were it so required. But by regu-
lating the feed of fresh ore and the speed of revolution, it
may be brought out calcined exactly as much as required.
The sulphurous acid from this process would then be con-
ducted into sulphuric acid chambers, and condensed in the
ordinary way.
An important invention for the utilisation of the copper
smoke is the calcining furnace of M. Gerstenhöffer.*
The application of the sulphur disengaged in the calcina-
tion of copper ores, to the production of liquid sulphuric
acid, has long been an important object to copper smelters.
In the case of a small proportion of copper ores, in hard
lumps free from earthy matter, containing a comparatively
large proportion of sulphur, this is affected in the ordinary
pyrites burner of the sulphuric acid works. But for the most
part copper ores are small, poor in sulphur, and with much
earthy matter; they are therefore calcined in reverberatory
furnaces with fire-grates. The gases from these calciners do
* Dr. URE's Dictionary of Arts, &c., i., 488.
M. GERSTENHOFFER'S CALCINING FURNACE.
149
not consist of more than o˚5 per cent by volume of sul-
phurous acid, and are, moreover, largely mixed with the
products of combustion from the fire-place; they are there-
fore unfit for the vitriol chamber, and are allowed to escape.
Gerstenhöffer's furnace consists of a rectangular chamber,
filled with rows of bars, having one of their faces parallel
with the top of the furnace, and so arranged that no two
bars in succeeding rows shall be directly one over the other.
The ore is introduced from above, in a finely divided state,
and the supply is regulated by rotating feed-rollers fixed in
the roof of the furnace. These feed-rollers are placed over
the first row of bars, and as the ore falls through it piles up
on them, until it at last flows over and falls then on the row
of bars below, and the supply being thus kept evenly flowing
from above, gradually fills the whole furnace. Between the
feed-rollers and the bars are channels in the side of the
furnace, by which the products of the calcination may escape,
and besides these, there are other apertures which may be
opened as necessity requires for cleaning the bars. Below
the tiers of bars-and here it may be mentioned that the
higher the furnace the more perfect will be the calcination-
there is an open space into which, at the commencement of
the calcination, a grate with some ignited fuel (say) wood or
charcoal, is introduced; and this fire is well fed with fuel,
until the charge is thoroughly ignited, when the grate is
withdrawn, and the charge burns by means of the ignition
of the sulphur evolved. This combination is kept up by a
current of air forced from beneath up through the furnace,
and the products of combustion are thus carried over through
the set of channels in the top of the furnace away into the
sulphuric acid chambers. The channels by which the gases
are here conducted are so arranged as to heat the current of
cold air forced into the bottom of the furnace. The calcined
ore having passed through the furnace, falls into a space
below, from which it can be drawn off into waggons, and so
transferred to the next operation.
Figures 41 and 42 represent perpendicular transverse
sections of Gerstenhöffer's furnace. Fig. 41 is a cross sec-
tion parallel with the front. Fig. 42 represents a section
150
COPPER.
from front to back. The length of the earthen bars is about
2 feet 6 inches; twenty rows occupy about 12 feet in height.
FIG. 41.

α
a is an iron box or hopper for containing a supply of ground
ore, having at the bottom two or more cast-iron grooved
rollers, which are worked without interruption at a speed
adapted to supply ore equal to a discharge of about 10 cwts.
of sulphur per 24 hours. As the ore sometimes has a ten-
dency to form a cake on the bars, it is necessary to scrape
them occasionally by an iron rod with a curved end. This
rod is introduced through the plug holes in cast-iron boxes,
shown in Fig. 42. The calcined ore collects in the cavity, b,
M. GERSTENHOFFER'S CALCINING FUrnace.
151
at the bottom, and is raked out when convenient.
The
necessary amount of air is admitted partly at the back
α
C
FIG. 42.

e
e
e
through the flue, c, and regulated by means of a screw valve,
and partly through the plug holes in front. After passing
upwards through the calciner, the air, now charged with
sulphurous acid to the extent of 6 or 8 per cent in volume,
passes over the bridge and through the flue, e, in the direc-
tion of the arrows, Fig. 42. Provision is made at the bottom
of these flues for collecting and removing particles of roasted
ore carried away by the current. f is a brick chamber for a
152
COPPER.
similar purpose. From its back the gas passes through the
flue, g, to the vitriol chambers.
The principles advantages gained by the use of this inven-
tion, says M. Gerstenhöffer, are the following:-" First, the
ores are operated upon in a very divided state, so as to expose
a large surface to the action of the air, and this surface is
always being renewed; second, the air which must be admitted
is forced to come perfectly into contact with the ores; third,
the ores are brought into the furnace by mechanical means,
continuously in such quantities as may be required, so as to
maintain a continual roasting process; fourth, the air or
draught is also brought into the furnace by mechanical
means, in order to be able to regulate the draft according to
the quantities of ore, and also to prevent the influence of
storms, which are sometimes very troublesome in working
with grate furnaces; fifth, the heat produced by the ores is
not led away uselessly by the gases, but is employed to heat
the new ores and the air.”
The Gerstenhöffer furnace is applicable to all varieties of
sulphuretted copper ores, although varying in proportion of
sulphur from 16 to 40 per cent; it is also adapted to the
crude metal or regulus obtained as the result of the first
reducing operations.
For roasting these substances Gerstenhöffer's furnace has
advantageously been applied both on the Continent and in
this country-in the works of Messrs. Vivian and Sons, near
Swansea-though the process has the disadvantage of re-
quiring ores in a finely divided state, and the necessity for a
blast apparatus, the current from which must be carefully
regulated, besides involving constant attention to the ma-
chinery used.
Mr. Peter Spence's furnace stands pre-eminently forward
amongst other furnaces for accomplishing the same ends as
that of M. Gerstenhöffer.
Mr. Spence,* in his patent of the 3rd of July, 1861, thus
describes his arrangements :—
The essential feature of this invention consists in submitting
Dr. URE'S Dictionary of Arts, &c., i., p. 889.
MR. SPENCE'S CALCINING FURNACE.
153
such ores to the action of a roasting heat as they are passed
from one end of a furnace to the other, during which trans-
ference a current of air is caused to travel over them in an
opposite direction. To accomplish this a furnace of con-
siderable length, having several doors for the purpose of
introducing apparatus by which the transference may be
effected, is used. It will be obvious that by this arrangement
the ores may be submitted to heat in a thin stratum, that
the amount of roasting may be modified by a quicker or
slower transference, and that the heat communicated may
be greater towards the termination of the operation than at
the commencement thereof.
This invention will be fully understood by the accom-
panying drawings of apparatus. Fig. 43 is a side view of the
furnace, and Fig. 44 is a longitudinal section. In both draw-
ings the two ends are shown, the whole being of considerable
FIG. 43.

d
C
b
هم
FIG. 44.

d
g
d
C
C
E
b
રી
length. One now in operation is 50 feet long, with twelve
doors for the transfer of the ores; but these dimensions may
be varied, and we merely mention them for the purpose of
154
COPPER.
showing that the furnace should be of considerable length.
Fig. 45 represents a cross section, and in describing the
FIG. 45.

d
apparatus, in the first instance, we suppose that the manu-
facture of sulphuric acid is the object in view.
At a are the fire-bars of the furnace, and at b is the fire-
chamber, formed by channels, as seen in Fig. 45, extending
under a partition, c, of fire-brick, which partition forms the
bottom of another and distinct chamber, d, furnished at one or
both sides with a number of doors, c₁, e₂, &c. At ƒ is an aper-
ture formed through the brick-work, and constituting a com-
munication from the external atmosphere to the chamber, d.
The ores from which the sulphuric acid is to be obtained are
passed through the door, e,, into the chamber, d, in sufficient.
quantity to lie about 2 or 3 inches in thickness upon the
bed, c, and to extend, say halfway towards the door, c. This
batch having been submitted to heat, passing from the fire-
chamber, b, for the required time, is pushed forward by any
convenient instrument so as to be brought opposite the
door, e,, and another batch is introduced through the door, c,.
The first charge is then removed from e₂ to c3, and the second
from e̟, to e̟2, and it is then pushed through the aperture, f,
into any receptacle placed to receive it. During this transfer
the material has gradually become heated, a current of air
entering at ƒ has been passing over it, and the operation has
caused the sulphur to be driven off from the ore and con-
veyed through the channel, g, to the ordinary sulphuric acid
apparatus. The end of the fire-chamber, d, leads into a
flue at i. The degree of roasting, and the way to secure
the best results, can only be ascertained by experience,
but when working with a furnace 50 feet long with
twelve doors, the first charge is allowed to remain for one
I
2
MR. SPENCE'S ROASTING AND SMELTting process. 155
hour, then transferred to the second position, and a fresh
charge put into the first, and so on, waiting an hour between
each charge.
In this patent Mr. Spence specified, as one of the pecu-
liarities of his furnace, that the current of air forced over
the material under calcination should travel in a different
direction to that in which the ore is moved; but towards.
the close of the same year he took out another patent, in
which he claimed, as a special improvement, the use of a
current of air passing in the same direction as the material
under operation.
On the 14th of June, 1864, Mr. Peter Spence and Mr. J. B.
Spence obtained another patent "for calcining and smelting
copper ores," which will be fully understood by the follow-
ing description:-
"Our invention consists in applying the heat used in
smelting copper ores to the purpose of calcining such
materials, and in transferring the calcined ores direct to the
smelting furnace. To accomplish these objects we place
the two furnaces in connection with each other and cause a
suitable flue to convey the heat used in smelting to the
material to be calcined; and when this operation is com-
plete, and the smelted ore is removed, we rake the calcined
ore direct on to the bed of the smelting furnace. The
calcining furnace we at present prefer to use is of that con-
struction for which Letters Patent were granted to Peter
Spence, bearing date July 3rd, 1861, No. 1695."
Mr. Peter Spence gives (1863) the following particulars
concerning his roasting process:-
"The furnaces, in full operation, have now been open to
the inspection of copper smelters for over three years;
20,000 tons of ore, Irish, Cornish, and Spanish, have been
calcined in them; and 95 per cent of the sulphur evolved
converted into sulphuric acid. The furnaces present no
difficulty, are simple in construction and operation, require
less labour than the present mode of calcining, the ores are
all calcined as they come from the mines, and some of the
furnaces have acted for two years without repair. The
adaptation to all kinds of ores is one advantage. We calcine
156
COPPER.
stamped ores in fine powder, crushed ores (the general form
of Cornish ores), and rough ores in sizes up to that of an
egg-all receive similar treatment. We so attach the
calciner to the ordinary smelter that the draught, or waste
heat, after leaving the smelter at the ordinary orifice over
the rabbling-door, instead of going direct to the chimney, is
passed along under the bed of the calciner, and without in
the least affecting the draught of the smelter, yields ample
heat for calcining in the ore furnace. So arranged, 25
to 30 tons of ore per week are all that the attached
smelter can work. Now, in the ordinary calciner not only
is fuel largely required for calcination, but the ore has to be
drawn out and drenched with water before removing from
the calciner to the smelter; and in Gerstenhöffer's furnace,
although fuel is chiefly dispensed with, yet the ore must fall
down in a red-hot state, and will have to be removed and
drenched before using. All this we have made a thing of
the past. From our combined calciner the ore is at once
transferred by the workman with his slice down an easy
inclination, in a red-hot state, right into the middle of the
bed of the smelter, and is, of course, melted in one-
fourth the time taken by a drenched charge of ore.
Gerstenhöffer's furnace does not provide for this. But,
again, while copper ores pure in sulphur can at once be
safely heated to a high red-heat, such rich ores as the Cobre,
Spanish, or Portuguese would, if suddenly heated, inevitably
flux. Regulus, which ought all to be calcined, and its fumes
condensed, is of a similar character. Any sudden heating
of these would result in clotting. Now, our furnace
gradually heats the ores for some hours before reaching the
highest temperature, and clotting never takes place either
with rich ores or regulus. Thus, the ore charged into the
calciner is calcined without the expenditure of an atom of
fuel, one furnace calcining all required by one smelter, and
the raw ore never coming out till it comes out as regulus
and slag. This arrangement has been twelve months in
successful operation."
The cost of calcination by his furnace Mr. Spence states
to be only 2s. 1½d. per statute ton of ore, which is less than
REPORT ON MR. SPENCE'S PATENT CALCINER.
157
the cost of calcination by the furnaces at present in use,
whilst for every 5 tons of ore calcined in Mr. Spence's
furnace £9 worth of sulphuric acid is obtained, at a cost of
not more than £1, from constituents of the ore which the
ordinary furnaces turn to no account whatever. Whether
Mr. Gerstenhöffer's furnace will prove capable of yielding
better results than Mr. Spence has thus for the last three
years obtained from his, may fairly be doubted. Be that
as it may, we really seem at last to have reached the
beginning of the end of the enormous waste which has so
long been going on at Swansea of a substance which is quite
as important to the chemical arts as iron is to the
mechanical ones.
The advantages anticipated by Dr. Percy and Dr. Angus
Smith from this patent copper ore calciner may be seen
from the following report which they have made to Mr.
Spence:-
"In the copper works of Swansea and other parts of
South Wales, the ores chiefly used contain a large amount
of sulphur, and are obtained either as dust or in small
fragments. Before smelting them, a large portion of this
sulphur is burnt off in reverberatory furnaces, called
calciners, the smoke of which must have an exit into the
atmosphere, and as the sulphurous fumes are mixed with it,
both escape together.
"The ores of copper used are, in fact, ores of sulphur
also. In cases such as these-viz., when the pounded
copper ore is deprived of its superfluous sulphur by roasting
in a reverberatory furnace-there is no manufacture of sul-
phuric acid.
"It has been estimated that £200,000 worth of sulphur is
sent yearly into the atmosphere, in the state of sulphurous
acid, in South Wales.
"This sulphur it is now proposed to use for the manufac-
ture of sulphuric acid; it may be considered, therefore, as
obtained free of cost, whilst the expense of manufacture
remains the same.
"The copper ore is bought at a price which is regulated
by the price of copper. As a large proportion of the sulphur
158
COPPER.
is burned off in the process of calcination, this amount can
only be viewed as a source of trouble and expense.
"The process of burning the sulphur for sulphuric acid.
will stand in the place of the preliminary process of calci-
nation adopted by copper smelters; one, therefore, of the
numerous operations of the copper works which has
hitherto been a source of loss becomes a source of profit.
"We find that Mr. Spence is able to reduce the sulphur
of the ore to any required extent by regulating the speed of
passage through the furnace; for his present purpose the
sulphur is reduced to five per cent.
"There can be no doubt of the ready combustion of
sulphur ores by this method, Mr. Spence employing no other
kind of furnaces for his large works at Manchester and at
Goole, nor can we imagine a reason for allowing the loss of
so much sulphur as now escapes into the atmosphere, when
it can be so readily collected in the form of a product so
valuable as sulphuric acid.
"The copper smelter dissipates in the atmosphere, at
considerable cost to himself, that which the soda maker
incurs considerable expense to obtain, namely, sulphurous
acid. It is therefore a very important question whether the
sulphurous acid evolved in the process of copper smelting
may not be in a great measure collected and profitably.
applied to the manufacture of sulphuric acid.
"We have during many years maturely considered the
subject, and from our own observations in various copper
works, from the opinions which some copper smelters of
great experience have expressed to us, and from our recent
examination of the working of the furnaces at Mr. Spence's
Alum Works for the production of sulphuric acid by the
calcination or roasting of cupriferous iron pyrites, we believe
ourselves justified in answering this question in the affirma-
tive. We believe that not only may a large proportion of
the sulphurous acid of the copper smelter be collected and
applied to the manufacture of sulphuric acid, without in
any way interfering with the usual system of copper smelting
at Swansea and elsewhere in this country, but that such
REPORT ON MR. SPENCE'S PATENT CALCINER.
159
collection and application would be a source of considerable
gain. All that is required to effect this important object is
to replace the ordinary copper calcining furnaces by furnaces
similar to those which we have inspected at Mr. Spence's
works.
"We have watched the operation of Mr. Spence's
furnaces with great care, and in our judgment nothing
could be more satisfactory or better adapted to copper
works. Furnaces of this construction have not to our
knowledge been previously employed for the production of
sulphurous acid by the roasting of sulphuretted ores, either
of iron or copper. They appear to us to have a decided
advantage over the ordinary copper calciners, irrespective
even of the collection of the sulphurous acid. Their action
is unintermittent, and the ore, as it passes down from one
end to the other, is exposed to a gradually increasing
temperature, whereby clotting is entirely prevented, and the
condition most favourable to calcination is realised.
"The advantages, therefore, which may reasonably be
anticipated from the combination of copper smelting and
soda making in suitable localities are :—
1. A large proportion of that which costs the copper
smelter money to throw away may become a source of profit.
"2. That which it costs the soda maker money to produce
may virtually be obtained for nothing.
3. In addition to the usual profits from copper smelting
and soda making, a considerable increase may be fairly
expected.
"4. The nuisance of copper smelting chiefly arises from
sulphurous acid evolved during the process of calcination,
especially of the ore, and the whole of this may be sup-
pressed.
"5. The outlay required does not exceed what would be
necessary for the erection of ordinary copper smelting and
soda works independently.
"6. By the utilisation of the sulphurous acid of the
copper smoke no change is necessitated either in the usual
system of copper smelting or soda making."
160
COPPER.
Production of Raw Matt (Regulus, Coarse Metal) from
Roasted and Raw Ores.-This operation purposes to
separate the copper from the earthy gangue, and to extract
part of the oxides of foreign metals contained in the roasted
ores, by a purifying and reducing smelting. In this the
sulphur acts the principal part, as the sulphides which remain
unaffected by the roasting process decompose oxides and
sulphates when fused at very high temperatures. At a certain
high temperature peroxide and sulphide of iron next react
upon each other, forming sulphurous acid and protoxide of
iron, which becomes scorified. At a still higher temperature
the oxide of copper contained in the roasting mass is acted
upon by the sulphides of iron and copper, forming peroxide
of iron and metallic copper. Part of the metallic copper is
dissolved in the formed matt, and part converted into sub-
oxide of copper by the peroxide of iron. The suboxide of
copper and protoxide of iron then combine at the highest
heat of the furnace with silica, which is either contained in
the ore or added to it. As, however, the matt permeates the
slags in the form of very small beads, if the smelting hearth
is constructed nearly horizontal, the suboxide of copper in
the slag becomes decomposed by the sulphide of iron in the
matt, forming sulphide of copper and silicate of iron; there-
fore but little copper will be scorified so long as sulphide of
iron is present in a well fused mass :—
3Cu₂0 + 3FeS + SiO3 3FeO, SiO3 + 3Cu₂S.
Hence it is obvious that a sufficient amount of sulphur must
remain in the roasting mass. If the roasting process is
carried on too far, raw ores may be added to the raw smelting.
If the roasting is not carried on enough a matt poor in
copper is produced. The formation of deposits of iron and
of black copper at the raw smelting is prevented by the pre-
sence of sulphur. If the mixture is siliceous or aluminous,
and difficult to fuse, additions of fluor spar and rich copper
slags may be made.
The chief aim of this smelting is, as we have before men-
tioned, to collect in a matt all the copper contained in the
ore, and to produce slags so poor in copper that they may
be thrown aside. As regards this point the process is not
PRODUCTION OF RAW MATT (REGULUS).
161
quite successful, as the resulting slags are usually richer in
copper than those from the smelting in cupola furnaces;
great care must therefore be taken to counteract the defi-
ciencies of the process by carrying it out in as perfect a
manner as possible.
Le Play infers from his investigations that the copper
contained in the slag is only mechanically mixed up matt,
and not a silicate in an oxidised state, as the sulphosilicate
of iron which is always present in the slags, immediately
sulphurises the oxidised copper present.
Percy is of opinion that the supposition of a sulpho-silicate
of iron is not necessary to an explanation of the chemical
reactions at the raw smelting, but such a combination does
exist, for instance in Helvin, and some iron slags.
Antimony and arsenic are partly volatilised by the roasting
process and partly by the smelting process, as the anti-
moniates and arseniates which have been formed in the
roasting are decomposed by silica. A great part of these
substances always enters the matt.
The fluor spar, when acted upon by sulphuric acid, seems
to form volatile fluoride of arsenic, and also volatile fluoride
of copper, thus causing a loss of copper.
Tin contained in the ores partly enters the matt and
partly the slags.
In order to prevent loss of copper, the production of black
copper in the ore smelting must be avoided, except when the
ores are very impure and a purification of the matt is in-
tended. The presence of zinc blende in the ore gives rise to
the formation of matt and slag, both very difficult to fuse.
Galena assists the purification of the copper from antimony
and arsenic.
The furnaces used for the smelting process differ from
those employed for roasting the ores chiefly by having
a smaller hearth and a larger fire-place. The hearth is
usually oval, seldom oblong or square, like that shown in
Figs. 46 and 47, in which the furnace is supposed to
have the following dimensions:-Length of the hearth,
Dr. URE'S Dictionary of Arts, vol. i., 875.
VOL. II.
*
M
162
COPPER.
Ι
II or II feet; breadth, 8 or 9 feet; fire-place from 3 to
4 feet long, and 3 to 3 feet broad.
FIGS. 46 and 47.

G.
A
B
B
C
IM
According to Percy, the hearths of the new furnace are
13 feet long and 9 feet broad at their widest part; they are
oval and truncated on both sides, so that they are 5 feet
6 inches broad on the fire-bridge, and about I foot on the
opposite side. The grate of these new furnaces measures
3 feet by 4.
The smelting furnaces usually have three openings; one
to the fire-place at D, a second one, o, in the side, generally
kept shut, and used only when incrustations are scraped off
the hearth, or when the furnace is entered for repairs, and
the third or working door, G, placed at the front of the fur-
nace beneath the chimney; through it the slags are raked
out, the melted matters stirred, puddled, &c.
The hearth is bedded with infusible sand, and slopes
slightly towards the side door to facilitate the discharge of
the metal. Above the door is a hole in the wall of the chim-
ney (Fig. 47) for letting the metal escape. An iron gutter
leads into a pit, K, with an iron receiving pot at the bottom,
which may be lifted out by a crane. The pit, M, is filled
with water, and the metal becomes granulated as it falls into
the receiver. These melting furnaces are surmounted by a
hopper, L, as shown in Fig. 46.
Melting furnaces are also sometimes used for calcination;
PRODUCTION OF RAW MATT (REGULUS).
163
some of the furnaces near Swansea serve this double purpose.
They are composed of three floors (Fig. 48). The floor, a,
FIG. 48.

C
B d
a
is for melting the calcined ore; the other two, B, C, serve for
the calcination. The heat being less powerful in the upper
sole, c, the ore dries upon it, and begins to be calcined-a
process which is completed on the next floor. The square
holes, d, which are left in the hearths, B and C, put them in
communication with each other and with the lower one.
Those perforations are shut during the operation by a move-
able plate of iron.
The hearths, B and c, are made of bricks; the bricks are
horizontal on the top and slightly vaulted beneath. They
are two bricks thick, and are larger than the inferior hearth,
as they extend above the fire-place. On the floors destined.
for calcination the furnace has two doors on one of its
sides. On the lower story there are also two, but they are
differently placed. The first, being in the front of the fur-
nace, is used for drawing off the scoriæ, working the metal,
&c., and the second, at the side, admits the workmen to
make repairs. Below this door is placed the discharge or
tap-hole, which communicates by a cast-iron gutter with a
pit filled with water. The length and breadth of this fur-
nace are nearly the same as those of the melting furnaces
above described. It is charged by either one or two hoppers.
The inner walling is formed of refractory bricks, the arched
roof usually of quartz bricks or dinas bricks.*
The smelting process comprises the following operations:-
The formation of the hearth, by smelting, either of refractory
* B. u. h. Ztg., 1862, p. 116.
M 2
164
COPPER.
sand, or an artificial mixture of sand and very siliceous.
slag. For this purpose a layer of sand 1 foot thick is placed
on the hearth, beaten down, and strongly heated, then some
matt slag is spread over it and fused. Upon this is placed
a second layer of sand some few inches thick and some slag
is spread over it. A third layer is sometimes added in the same
manner, so that the thickness of the hearth after fritting is
frequently as much as 20 inches.
The durability of the hearth depends chiefly on the state
of aggregation of the smelting mass, whether it is treated
in the form of fragments or of schlich. When heating ores
in fragments the easily fusible sulphides drop out first
upon the sole of the hearth, coating it without exciting a
destructive action. When treating ores in a pulverised form
the fused sulphides cannot drop upon the sole, which is
exposed to the corroding influence of the oxides, metallic
copper, &c., with which it comes into contact. At the
treatment of the latter the superficial smelting proceeds but
slowly, as the air enclosed between the particles of schlich
is a bad conductor of heat, preventing it from entering the
lower portion of the ore.* When this is the case some fuel
may usually be saved by employing a gradually increasing
temperature instead of a sudden high temperature. The
furnace hearths often last for 12 months, frequently absorb-
ing a considerable quantity of copper. Their removal is.
often very difficult.t
The ores and fluxes are now charged by the funnel; they
are spread uniformly upon the hearth, slags being thrown in
as required by the working door, which is afterwards closed,
whilst, by a continued firing, the masses are fused. The
raking of the mass commences when the first reaction of
the components of the mixture takes place; this reaction
may be observed by the cessation of the frothing and of
caked substances swimming on the surface of the bath.
This raking is intended to separate the fritted quartz
particles and to dissolve them in the slag, and also to
* Oesterr. Ztschr., 1859. p. 309.
+ B. u. h. Ztg., 1859, No. 11.
PRODUCTS OF THE RAW SMELTING PROCESS.
165
liberate the matt grains that they may sink to the bottom.
After the raking the working door is again closed and the
temperature raised to the highest degree.
In a short time the matt is tapped off and conducted in a
thin jet into a reservoir with a stream of water passing
through it. The matt is thus converted into grains of the
size of a lentil, and suitable for roasting. In some copper
works in Wales the matt is first made to run into iron
moulds and granulated, after re-melting the blocks, which
yield a better copper (Hafod), or the matt flows into a
mould of sand and is stamped. Previous to this tapping
off the slag is removed by iron rakes on layers of sand
through the working door, and separated according to the
quantity of particles of matt. The matt is usually tapped
off after two smeltings, as the quantity obtained by one
smelting is too small.
The chief products are-
1. Raw Matt (regulus, coarse metal), brittle, crystalline
in its fracture, uneven, more or less granular, frequently
blistered and bronze-coloured. It must not contain more
than 35 per cent of copper at the most or the resulting
slags will be too rich and the grains obtained too coarse.
Matt in coarse grains roasts badly either in reverberating
furnaces or in kilns. When using kilns very little air must
be admitted, or the cooling will be too strong.
Matt must have about the composition of copper pyrites.
Its highest and lowest per centage of metal was found by
Napier to be:-
I.
II.
Cu.
Fe.
S
21'I 39'5
33°2 36'4
45'5
25'0
According to Percy, the formula of raw matt as given by
Le Play,-
3Cu₂S + Fe₂S¸ + 4FeS,
is not always correct. Matt containing 337 per cent of
copper has a specific gravity af 4'56.
2. Raw Slag* (ore furnace slag); it is brittle, compact,
* V. LEONHARDT, Hüttenerzeugnisse, 1858, p. 12.
166
COPPER.
dark, and frequently like porphyry, containing angular frag-
ments of quartz. Intermixed particles of matt are found
chiefly on the lower side of the pieces of slag, which are
therefore broken, and the matt separated as much as
possible. Pure slag must not contain more than per cent
2
of copper. The nature of the slag (chiefly mixtures of bi-
and mono-silicates) exerts an essential influence upon the
process; if the slag is too fluid it mixes easily with matt
during its removal, and also leaves deposits on the hearth,
particularly if the mixture is zinciferous; if too pasty the
separation of the slag from the matt will be more imperfect.
Generally the formation of a pasty slag is preferred as
causing less loss of copper.
The following modifications of this process have been
proposed and partially tested.
Schneider, at Swansea, adds sulphate of soda and
powdered coal to the smelting mass just before the matt is
tapped off, thus forming double sulpho-salts of arsenic,
antimony, and sodium, which dissolve when the matt is
made to run into water; in that way arsenic and antimony
are removed.
At the Spitty copper works, in Loughor, near Swansea,
and other English works, Napier'st process for treating
antimonial, arsenical, and stanniferous ores, has been
adopted. The sulphuretted ores are roasted for 9 hours,
and afterwards melted in admixture with 120 lbs. of sulphate
of soda, 40 lbs. of slaked lime, and 60 lbs. of coal, in order
to form soluble sulphides of antimony, arsenic, tin; and
sodium. After extracting these salts by means of water, the
remaining matt is smelted for the production of black.
copper; Percy states the effect of these fluxes to be
imperfect.
The slags resulting from the ore smelting, after Napier's
addition of lime and common salt, are composed as follows:-
* TUNNER'S Jahrb., 1852, p. 157.
+ Bgwkfd., xi., 584; xii., 270. B. u. h. Ztg., 1852, p. 285.
+
PERCY, Metallurgy, i., 382. Mining and Smelting Magazine, 1862,
No. 2, p. 124.
DIFFERENT METHODS OF RAW SMELTING.
167
I.
II.
III.
IV.
SiO.
NaO
CaO
49°26 49*60 42°20 35.60
7'93 0'70 3'44
7.84 I'23 643 448
MgO
2.62 O'II
2*14
0*16
A1,0₁
12.37
14'00 10.80
6.85
FeO
18.60
3100
50 46
Fe₂O₁
32'94
CuO
0'70 I'06 0'45 2*32
NaCl
•
0*48 0°43
0'34
MnO
trace trace
S
trace
1'43
Loss
0'20
0'21
0'13
Residue
1'56
Nos. 1-3 are slags produced in the copper smelting
furnaces of the South American and Mexican Company in
Chili, analysed by Field. Nos. 1 and 2 are pellucid and
uniform, being produced by an admixture of 20 per cent of
common salt; on cooling, they form two distinct schists;
the upper being amorphous (1), and the lower one of
crystalline structure (2). No. 3 is produced by an addition
of 10 per cent of common salt. No. 4 is a slag from
Coquimbo, by Field.
Oxidised ores are treated more easily by Napier's method
than sulphuretted ores.
Rivot and Phillips have tried to produce pure copper
direct from roasted ores by means of the electric current in
one operation. This method may perhaps be practicable
when treating very rich ores.
Beudant and Benoit'st patent aims at the extraction of
antimony and arsenic from raw matt by an addition of galena
and iron.
Low adds to the smelting a mixture of brownstone,
graphite, saltpetre, and carbon, with the object of shortening
the process and lessening the scorification of copper. This
method is employed at Low's copper works in Penclawdd;
Ann. d. min., 4 sér., tom. xiii., 251; livr. i. de 1848. Bgwkfd., xi., 696;
xii., 406; i., 266, 539. B. u. h. Ztg., 1850, p. 593; 1852, pp. 268, 303. PERCY,
Metallurgy, i., 385. DINGLER'S Polyt. Journ., xv., 192; lxix., 265.
† PERCY, Metallurgy, i., 372.
+
† Bgwkfd., xii., 669. B. u. h. Ztg., 1860, p. 62.
168
COPPER.
according to Tunner* the ores are previously submitted to
a treatment with mineral waters.
The Roasting of the Raw Matt is effected in the same
way as the roasting of the ores, excepting that a higher
temperature is employed towards the end of the process.
After the matt has been pounded or rolled it is roasted either
in a granulated form or by Trueman's method ;† if the
roasting process is well conducted the material keeps its
original form. The degree of roasting depends on the
quantity of pure ores which are to be mixed in the subsequent
smelting.
According to Napier, the composition of the matt before
(a) and after (b) roasting was as follows:-
Cu.
Fe
S
O
Insoluble.
a.
b.
32
32
36
36
25
13
II
7
7
The Roasted Matt for the production of different con-
centrated matts is effected in an ore smelting furnace with
a hearth consisting of two layers of sand, the lower one
9 inches, and the upper one 3 or 4 inches thick; the lower
layer seldom requires repairing.
Concentrated matt may be produced, varying in compo-
sition (blue matt, white matt, pimple metal) according to
whether more or less oxidised ores and products (rich
slags) are available, the purity of the ores thus allowing
a quicker or slower production of metallic copper; this
yield of matt of different composition may then be obtained
by an addition of the named oxidised substances, or by a
more or less strong roasting of the raw matt. The matt
may then be purified by a repeated concentration, at which
more or less copper will be extracted.
According to Field,‡ raw matt consists chiefly of—
3Cu₂S+Fe₂S3+ FeS+2Fe₂S (analysis a),
TUNNER'S Jahrb., 1852, p. 157.
† Allgem., B. u. h. Ztg., 1859, p. 180.
Mining and Smelting Magazine, 1862, vol. i., p. 324. B. u. h. Ztg., 1Ɛ62.
P. 343.
PRODUCTION OF BLUE METAL.
169
and may be gradually converted by a further concentration
into-
6Cu₂S+Fe₂S₂+ FeS+2Fe₂S (analysis b), and
12Cu₂S+Fe₂S₂+FeS+2Fe₂S (analysis c);
afterwards into Cu₂S, and finally into metallic copper—
2CuO+Cu₂S=4Cu+SO2.
The richer the matts are in Cu₂S, the less they are attacked
by muriatic acid.
The following analyses confirm the compositions stated
by Field :-
Cu.
a.
b.
C.
36.12 49'71 61*34
Fe.
S
36.78 25°35 15.61
27.08 24.85 22'90
Production of Blue Metal, Cu₂S+yFeS+Cu.—When
treating impure ores which require a repeated concentration
of matt, or if the best selected copper is to be produced
from pure ores, no addition of oxidised flux is made to
the smelting of the impure raw matt, or only a moderately
rich addition of the sulphuretted ores which have not
been strongly roasted. When treating very pure ores the
roasting of the raw matt must not be carried on too far,
or the sulphide of iron in the matt cannot be completely
decomposed on account of the absence of oxidised com-
pounds.
The resulting matt is poor in copper, and contains in its
numerous blisters metallic copper, the formation of which
is explained by Plattner and by Le Play, but in a more pro-
bable way by the former; experiments made by Percy have
also confirmed Plattner's views. Plattner's explanation† is
as follows:-
Rich copper matt consists of
Moderately rich matt consists of.
Poor matt consists of
or of
PERCY, Metallurgy, i., 352.
+ B. u. h. Ztg., 1855, p. 303.
"Cu,S,FeS,
Cu,S,FeS, and
n(Cu₂S, Fe₂S) FeS,
(Cu₂S, FeS) FeS.
PLATTNER'S Röstprocesse, p. 210.
170
COPPER.
On slowly cooling, the bluish black copper matt, µCu₂S,FeS,
liberates a small portion of its copper, probably because
during the smelting of the copper matt the sulphide of iron
gives part of its sulphur to the copper, transforming it into
Cu₂S, while the FeS itself is converted into Fe₂S, and thus
a combination of Cu₂S, Fe₂S results, which does not become
modified when in a fused state at a certain temperature,
or when rapidly cooled. If, however, a gradual cooling
takes place, the Fe₂S is re-transformed into FeS, at the
expense of the Cu₂S, owing to the respective affinity of the
Cu₂S and FeS. The reconstituted combination, Cu,S,FeS
goes to the outside, while in the cavities the liberated copper
deposits in a capillary form.
According to Le Play, this copper from a blue copper matt
(blue metal) of Wales, contained-
Cu.
98.2
Fe.
0'4
Ni. Sand and Carbon.
0'6
0°2
As the raw matt is more impure, the production of a blue
metal poor in copper should be attempted.
After being more or less perfectly roasted the blue metal
yields by the roasting smelting, or when smelted with an
addition of oxidised products, a more or less pure white metal
with or without any separation of metallic copper, and when
re-smelted yields a regulus matt.
The blue metal (xCu₂S+yFeS+zCu) may thus be trans-
formed into white metal (Cu₂S), while noxious admixtures are
gradually extracted :—
(xCu₂S+2FeS+2Cu) +2CuO+xSiO, = xCu₂S+2Cu₂S+
2 FeO,SiO2, or
3
(xCu₂S+3FeS)+3Cu₂O,xSiO¸=xCu₂S+3Cu₂S+3FeO,xSiO¸
The silica required for the scorification of the iron is taken
from the sand adhering to the pieces of matt and to the
bricks and the hearths of the furnace.
The blue metal slags are poorer in oxidised copper if the
matt is richer in iron. They are added either to the ore or to
the matt smelting.
Production of White Metal.—CuS, containing upwards
of 75 per cent of copper, and usually containing from 65 to
70 per cent of copper and 21 or 22 per cent of sulphur.
SMELTING THE SLAGS WITH PYRITIC ORES.
171
It consists chiefly of Cu₂S, without a large amount of FeS;
it is compact, brittle, uneven in fracture, of granular or
crystalline structure; it is bluish-grey in colour, and has
little metallic lustre. It may be produced direct in larger
quantities from purer ores with an addition of oxidised ore or
products, if common sorts of copper only are to be produced.
When heating this carefully formed admixture, the unde-
composed sulphides contained in raw matt smelt next, and
run to the furnace sole, protecting it from the reaction of the
oxides. At an increased temperature, the copper of the
oxidised ores and products becomes sulphuretted by the re-
action of the sulphide of iron and the silica present, whilst
the oxidised iron becomes scorified:
Cu₂S+ FeS+xSiO3 = Cu₂S+ FeO,SiO3
2CuO+2FeS+xSiO¸=Cu₂S+S+2FeO,xSiO3.
The white metal slag is similar to that of the blue metal,
and contains about 1.83 per cent of oxidised copper; it is
added to the ore and matt smeltings.
Production of Pimple Metal.—In order to produce
copper as quickly as possible from very pure raw matt, the
matt is melted with a surplus of oxidised ores and products;
the consequence is the formation of a matt similar to the
white metal in composition, colour, and fracture, with many
cavities and blebby knobs, containing greater quantities of
separated metallic copper, frequently in crystals an inch
long; hence the white metal stands between the blue and
pimple metal.
If the matt is very rich in separated copper
it is called close regulus. The pimple metal may be smelted
either direct for the production of black copper, or be con-
centrated to a second pimple metal.
Smelting the Slags with Pyritic Ores for the Pro-
duction of White and Red Matt, which are produced in
the production of blue metal slags poor in copper; black
copper slags containing 10 or 12 per cent of suboxide of
copper, are sometimes submitted to a smelting with pyritic
ores for the production of white or red matt.
The process is carried out by putting, first non-caking coal
upon the furnace hearth, then pure pyritic ores, and upon
these the slags. The mass is then fused while keeping the
172
COPPER.
furnace doors closed, and the fused mass is made to run into
sand moulds.
By the reaction of the suboxide of copper upon the pyrites.
a cupriferous matt is formed, and with it combines the matt
which had been mixed with the slags. A considerable
portion of metallic tin and copper is reduced by the contact
of the suboxide of copper with coal. These reduced metals
flow down to the hearth sole, combining on the way with
the impurities (arsenic, nickel, cobalt, &c.) of the matt; a
very pure white metal is consequently formed, from which, by
further treatment, the best selected copper is produced. If
the mixture contains much sulphur, a red matt results similar
to the blue metal, and usually differing from it only by the
absence of metallic copper, a more compact fracture, and a
larger amount of copper.
Beneath the red matt two distinct layers are found; the
lower one consists of a very impure coarse copper (bottoms),
and the upper one (hard matt) of a brittle alloy of copper
and tin, which is sold as it is. The matt attains its good
quality chiefly by the extraction of the tin. Plattner has
suggested a method of extracting silver and copper from
such products in the wet way.
Roasting Concentrated Matt for the Production of
Black Copper (Coarse Copper, Blister Copper).-The
concentrated matts obtained by the preceding operations are
heated without fluxes in an ore smelting furnace with a fire-
bridge furnished, like the roasting furnaces, with channels for
admitting air. The charging is not effected through a funnel
but by a side door. This process aims at the evolution of
the sulphur as sulphurous acid, and the separation of the
foreign substances by volatilisation or by scorification.
In the first period of the slow drop-like melting at the
admission of air, a slag is formed of the foreign metallic
oxides; this slag is several times removed, and the suboxide
of copper formed reacts already to a certain degree upon the
sulphide of copper present.
The temperature is then lowered in the second period,
while air is also admitted, and a solidified crust rich in
* B. u. h. Ztg., 1853, p. 614.
PRODUCTION OF COARSE COPPER.
173
suboxide of copper is formed superficially. When re-heating
the mass in the third period the oxides react energetically
upon the sulphides, thus yielding a blistered, coarse, or black
copper, while sulphurous acid is evolved. This copper still
contains sulphur even if the preceding operations of cooling
and re-heating have been several times repeated, and the
sulphur is only completely separated by the refining. Before
finishing the process, the air channels are closed, a strong
heat is given, and the copper is tapped off into sand moulds
after the removal of the slags.
According to Parkes and Keatest, the hot blast accele-
rates the process and facilitates the formation of purer pro-
ducts.
It is very necessary to the success of the process that the
smelting of the matt should be effected gradually, otherwise
the desulphuration by the atmospheric air will be but slight.
When blue and white metal are treated they are next gra-
dually converted into pimple metal, and afterwards into
coarse copper.
The black copper slag (roaster slag) is blebby, without
metallic lustre, reddish brown, or greyish black, with grains
of metallic copper. According to Le Play, it contains 20 per
cent, and according to Percy, 44 per cent, of copper as sub-
oxide, and o'85 per cent in a metallic state. As it is either
unfused or only partially fused it may contain larger quan-
tities of sulphides as well as the oxides.
If the production of the best selected coppert is intended,
the purest ores are employed, and the roasting process,
which in this case is called the selecting process, is conducted
in such a manner that from the white metal a portion of coarse
copper becomes reduced as bottoms, which then absorbs a
great part of the foreign impurities. A pure matt called
best regulus, spongy metal, then results; if necessary, this
matt is submitted a second time to the selecting process.
The best regulus is then converted into coarse copper. The
reason the greater part of the foreign metals concentrate in
* B. u. h. Ztg., 1852, p. 340.
Berggeist, 1857. p. 632.
+
PERCY, Metallurgy, i., 36.4.
174
COPPER.
the bottoms is that the separated copper decomposes the
sulphides of tin, antimony, arsenic, &c., contained in the
white metal, and these substances then combine with the
surplus copper, rendering it very impure, whilst the matt
resulting at the same time is purer. When refining these
copper bottoms an addition of lead- or copper scale is made.
When refining gun metal* the tin is extracted by an addition
of common salt.
According to Le Play (a) and Napier (b) these copper bot-
toms have the following composition :-
Cu.
Sn. Sb. As. Pb.
S. Fe. Ni. Mn.
a. 92.5 0'2
0'4
1.6
b. 740 13.8 45
4.8
0.8 3.9 2.5
Vivian, Herrmann, and Morgant have suggested a method
of extracting copper, silver, and gold from these bottoms.
Refining the Coarse Copper.‡
This process is performed in reverberatory furnaces, dif-
fering from the furnaces used for ore and matt smelting in
the following points chiefly :-They are usually smaller,
having no opening in the arched roof, and no tapping hole,
as the charging of the copper is effected by a side door, and
as the working on the hearth, particularly the ladling of the
copper, is effected by the working door situated opposite to
the fire-place. The hearth close to the working door is
furnished with a sump towards which all other parts of the
hearth are inclined.
The sole of the furnace is carefully prepared of quartz
sand, upon which the coarse copper is melted with the ad-
mission of air. The more slowly this melting is effected the
more impure will be the copper; hot blast accelerates the
process. The foreign admixtures are then scorified as much
as possible by increased blast, and removed as slag. As soon
as the slag has obtained a certain fusibility and a reddish
* B. u. h. Ztg., 1859, p. 261.
† Ibid., 1857, p. 180.
+
Grützner die Augustin sehe Silberextraction, 1851, p. 99.
|| B. u. h. Ztg., 1852, p. 356.
REFINING THE COARSE COPPER.
175
colour from the presence of suboxide of copper, nearly all
the iron and zinc will be scorified, and the greater part of
the lead extracted, whilst part of the antimony, nickel,
cobalt, and sulphur remain in the copper.
In order to remove these substances a slag rich in sub-
oxide of copper is formed upon the metal-bath by an admis-
sion of air, and at a certain temperature the oxygen of the
slag reacts upon those substances. Cobalt, nickel, and tin
are thus scorified, whilst the greater part of the antimony
and sulphur is volatilised.
In some copper works the copper is allowed to solidify at
the end of this period, and it is then rapidly fused by raising
the temperature to a white heat; in this manner the suboxide.
of copper extracts the last traces of impurities.
The copper then has the qualities of the rosette copper as
produced in small hearths or in reverberatory furnaces. To
toughen it the metal is stirred with green wooden poles.
The dry copper, as the metal containing a large quantity
of suboxide in solution is called, is first covered with a layer
of anthracite or charcoal for the purpose of preventing the
absorption of more oxygen.
The suboxide that remains dissolved is now reduced by
thrusting into the liquid mass a large pole of green birch or
oak, which is decomposed by the great heat. Carbonaceous
gases are evolved, which escape through the metal and act
strongly reducing themselves, and by the commotion pro-
duced cause every part to be brought rapidly and thoroughly
in contact with the anthracite or charcoal on the surface;
in this way all the suboxide is reduced to metallic copper,
whilst its gaseous portion escapes in combination with the
carbon into the air. This action is continued for from a quarter
to half an hour, during which time samples are taken at
short intervals and tried under a hammer or in a vice to de-
termine their malleability. When the metal has reached
what is termed the proper pitch the assay bends readily
without showing any cracks at the edges, and its fracture
should have a fibrous silky lustre, and be of a light colour.
When this pitch is attained, the pole is withdrawn and a
large sample termed a test is taken out. This test is cast
176
COPPER.
in a small mould, and beaten with sledge hammers into a
plate. If the copper is good this plate does not crack at the
edges, and the surface is smooth and free from scales. It
occasionally happens that the poling is continued too long
(over-poling), in which case all the suboxide is reduced,
and the metal assumes a light colour, loses its toughness,
and becomes hard. These defects are variously ascribed to
the absorption of carbon from the pole and coals, and to the
action of impurities in the metal, which are neutralised by
the presence of a small quantity of the suboxide of copper;
to remedy them, the air is again allowed to play over the
surface, in order to oxidise the carbon or produce some sub-
oxide, and the operation of poling repeated. When the
assay or trial shows the copper to be neither insufficiently
deoxidised (under-poled or dry) nor over-poled, but of a good
colour, soft, malleable, and tenacious (tough pitch), it is
ladled from the furnace into iron moulds, where it is formed
into ingots, tiles, or wire bars, according to demand. When
intended for the manufacture of brass the copper is tapped
off into vats of water and granulated.
During the ladling out of the copper the fire is kept up,
and samples continually taken. As soon as the samples
show the copper to be again somewhat oxidised, it must be
treated with wooden poles. The mould for receiving the
copper consists of a sole plate upon which stands a case
somewhat smaller above.
In order to obtain several thinner plates by the same
mould, instead of a large
a large copper ingot, the mould is
partially filled with copper by means of heated wrought iron
ladles coated with lime; the copper is allowed to cool, it
thus becomes coated with oxide, and copper is again poured
upon it; this is repeated until the mould is filled, when the
moulding case is removed and the single copper-plates are
broken asunder. The cases are also previously warmed, and
then brushed judiciously with bone dust and burned clay
in order to remove the rough spots.
Details of the refining process will be given subsequently
in the illustrations of the reverberatory process.
SMELTING PROCESSES IN SOUTH WALES.
177
The loss of copper in this process is caused—
1. By the roasting process, chemically and mechanically;
the extent of this loss is not known, and is estimated by
Le Play and Percy as not inconsiderable; but Vivian judges
it to be slight.
2. By the smelting, chiefly in the slags which are thrown
away and which usually contain not less than per cent of
copper. According to Le Play the loss at the ore smelting
amounts to 25 per cent of the total amount of copper con-
tained in the ores. Fluor spar if present in the ore increases
the loss.
3. By the sand hearths of the furnaces. When charging
24 cwts. the hearth of a furnace absorbs annually from 4
to 5 tons at the ore smelting, and 7 or 8 tons at the matt
smelting; a black copper furnace about 10 tons, and a
refining furnace about 5 tons of copper; but the greater
part of this copper is regained by breaking up the hearth
and working it up.
Illustrations of the Reverberatory Process.
A. Different Methods in South Wales.*-The methods as
here described are partly of historical interest only, as they
have recently undergone important modifications, chiefly in
the operations of calcining the ores and regulus. Gersten-
höffer's furnace before described has been recently introduced
in South Wales with great success.
The Smelting Process* as described by Le Play, comprises
all possible operations. This process, which is also adopted
by Rivot, includes the following manipulations:
Ann. d. min., 1 sér., ix., 827; X., 331, 401; xi., 207; 2 sér., vi., 3; 3 sér.,
v., 657 ; 4 sèr., xiii., 3. KARSTEN'S Archiv., 1 R., viii., 160; xiii., 60, 144; 2 R.,
XXV., 578. HARTMANN'S Repertoire, ii., 506. RUSSEGGER'S Reisen, iv., 428,
467. B. u. h. Ztg., 1848, pp. 757, 780; 1849, pp. 241, 305; 1851, p. 463; 1852, pp.
265, 352; 1853, p. 909; 1862, p. 345. TUNNER'S Jahrb., 1852, p. 27. Bgwkfd.,
vi., 369. DINGLER'S Polyt. Journ., xii., 124; lxxi., 50; lxxiii., 435; lxxvi.,
193; lxxxii., 275. LAMBORN'S Metallurgy of Copper, 1860, p. 121.
Description des Procédés Metallurgiques Employés dans les pays des
Galles pour la Fabrication du Cuivre, par M. F. LE PLAY; Paris, 1848.
Deutch von HARTMANN, unter d. Titel. Beschreibung der Hüttenprocesse,
welche in Wales zur Darstellung des Kupfers angewendet werden, 1851.
+
Métallurgie du Cuivre, par M. L. E. RIvor: Paris, 1859, PP, 117, 194.
Deutch von HARTMANN : Leipzig, 1860, pp. 73, 116.
VOL. II.
N
178
COPPER.
1. The Calcination of Copper Pyrites* associated with
much iron pyrites and quartz, and containing 35 per cent of
copper, is effected in reverberatory furnaces in charges of
3°45 tons, and finished in 12 hours, during which time two
charges are given, yielding 6'4 tons of roasted ores, and
consuming about 6 cwts. of anthracite and 2 cwts. of coal.
When roasting in Parkes's double furnaces 4 tons of ore are
charged upon each hearth and raked for 4 hours. The ore
of the upper hearth is then removed to the lower hearth,
and after being raked there for 4 hours it is removed from
the furnace; 24 tons of ore are therefore roasted in 24 hours,
consuming 1 tons of fuel.
I
2. Raw Smelting.-A charge consists of—
0.896 tons of roasted ores.
O'104
وو
0'051
0'071
0.106
0*063
0'009
""
وو
I'300 tons.
raw oxidised siliceous ores, containing
little iron pyrites, and 12 to 20 per cent
of copper.
fluor spar.
slags from the same process.
poor slags from the raw melting.
poor slags from the matt smelting.
poor slags from the process for producing
blue metal.
This is melted in reverberatory furnaces in about 4 hours.
100 cwts. of mixture yield about 27.5 of raw matt, containing
33 or 34 per cent of copper, and 65 per cent of slags con-
taining 0.8 per cent of copper. In 12 hours, or for melting
3 charges, 1677 tons of fuel is consumed, consisting of
68 per cent of small anthracite, and 32 per cent of small
coal.
3. Roasting the Raw Matt (bronze matt).-The raw
matt is first granulated, and then roasted for 36 hours at a
gradually rising temperature in furnaces similar to those
used for the ore roasting. Charges of 4'5 tons of matt are
given, consuming 1'543 tons of fuel, consisting of 77 per
cent of anthracite and 23 per cent of coal. 100 parts of raw
matt yield 94°4 roasted matt, 66'6 of which are used for the
* Erzvorkommen in Cornwall. COTTA, Erglagerstätten, ii., 464. B. u. h.
Ztg., 1862, pp. 144, 345.
SMELTING PROCESSES IN SOUTH WALES.
179
production of the common white concentrated matt, and 30.8
for that of blue metal.
4. Production of White Concentrated Matt.-Charges
consisting of—
I'ο01 tons of roasted raw matt.
oxidised copper ores, containing 25 to 45
per cent of copper.
0.435
0.108
وو
black copper slags.
0'043
""
O'012
0'107
0'075
refinery slags.
cupriferous products.
furnace ends.
hearth sand.
""
O'SII
bricks of the furnace walls.
I'739 tons
are melted in about 6 hours time.
100 parts of mixture yield 40°2 per cent of white metal,
the composition of which is shown in the following analyses.
by Le Play.
1
Cu
Fe
Ni
Co
Mn
Sn, As
S
Residue.
I.
II.
77'4
64.8
0*7
9'0
traces
traces
0°5
traces
O'I
0*7
21'0
22.6
1.8
0°3
No. I resulted from melting the roasted matt with rich
ores. No. 2 the same, somewhat bluish. There are besides
26.1 per cent of poor slags and 28'1 per cent of rich slags,
composed as follows:-
SiO3
II.
1.
33.8
33'0
FeO
56'0
55'0
Cu₂O
0'9
2.7
Other oxides
2.I
2.0
Al2O3
CaO
MgO
Cu
1'5
1.6
1
1'4
I'4
0°3
0°3
2'9
2.9
Fe
0°3
0'3
S
0.8
0.8
N 2
180
COPPER.
The fuel consists of 74 per cent of anthracite and 26 per
cent of coal, of which 3'524 tons are consumed in 24 hours.
5. Production of the Blue Metal.-In 5 hours a charge
is smelted composed of
1*592 tons of roasted raw matt.
وو
roasted ores, containing from 15 to 35 per
cent of copper.
0*408
0*185
0.018
""
bricks.
""
hearth sand.
2*203 tons.
100 parts of mixture yield 49°5 per cent of blue metal,
composed as follows:-
Cu.
Fe.
Ni.
Sn.
S.
Residue.
23°0
0'5
65'7 16:3 16 12
and 43'4 per cent of slags of the following composition :-
Other
SiO3. FeO. Cu2O. Oxides. Al2O3. CaO. MgO. Cu. Fe. S.
54'4 0'7 2.5 0.8 I'2 0°2
4.2
36'0
6. Melting the Slags.-This is effected in mixtures of the
following composition :-
1'718 tons of slags resulting from the production of the
common white metal, the white extra
metal, and the last concentration matt.
siliceous pyrites.
وو
0'166
0*116
0'099
""
0'076
0'020
2*195 tons.
scale from the rolling mills.
coal.
hearth sole.
bricks.
This charge is melted in about six hours, and 100 parts
of mixture yield about 5'7 per cent of white metal composed
as follows:
Cu. Fe. Ni. Co. Sn. S. Residue.
I'I
1.6 per cent of red metal of the following composition:-
74.6 3.1 trace o°3 20'2
Fe.
62°1 11'9 0°2 1.8
Ni. Co. Sn. As.
trace 22.8 0.7
S. Residue.
Cu.
SMELTING PROCESSES IN SOUTH WALES.
181
0.5 per cent of hard metal, with 66 per cent of copper, and
28 per cent of tin.
90 per cent of poor slags, composed as follows:-
Other
SiO 3. FeO. Oxides. Al2O3. CaO. MgO. Cu. S.
40'0 52'9 I'5 1.8 2.4 0°3 0'4 0'35
and o 8 per cent of copper bottoms, containing-
Cu
Fe, Mn, Ni.
As
Sn
S.
86'5
3°2
1.8
0'7
6'9
7. Production of White Extra Metal.-Two tons of blue
metal are smelted without any addition of fluxes for about
twelve hours, producing, by the reaction of the hearth sand
and the furnace bricks, 59 per cent of white extra metal,
consisting of—
Cu.
Fe.
S
Ni, Co, As
77'5
2'2
20'I
traces
This method also yields 10'3 per cent of poor slags, which
are given to the raw smelting, and 18 per cent of rich slags,
which are given to the slag smelting.
8. Production of the Concentrated Matt (Regulus).—
The white extra metal and the white and red metal of the
slag melting are smelted in about four hours in charges of
1*482 tons without any addition of fluxes. The smelting is
assisted by about 5 per cent of hearth and bricks of the fur-
nace, which facilitate the scorification. 100 parts of mixture.
yield 64 per cent of concentrated matt, composed as follows:-
Cu.
Fe.
SI.I
S
Ni, Co, Mn, Sn, As
0.2
18.5
traces
10 per cent of cupriferous bottoms containing 93 per cent
of copper, 14 per cent of slags, composed thus:-
SiO 3
FeO
Cu,O
A1203
CaO
34°0
52°0
12'0
ΙΟ
ΙΟ
182
COPPER.
9. Black Copper Smelting.-Charges of from 275 to
3'75 tons of common white metal, concentration metal, and
cupriferous bottoms are smelted in 24 hours with an admixture
of very pure, rich, sulphuretted and oxidised ores containing
from 60 to 80 per cent of copper. About 60 per cent of
black copper is thus obtained from 100 parts of smelting
materials, and 8 per cent of slags containing 20 per cent of
copper, which are added to the smelting of the white con-
centrated matt.
10. Refining the Black Copper.-Seven tons are treated
in 24 hours, yielding 90 or 91 per cent of fine copper varying
in quality, and 5 per cent of refining slags composed as
follows:
0'4
SiO 3.
Cu2O. FeO. NiO. MnO. SnO. Al2O3. CaO. MgO. Cu.
47'4 36.2 3'I
0°2 2.0 ΙΟ 0°2 9'0
Three sorts of copper of different purity are usually pro-
duced (vide analyses on page 77). The first kind is the best
selected copper, perfectly free from antimony, and principally
adapted for the manufacture of brass plates and wire. The
second is used for rolling purposes; and the third for casting
brass and other alloys. o'001 of antimony renders copper
unfit for the first-named use, but not for rolling purposes.
The best selected copper results from the purest ores and from
the slags. The common purer ores yield chiefly two sorts
of copper called tough copper, whilst that resulting from im-
pure ores is called tile copper. English copper is also sold
in the form of Japanese copper* in small bars, but it is less
pure than the real Japanese. For the convenience of brass
manufacturers the copper is sometimes granulated by pouring
the fused metal into cold water; this copper is called
feathered shot copper. If it is poured into hot water the re-
sulting grains are more or less round, and are called bean
shot copper.
According to Le Play and Rivot the smelting cost of
common pure and impure ores amounts respectively to
17s. 4d. and £1 2s. per ton of ore, and to £13 14s. and
£21 12s. per ton of copper.
*
Bgwkfd., vi., 379.
SMELTING PROCESSES IN SOUTH WALES.
183
The production of copper amounted in 1859 to 32,715 tons.*
*
B. Smelting Processes according to Percy,+ Napier,
and Hyde Clark.‡
These processes are carried out in the English copper
works with the following modifications:-
1. Smelting Partially Oxidised Ores of good quality,
containing from 8 to 10 per cent of copper.
a. Calcining of 3 or 3 tons of ore for from 12 to 24 hours
in furnaces of the older form.
b. Raw smelting of the ores with slags of d in the newer
melting furnaces, which are described on page 162. One
charge requires five hours' firing to fuse it; it is then raked.
The slags are removed, and a new charge given and treated in
the same way. The matt resulting from both charges is then
tapped off either into a reservoir containing water or into
iron moulds.
c. Calcining the raw matt for 24 hours.
d. Melting the roasted raw matt with slags of e and ƒ and
rich oxidised ores producing white metal.
e. Roasting smelting of the white metal for the production
of black copper, without an addition of fluxes. The matt is
melted with an admission of air in six or eight hours; the
resulting slags are twice removed, then the temperature is
lowered whilst air is admitted, and afterwards a strong heat.
is given with the doors closed. The mass is now raked, the
slags removed, and the black copper tapped off into sand
moulds.
f. Refining black copper. From 6 to 8 tons of black copper
are melted with an admission of air in fifteen hours, and
converted into the form of dry copper, whilst the slags are
repeatedly removed; they are next stirred with wooden poles
until of tough pitch, and then ladled out. Copper intended
for rolling is treated with a small addition of lead, thus
causing a dense casting with a smooth surface. This addition
B. u. h. Ztg., 1856, p. 4; 1858, p. 87; 1859, p. 360 ; 1861, pp. 170, 448.
† Ibid., 1862, p. 345.
Ante, page 135.
184
COPPER.
varies in quantity according to the purity of the copper; it
amounts in some instances to 80 lbs. for 6 tons of copper,
while in other works only from 14 to 30 lbs.
2. Smelting Ores with 7 or 8 per cent of Copper.
a. Ore calcining for twelve hours.
b. Ore smelting in charges of 21 cwts. with from 2 to
5 cwts. of slags of d, in four hours for the production of raw
matt.
c. Calcining the raw matt in from 24 to 30 hours.
d. Smelting roasted raw matt in quantities of 20 cwts.
with 4 cwts. of slags of e and f, and 3 cwts. of native car-
bonates, in six hours, for the production of concentrated
matt.
e. Smelting the concentrated matt in charges of 24 cwts.
with 2 cwts. of refinery slags of g, in six hours, for the pro-
duction of fine metal.
f. Roasting smelting the fine metal. A charge yielding
2 tons of copper is melted in twelve hours, the slags are
removed and the resulting matt and regulus cooled and re-
heated until fused, so that pimple or blister copper is
produced; the whole process is finished in 24 hours. If the
production of best selected copper is desired, the 2 tons of
fine metal are submitted to a roasting smelting and tapped
off into sand moulds, in which case then the first five moulds
will contain bottoms, while the remaining moulds contain
regulus (matt). The regulus is then re-submitted to a
smelting as before, and the resulting regulus is treated for
the production of black copper, which is refined and yields
the best selected copper.
g. Refining the black copper. From 5 to 7 tons of
coarse copper are melted with an admission of air, blisters
rise to the surface of the metal and burst; the slag is
removed and the metal stirred with wooden poles. If
instead of black copper pimple metal is under treatment,
the stirring with wooden poles is effected after melting; the
mass is then cooled and fused again, the slags are now
removed and the mass again stirred with poles till the copper
is tough. If a sample shows that it is required, from 16 to
20 lbs. of lead are mixed into the fused copper, the slags are
SMELTING PROCESS IN SOUTH WALES.
185
removed, and the metal bath is covered with charcoal or
anthracite; the copper is ladled out in the usual manner
when a sample shows that it is sufficiently fine. The whole
process takes from 70 to 96 hours, and from 13 to 18 tons of
coal are consumed per 1 ton of copper produced.
3. Treatment of Ores with 9 per cent of Copper.
a. Calcining 7 tons of ores in from 12 to 24 hours.
b. Ore smelting in charges of 22 cwts. with 6 cwts. of
slags of d, f, and h, for the production of raw matt.
c. Calcining the raw matt in from 15 to 18 hours.
d. Smelting roasted raw matt in charges of 45 cwts. with
from 6 to 12 cwts of slags of ƒ and g, or with native car-
bonates, for the production of fine metal.
e. Roasting the granulated fine metal for 18 hours.
f. Smelting the fine metal for the production of blue
or pimple metal, charging 50 cwts. of matt with from 3 to
6 cwts. of slags of h.
g. Roasting the blue or pimple metal for the production
of black copper.
2 tons of metal of ƒ are melted in
4 hours, descorified, and kept in a fused state for 18 hours
with the side doors open; during this time the slag is twice
removed. Towards the end, the heat must be increased, as
the more purified mass chills easily; the black copper
is now tapped off. If the production of the best sort of
copper is intended, the roasting is only carried on so far that
about half the result is metallic copper, and the matt (best
regulus, spongy metal) resulting at the same time is smelted
for the production of black copper.
h. Refining the black copper. 6 or 7 tons are melted in
18 hours, then descorified three times, and afterwards stirred
with wooden poles in the usual manner.
4. Smelting Modification in the case of Ores containing
9 per cent of Copper.
a. Ore smelting. 22 cwts. of ore are melted with 4 cwts.
of matt slags, for the production 'of raw matt containing
38 per cent of inatt. The matt is granulated and rolled.
b. Roasting the raw matt in charges of 5 tons for 24
hours.
c. Smelting the roasted raw matt for the production of
186
COPPER.
blue metal in charges of 36 cwts. mixed either with 3 cwts.
of slags from e and f, or with native carbonates.
d. Roasting smelting the blue metal for the production of
pimple metal, in charges of 4 tons. In order to produce the
best selected copper eight or ten of the blocks situated nearest
to the tapping hole are set apart and used for the production
of common copper. The remaining fourteen blocks are
then submitted in the following manner to the selecting
process (page 173):-5 tons of the latter blocks are heated
in the furnaces until red hot, the furnace doors being kept
closed; air is then admitted by the air channels of the fire-
bridge, and the temperature is raised so that the charge may
melt in 6 hours. The air channels are now closed, the slag
removed, and air is now again admitted. After removing the
newly formed slag the black copper is tapped off and refined.
e. Roasting smelting of the pimple metal in charges of
4 cwts. for the production of pimple copper.
f. Refining black copper in charges of 8 tons; if the
refined copper is intended for rolling mills, 35 lbs. of lead
are added.
According to Schwarz, a black copper newly imported into
England in large quantities from Australia, contains-
Cu.
99'17
Bi.
0*280
S.
Pb.
Sn. Sb. As.
0*123
0'002
0*244
Copper matt and black copper are also imported into
England from Chili.*
The smelting cost varies according to the prices of the fuel,
fire-bricks, and labour, the quality of the ores, &c. In the
last six months of the year 1859 the mines were paid £90
per ton of copper contained in the ores, whilst the market
price of copper amounted to £112 10s.
I
The smelting cost.
per ton of ore amounted to £1 3s. 3d., whilst the cost for
producing I ton of copper was not more than £10.
Copper works must produce annually at least 1,100 tons
of copper in order to be profitable, and therefore require
6 roasting furnaces, each costing £240
12 smelting
22
B. u. h. Ztg., 1862, p. 260.
£200
COPPER WORKS AT STEINWERDER.
187
and a total amount of £9,500 to £10,000, besides a floating
capital of £35,000. If such works are erected with a view to
enlargement, their capacity may be doubled with a saving of
half of the original capital.
Works of this description consume annually 20,000 tons of
coal, or 18 tons of coal to every ton of copper produced from
ores containing 10 per cent of copper; the expenses for labour,
refractory materials, &c., per ton of copper amount to
£8 15s., to which various sundries, such as freight, interest
of capital, &c., must be added. As the profit depends chiefly
on the judicious buying of the ores, and as the price of copper
is so very variable, the profit sometimes amounts to 13 percent.
interest on the original capital, whilst at other times losses
occur. If the market is brisk the capital employed may be
turned annually 2 times, and at a moderate market 24 times.
Copper Works at Steinwerder near Hamburg.*—These works
smelt copper ores of different countries and varieties,
namely:-Almost pure copper pyrites of Norway, containing
27 per cent, and 6 or 8 per cent of copper if in the form of
schlich; siliceous copper ore of the Rhine, containing 2 per
cent; pyritic ore with magnesian gangue of Tuscany, con-
taining 12 or 14 per cent; argentiferous copper ore of Peru,
containing 10 per cent; oxidised copper ore of Copiapo, con-
taining 15 per cent; and argentiferous copper matt of Chili.
All the sulphuretted and most of the oxidised ores are worked
in reverberatory furnaces, and part of the oxidised siliceous
pure ores in cupola furnaces, for the production of matt con-
taining 30 per cent of copper and slag containing per cent.
The matt is then further treated in reverberatory furnaces.
Very pure and rich oxidised ores are exceptionally treated
in cupola furnaces for the direct production of black copper.
The extraction of copper is generally effected by simpler
processes than those used in England, because antimonial
and arsenical ores are avoided as far as possible. If they
now and then occur they are worked by themselves. The
pyritic ores are pounded to the size of a nut, and mixed in a
Schles. Wochenschrift, 1859, No. 6. B. u. h. Zig., 1859, p. 100. LEON-
HARDT'S Hüttenerzeugnisse, 1858, p. 59.
188
COPPER.
raw state with such a quantity of oxidised ores that the mix-
ture, containing on an average 10 per cent, possesses the
ingredients for forming a nearly neutral slag, and the sulphur
required for the total amount of copper.
The process comprises the following operations :
1. Raw Smelting.-The ore is mixed with matt slags,
chiefly with refining slags, and more rarely with oyster
shells. It is then smelted in a reducing flame in order to
avoid the formation of black copper by the oxidation of the
pyrites, in reverberatory furnaces about 20 feet long and
15 feet broad. The hearth is composed of a quartz sand
layer from 1 to 3 feet thick. The arched roof, 8 inches
thick, is formed of Ramsay stones, and is said to stand for
five years, but if it is thicker, or covered with ash, it will
only last one year. The inner walling of the furnace is
formed of the same material, and separated from the outside
walling by a hollow space. In order to produce combustible
gases, i.e., a reducing reaction, the grate is laid about 3 feet
beneath the fire-bridge. If an oxidising reaction is desired,
two air channels attached to the fire-bridge are opened.
A furnace holds 5 tons of mixture, and yields 30 cwts. of
matt containing 30 per cent, and slag containing per cent
of copper.
For producing equal quantities of matt in reverberatory
and in cupola furnaces, 100 and 75 parts in weight of fuel
are respectively required; three smelters attend two furnaces.
2. Matt Smelting. The pure matt is roasted and smelted
direct in reverberatory furnaces, but matt containing anti-
mony and arsenic is previously treated by a separate roasting.
Argentiferous matt is treated by Ziervogel's mode of extrac-
tion (vide vol. i., page 378), and the resulting residues are
given, according to their amount of copper, either to the
concentration or the black copper smelting.
The concentration smelting is effected in the same kind of
furnaces. Raw, roasted, and desilverised matts are either
smelted each by itself or together in admixture with siliceous
oxidised ores of Chili, as well as with residues of the re-
fining process and of the smeltings in cupola furnaces. A
furnace is charged with 5 tons of mixture, which yield a
COPPER WORKS AT DUISBURG.
189
copper glance-like matt containing 60 per cent of copper,
and containing metallic copper separated in a capillary form,
besides slags containing from 6 to 8 per cent of copper. The
matt is either worked for the production of black copper, or
previously desilverised by Ziervogel's method.
3. Black Copper Smelting.-The furnaces are somewhat
smaller than the ore furnaces. The matt is usually smelted
by itself, in rare cases with an addition of residues of the
refining process. Black copper results containing 96 per
cent, and slag with from 12 to 15 per cent of copper; the last
is given to the ore or matt smelting.
4. Refining the Black Copper.-Five tons of black copper
are refined in 24 hours in from six to eight charges, con-
suming 10 tons of coal, and yielding 90 cwts. of fine copper
and 25 cwts. of slags.
Copper Works at Duisburg (Aggerthal).-The following
ores are here smelted :-
a. Barilla copper sand (page 3), containing from 75 to
80 per cent of copper; it is refined direct, yielding copper of
the first quality or crown copper. The refinery slags are
added to the pimple smelting.
b. Pure oxidised ores of Chili containing from 17 to 22 per
cent of copper, and Norwegian pyrites containing 4 or 5 per
cent of copper.
They are mixed together and melted in
charges of 34 cwts. for 7 or 8 hours, producing a raw matt
containing 33 per cent of copper, and slags with from
5 to 1 per cent, which are thrown away, consuming
about 36 cubic feet of coal. The raw matt is smelted in ad-
mixture with oxidised ores for the production of white metal
containing 75 or 80 per cent of copper, and slags containing
4 or 5 per cent, which are given to the raw smelting. The
white metal is treated in a smaller furnace for the production
of black copper, 60 cwts. consuming about 34 cubic feet of
coal. The black copper is refined in quantities of 70 or
80 cwts. in 24 hours, with a consumption of 53 cubic feet of
coal, yielding the second quality of copper.
c. Impure pyrites, containing arsenic, antimony, and lead,
of Siegen, is smelted for the production of a matt with
30 per cent of copper. Part of it is roasted and smelted in
靠
​190
COPPER.
admixture with the unroasted matt, producing pimple metal,
which is converted into black copper. The black copper is
then refined, yielding the third quality of copper.
d. Rich oxidised ores. These are smelted direct in admix-
ture with some pyrites and equal parts of reverberatory con-
centration slags, producing black copper. The smelting is
effected in cupola furnaces with two open eyes, 6 feet high,
and from 150 to 160 cwts. of mixture (80 cwts. of ore), are
smelted in 24 hours, producing black copper containing about
85 per cent, and matt containing from 50 to 56 per cent of
copper; thus I lb. of coke is consumed for every 6 or 7 lbs.
of mixture, the black copper is refined, and the matt smelted
for the production of black copper.
The refining is performed in the following manner :-
After charging from 50 to 60 cwts. of black copper the
furnace doors and air channels are hermetically closed, and
the air channels re-opened as soon as the copper becomes
red hot. It is now allowed to melt at a moderate tempe-
rature, air being admitted, when the greater part of the
copper is fused, and sulphurous acid is evolved; the air
channels are now closed and all the copper is fused at an
increased temperature, and if the hearth is free from deposits
the tough reddish slag is removed and the oxidising begins.
The more fusible slag thus formed must be removed by
spreading coke cinders upon it, while extra firing is always
given before such removal; after the first removal of slag
the metal is perfectly quiet, a ladle sample rises but without
breaking on its surface, and the fracture shows a large blister
on the upper part, while the lower part is interwoven with
small golden yellow cavities. The scale sticking to the
under part of the ladle is nearly smooth.
At this period the temperature of the furnace is very high
and the metal will remain quiet as long as that temperature
is sustained, but as soon as it is lowered so far as to allow
observations of the fire-place by the working door, a movement
or sort of ebullition of the fused mass at once takes place, as at
a certain temperature sulphide and sub-oxide of copper react
upon each other, evolving sulphurous acid. During this ebul-
lition but little slag is formed, but more is formed later on, and
COPPER WORKS AT DUISBURG.
191
the copper during this oxidising action is modified in the
following manner :-The appearance of the ladle sample
remains as before, except that the lower scale becomes
porous. Later on, the samples rise more and more vehe-
mently, so that the surface becomes broken and copper runs
out; afterwards the whole sample becomes blistered, the
cavities appear no longer yellow but a dirty red, and the
lower scale is smooth. At Mansfeld, the stirring with wooden
poles now begins; the oxidation is thus carried on still
further, the samples rise less freely and become more dense,
and the cavities again appear of a golden yellow. If the sample
shows a decidedly crystalline texture-a characteristic sign
that the copper contains suboxide of copper-the porous.
parts of the sample appear a bright white. The stirring
with wooden poles begins when the samples have become
very crystalline and show only a stripe of small cavities in
their upper parts, the cavities being violet in colour. Some
pounds of lead are now usually added to the copper before
commencing the stirring with poles, the air channels are
closed, and charcoal is thrown upon the metal bath.
Often the copper loses its oxygen very slowly by the stir-
ring; the fracture becomes purer and the crystalline appear-
ance cubic, and the porous parts are first bright whitish,
and later bright yellowish.
If the sample is rose-red in colour, perfectly dense, and
shows a radiated texture, the stirring is continued uninter-
ruptedly; but if golden yellow cavities again appear on the
fracture the stirring is discontinued, and it is instead again.
somewhat oxidised, but without opening the air channels.
If the stirring with poles is continued the samples show
golden points, become fibrous, compact, and of a silky lustre.
The surface of the ladle sample is also characteristic in
appearance, showing concentric rings extending more and
more towards the middle and having an appearance similar
to the fine silver assay.
If the copper is considered tough, judging from the sample,
two bars of it are cast; one bar is bent whilst hot, the other
when cold; no breakage should take place. If the latter
bar breaks the fracture will indicate whether suboxide is still
192
COPPER.
present, if some coal has combined with the copper, if
the copper has been stirred too much, or if it is not yet.
quite dense enough; if red-short it is best to somewhat
oxidise again and afterwards to stir with the wooden
pole.
The differences between this mode of refining and the
mode in use at Mansfeld will be pointed out in a later
chapter.
EXTRACTION OF COPPER BY COMBINED PROCESSES IN
CUPOLA AND REVERBERATORY FURNACES.
When comparing the two processes the English process.
appears to have advantages-first, in roasting ores and
matt, though the higher prices of fuel may make roasting
matts in mounds preferable, and secondly, in the refining
of black copper, as large quantities are toughened at once.
On the other hand, the process in cupola furnaces also
has its advantages, especially at the ore and matt smeltings,
as they produce purer slags and therefore a better yield;
arsenic and antimony are more perfectly volatilised, and
the copper is obtained quicker and concentrated in fewer
operations, as shown by the following:--
English
Reverberatory
Process.
Mansfeld
Process in Cupola
Furnace.
2'75 per cent
Roasted ore
20-25 per cent
Raw matt.
33
""
47
""
Roasted raw matt
34
51
White metal .
77
""
"?
Blue metal
56
Slag matt.
62-74
""
Extra metal
•
77'5
""
Regulus
Black copper .
Refined copper
If it is desirable to purify matt from admixtures, such as
lead, zinc, antimony, arsenic, &c., previous to its being
desilverised by Augustin's or Ziervogel's method, the
81.5
""
""
92
""
97-99'5
88-95
98.5-99'5
""
SMELTING THE MANSFELD COPPER SCHIST.
193
concentrating of the matt may be advantageously effected
in reverberatory furnaces. We have before stated that matt
may be concentrated to a higher degree, without the forma-
tion of black copper, in reverberatory furnaces than in cupola
furnaces.
The loss of copper is less when performing the black
copper smelting in cupola furnaces, chiefly when treating
desilverised residues containing common salt, and the anti-
mony and arsenic are more completely extracted; whilst in
the reverberatory process a great deal of copper enters the
hearth, and is only extracted with difficulty. As, however,
the reverberatory process allows a larger production and
produces a richer black copper and little matt, it is still
sometimes preferred to the smelting in cupola furnaces.
Illustrations of the Combined Smelting Processes.
Smelting the Mansfeld Copper Schist, or Bituminous Kup-
ferschiefer. The ore here treated is Copper Slate,† a
mixture of lime, alumina, bitumen, and different metallic
substances, chiefly copper glance, copper pyrites, variegated
copper ore, native copper and silver, red copper ore, malachite,
azurite, cobalt and nickel ores, galena, blende, molybdenum
and vanadium combinations, &c. The top of the copper
schist is formed of a cupriferous, calcareous layer (Noberge
und Dach), and the sole of a sandstone, impregnated a few
inches deep with ores (sand ores). The copper schist is not
usually more than 20 inches thick, and contains from 18 to
3.7 per cent of copper, and I cwt. of produced copper con-
tains from 0'53 to 0.58 lb. of silver.
The ores are separated by hand picking, and divided into
those worth smelting and those to be thrown away, the
Lan, in B. u. h. Ztg., 1852, No. 25; 1856, p. 162; 1859, PP. 371, 412;
1860, p. 501; 1861, pp. 67, 449; 1862, p. 140. Bgwkfd., xxii., 229. BEZARD,
in Bulletin de la Société de l'Industrie Minér., Tom. i., livr. 4, de 1856, p. 606.
RIVOT, Métallurgie du Cuivre, 1859, p. 405. LAMBORN'S Metallurgy of Copper,
1850, p. 146. PERCY, Metallurgy, i., 413.
+ VELTHEIM, in Karsten's Archiv. 1 R., xv. 89. COTTA, Erzlagerstätten
ii., тоб.
VOL. II.
194
COPPER.
appearance of the fracture being the leading point. They
contain about-
Sio,
CaO and MgO
•
Al2O3
Fe2O3
or, according to Berthier:-
SiO3
ALO
3
Fe₂03
CaO, CO,
MgO, CO₂
ко
Copper pyrites .
Water & Bitumen
40 to 50
13 to 22
IO to 18
3 to 5
40
10'7
5
19'5
தம்
6'5
26
10'3
The sand ores are classified into rich ores containing from
7 to 75 per cent of copper, and poor ores containing 25
per cent. They only contain o'18 lb. of silver per cwt. of
copper.
The ores of the top of the copper schist are very poor,
containing on an average 100 lbs. of copper in 60 cwts. of
ore, and are considered worth smelting if containing 85 lbs.
in 60 cwts.
The sand ores and the ores of the top are not roasted.
The smelting of the bitumenous copper slate comprises
the following operations :-
1. Roasting, which is less for the purpose of oxidising
the metallic sulphides than to remove the bitumen and
water contained in the slate, as both substances injure the
smelting process and increase the consumption of fuel.
The roasting is effected in large heaps containing 100 tons.
interstratified with brush wood, and with some slate rich in
bituminous matter mixed with the mass. These heaps are
about 6 feet high, and burn 15 weeks in fair weather and
20 weeks in rainy weather. The bitumen is decomposed, the
sulphur is dissipated chiefly in the form of sulphurous acid,
the metal becomes partially oxidised, particularly the iron,
which is a very desirable circumstance for the production of
good smelting slag. The calcined ore is diminished 1-10th
SMELTING THE MANSFELD COPPER SCHIST.
195
in bulk and 1-8th in weight, becoming of a friable texture
and a dirty yellow colour. About 10 lbs. of wood are used
per ton of ore.
The calcined slate has the following composition --
III.
IV.
V.
VI.
VII.
52.72
50*6 43°8 49′00
17°2
590 46.2 55'60
15°31 180 16'0 13.85
15.67
4'65
I.
II.
SiO,.
Al₂03
MgO J
23°4
Fe₂O,
CuO.
3
9'0 7.2
2.8
2.5
Cao.
S.
Loss.
4°14 3°3 4'5 3'74
8.00 8.0 8.0 8.00 8.00
7.8 18.0 18.10 13°2 15'6 10′97 9'99
4'0 2.4
0.8 6'0
2. Raw Melting.—An operation for scorifying the earthy
substances of the ore, and combining the sulphides into a
matt. If the roasting is carried on too far a loss of copper
takes place by scorification, and deposits of iron are formed
owing to a want of sulphur; in such cases raw ores are added
to the mixture.
The cupreous slate is sorted according to its composition
into lime, clay, iron, slate, &c., and the smelting is facilitated
by a proper admixture. For example, one part or charge
may consist of 20 cwts. of ferruginous slate, 14 calcareous,
6 argillaceous, with of 3 fluor spar, 3 of rich copper slags,
and other refuse matters.
The smelting furnaces are cupolas 14 or 18 feet high,
represented by Figs. 49, 50, 51, 52.
Fig. 49 is a vertical section through the tuyere in the
dotted line, A, B, of Fig. 51. Fig. 50 is a vertical section
through the dotted line, C, D, of Fig. 52. a is the shaft of the
furnace; b, the boshes; c, c, the tuyeres ; d, the sole or hearth
stone, which slopes 3 inches towards the front wall; e, c,
casing walls of fire bricks; f, f, &c., filling up walls, built of
stone rubbish; g, g, a mass through which the heat is slowly
conducted; h, h, the two holes through which the product of
the smelting process is alternately run off into the fore-
hearth. Beneath the hearth sole is a solid body of loam ;
and the fore-hearth is formed of a mixture of coal dust and
Dr. URE'S Dictionary of Arts, &c., vol. i., SS5.
0 2
196
COPPER.
FIG. 49.
H
C
不
​α
b
D
C
B
F
k
576
FIG. 50.

a
b
C
G
E
A
h
FIG. 52.
FIG. 51.
clay; k is the escape for moisture. Fig. 51 is a horizontal
section of the furnace through the hole or eye on the dotted
line, E, F, of Fig. 49. Fig. 52, a horizontal section of the
shaft of the furnace through the form along the dotted line,
G, H, of Figs. 49 and 50. The height of the shaft from the
line, E, F, to the top is 14 feet; from E to G, 25 inches; from
c to the line below b, 2 feet; from that line to the line oppo-
site g, g, 2 feet. The width of the line, g, g, is 2 feet
3 inches, and at e, 26 inches. The basins, i, i (Fig. 51), are
3 feet in diameter and 20 inches deep.
These furnaces have two tuyeres. The smaller furnaces
of this kind furnished with one tuyere are now being super-
seded by the larger ones.
In some of the Mansfeld copper works hot blast is advan-
tageously employed. Coke is now used exclusively for fuel
instead of charcoal. The large furnaces last one year, some-
times even two or three, in continual operation. The smelting
products are
SMELTING THE MANSFELD COPPER SCHIST.
197
a. Raw Matt.-(vide analysis on page 43).-60 cwts. of
copper slate yields at the
Cu.
lbs.
Ag.
lb.
4*50 containing 35°33 and o*195
Raw Matt.
cwts.
Eisleben Smelting Works.
Friedeberg
5'00
""
Mansfeld
>>
5°25
""
Kupferhammerhütte
""
5'50
Sangerhäuse
6.10
""
""
22.00
33'00",
0*125
0*190
26.36,, O*109
4800,, 0097
The rich copper matt is submitted to Ziervogel's method
for the extraction of silver; poor matt containing less than
40 per cent of copper is previously concentrated in a rever-
beratory furnace.
b. Raw Slags, part of which are used as a flux for later
smelting processes, besides some deposits of iron, smoke,
furnace ends, &c.
The furnace gases have been investigated by Heine;
they were produced in a cupola furnace furnished with two
tuyeres, coke being used. The pressure of the cold blast
was 10 lines of mercury, and that of the hot blast (212° C.)
from 11 to 13 lines of mercury. The following are Heine's
results:-
Charcoal.
Feet.
HOT BLAST.
Coke.
Feet.
Feet.
Feet.
Beneath the furnace top
3'00
6.00
3°00
6.00
Nitrogen
67.85 69 10
6098
6127
Carbonic oxide .
10'04
11'95
28*90 30'77
Carbonic acid
20.80 18.67
6°34 4'14
•
Carburetted hydrogen
Hydrogen
Sulphurous acid
0°30
0'22
2.68
I'94
1'31 o'99
o So
1.66
COLD BLAST.
Beneath the furnace top
Nitrogen
3'00
46.41 66.09
6'00
3'00
6'00
33.38
6102
Carbonic oxide.
22*47 23.04
35.80 32 48
Carbonic acid
Hydrogen
•
Carburetted hydrogen
Sulphurous acid
0'95
These results correspond with those obtained by Bunsen
(page 114), and prove that coke yields more carbonic acid,
and is therefore more profitable than charcoal; that the
quantity of carbonic oxide gas produced is largest when
25*25 8.00
4'92 I'97
198
COPPER.
employing charcoal and cold blast, and smallest when using
coke and hot blast, and that the loss of fuel is very con-
siderable.
3. Roasting the Raw Matt.-This is sometimes performed
in mounds, but usually in muffle furnaces combined with
leaden chambers for the manufacture of sulphuric acid. The
matt is taken from the fore-hearth of the furnace when still
hot and broken, and water is poured over it, which makes it
divide into fine powder; it is then roasted in quantities of
Io cwts. for about 11 hours at a red heat, while it is con-
tinually raked. 5 tons of matt consume 18 tons of lignite
and coal, 75 lbs. of saltpetre, and 75 lbs. of sulphuric acid
of 60° B., yielding 18 cwts. of sulphuric acid of 66° B. The
dust collected in the chambers is very rich in selenium.
4. Concentration of the Raw Matt in Reverberatory
Furnaces. These furnaces are employed in preference to
cupola furnaces in order to avoid the formation of black.
copper, which is required for the subsequent desilverisation
of the matt by Ziervogel's method. Lead, when present,
FIG. 53.

а
С
α
d-
1k
b
Ъ
is also more perfectly extracted in these furnaces. Those
of later construction are represented by Figs. 53, 54, 55.
For the formation of the hearth, raw slags 6 or 7 inches
high are laid upon the foundation, next 6 inches of sand,
SMELTING THE MANSFELD COPPER SCHIST.
199
which is then heated for drying, then an admixture of quartz
sand and pounded raw slag (56 cwts. of quartz and 8 cwts.
of slag) 2 inches high, is spread repeatedly upon the sand
FIG. 54.
f_
7721

d
་་
until the mixture is 8 inches high in all; a strong fire is given,
after each charging of the mixture, for about two hours.
When the hearth is pasty from the last heating it is made.
smooth and beaten down into the suitable form by deepening
it towards the middle and giving it a slope towards the
FIG. 55.

aka aka
d
tapping hole. The hearth is now closed and heated for 16 or
18 hours till it frits; it is afterwards gradually cooled and
charged with matt. The matt roasted in the muffle furnaces
is charged in quantities of 50 cwts. with an addition of 4 cwts.
of sand and 2'6 cwts. of raw slag.
The resulting concentrated matt is made to run into a
current of water and is thus granulated. 100 cwts. of raw
200
COPPER.
matt yield the following amount of concentrated matt in
the different copper works of the Mansfeld district :—
وو
Eisleben Works
Kupferhammer
Friedeberg
Mansfeld
Sangerhäuse
وو
cwts.
Cu.
per cent.
Ag.
per cent.
50'6 containing 696 and 0*384
65°25 0'345
64'70
""
,, 0*368
50.6
""
48.7
65'0
>>
67.61 0'390
68.00
"" O'135
This concentrated matt, or spurstein, is desilverised by
Ziervogel's method, and the remaining residues treated for
the production of black copper.
Slags and cupriferous hearth are also yielded at the con-
centration smelting. The slags contain-
SiO3 ·
Cu₂O.
FeO.
CaO.
A1,03
· 33.6
3°0
· 37°0
5'0
5'6
5. Treatment of the Desilverised Residues for the
Production of Black Copper.-The residues are mixed
with from 3 to 6 per cent of pounded Dünnstein in order to
lessen the scorification of copper, and then intimately com-
bined with moistened clay and formed into lumps; they are
next dried and then melted, with an addition of 10 per cent
of raw slag, in cupola furnaces with open eyes, and furnished
with one tuyere, the hearth sole of which is formed of quartz.
From 60 to 8o cwts. of mixture are smelted with coke fuel
in 24 hours. The products are black copper, thin matt, and
slags. According to Gerhardt, the black copper is composed
as follows:
Cu.
Pb.
Fe.
Ni and Co
Zn.
Ag.
S.
• 92.677
1*870
I'931
0.626
1*721
Ο ΙΙΟ
O'993
SMELTING THE MANSFELD COPPER SCHIST.
201
The thin matt containing 65 per cent of copper and o‘o11 per
cent of silver is collected in sufficient quantity and roasted
and treated for the production of black copper. The residual
slag, containing first I per cent and afterwards only o°2 or
0'3 per cent of copper, is thrown away.
6. Roasting or Oxidising Smelting (Purifying) the
Black Copper in Small Hearths* (vide page 58).—6 or 7
cwts. of black copper are charged at a time; 100 cwts. yield
89 cwts. of copper in rosettes or blocks, and 16 cwts. of hearth
ends.
According to Nauwerk, the rosette copper contains-
Cu.
Ag.
Pb.
Fe.
Ni .
S
)
a
ن
C
· 97°17
0°02
0'40
0'07
0°34
FIG. 56.
0004 (= 0.019 Cu,S)
1*755 (=15⋅643 Cu₂O)

Ch
2
* B. u. h. Ztg., 1861, p. 465.
72
770
202
COPPER.
This large amount of Cu₂S is derived from the coating of
the very thin rosettes or shavings, as 100 or more are formed
in one hearth. The possibility of forming so many cakes
also indicates the good quality of the copper.
Refining Black Copper in Reverberatory Furnaces.
-At the copper works at Hettstädt, gas reverberatory fur-
naces, represented by Figs. 56, 57, are in use. a is the shaft
of the gas generating furnace; b, grate; c, shaft for charging
the gas generating furnace; d, cover, to be made tight by
sand; e, channels to conduct the air for combustion; f, fire-
bridge provided with an opening, g, for cooling; h, openings
for admitting the air for combustion, if the channels do not
conduct it in sufficient quantity; i, channels for conducting
the air for oxidation; k, hearth walling; l, door for charging
the furnace, opposite to which a tapping hole is made that
the hearth may be quickly emptied of the liquid copper when
necessary; m, working door; n, flue communicating with
a chimney, o, 60 feet high; the chimney stands at some
FIG. 57.

IN 12 6
ހ
ހ
E
In
K
2
Z
0
3
6
9
12 FI
REFINING BLACK COPPER AT MANSFELD.
203
distance from the furnace and so lasts longer; flue walls
5 inches thick last longer than thicker ones; p, refractory
admixture of three parts quartz and one part loam ; q, out-
side walling.
The hearth is formed like the hearth of the concentration
furnace (page 198), of 10 or 12 parts of quartz and I part slate
slags in six or seven layers, each 1 or 2 inches thick, sloping
1 or 2 inches towards the working door, m, and 1 inch towards
the tapping hole. The hearth rests upon the hearth walling, k.
The same refractory mass projects 3 inches high round the
side walls and the fire-bridge.
The refining process comprises the following operations:-
a. Charging (3h.).*-5 tons of black copper are charged
into the red hot furnace by four workmen, and so divided
over the hearth that the copper lies highest on the fire-bridge.
b. Smelting (4h.)—The generator is charged with char-
coal about every 2 or 3 hours during the whole process, the
doors and channels for the oxidising air are hermetically
closed, and the channels for admitting air for combustion are
opened 10 inches; the chimney lid is also somewhat opened.
A strong fire is now given, and the copper becomes red hot
in about 4h. 30m., and melted at about Ioh. As soon
as the copper begins to flow the oxidation air channels
are opened for some time, and re-closed when the copper
is so far melted as to show no more hollow spaces between
the single pieces. The melted copper is now stirred with a
rake in order to remove pieces of copper which have remained
unfused on the fire-bridge or the side walls, and a thick un-
fused crisp crust of metallic oxides appears on the surface
of the metal. In order to liberate this crust from the
mechanically enclosed copper a short but strong heat is
again given. At 10h. 35m. quartz sand is sprinkled before
the working door, a screen of iron plate put upon the floor
for protecting the workmen, and the crisp black slag is
carefully removed, whilst small coal is thrown into the fur-
nace. The coal rubbish is strewn over it to prevent its
*
3h. means 3 o'clock, the beginning of the charging; figures behind the
h denote the minutes.
204
COPPER.
caking; it is cooled with water and broken with a pointed
iron. The working door is cleaned from deposits, which are
thrown back on the hearth, as they contain mechanically
enclosed copper. The metallic bath is sea-green in colour,
and evolves zinc in the form of gas.
According to Gerhardt, the first scum is composed as
follows:-
Cu₂O
FeO
PbQ
Coo
ZnO
SiO
MgO
CaO
Al₂03
50*16
27'70
2.23
4'05
10°37
3'70
O'14
I'06
0'34
If the alumina is considered as an acid, the oxygen of the
acids stands in proportion to the oxygen of the bases as I: 7'5.
According to this analysis Fe, Co, and Zn are scorified first,
but no Ni.
(At Duisburg (page 190) the smelting of the purer copper
containing 97 per cent, and richer in oxygen and poorer in
sulphur, is effected with an admission of air and in two hours'
longer time, which facilitates the subsequent purification.
Owing to the greater purity of the copper, easily fusible red
coloured slags result at once, and the smaller amount of
sulphur causes less sparkling, as the greater part of the
sulphur is already removed at the smelting).
If the more impure Mansfeld copper, containing 92 or
94 per cent of copper, little oxygen, and about 1 per cent of
sulphur, were smelted with an admission of air, the loss of
time would perhaps outbalance the gain by oxidation.
c. Oxidising (11h.) The oxidation channels are entirely
opened, and the working and charging doors partly so if
the temperature allows it. The formation of slag now
begins, and is indicated by a rotating movement of the metal
bath and a peculiar faint noise. The slag is then moved back-
wards and forwards by a wooden pole in order to bring it
into contact with the air; two or three poles are sometimes
thrown upon it, as their movement causes the surface of the
REFINING BLACK COPPER AT MANSFELD.
205
metal bath to be partially freed from slag. If much fusible slag
is formed, coke cinders are thrown upon it, and it is removed.
In order to render the slag more pasty small charcoal is
spread upon it; small coke has the reverse effect. The oxi-
dation air is turned off during the removal of the slag, which
contains, according to Nauwerk,-
Cu₂O
РЬО
FeO
NiO
ZnO
AgO
SiO3
75'61
5'09
3'06
I'15
2.16
0*06
II 18
proving that at this period chiefly Fe and Pb and some Ni
become scorified by the Cu₂O which is formed in great quan-
tity, whilst at the same time much zinc volatilises. As the
re-opened air channels give rise to the formation of a tough
slag mixed with copper, the temperature is raised to keep
the slag sufficiently liquid. After the second removal of the
slag the copper attains an undulating movement, and as soon
as the metal has combined with more Cu₂O the oxidation of
the Cu₂S still present begins, which at a certain proportion of
both and at a moderate temperature causes a sparkling,
whilst sulphurous acid is briskly evolved (12h.) Cu₂O
and Cu₂S react less upon each other at a high temperature
than at a certain low temperature, which is employed at
Duisburg, for instance, at the commencement of the oxidation
period. Through this low temperature the copper is ren-
dered dense and tough by stirring it once with wooden poles.
About one hour after the commencement of the oxidation
the copper often sparkles suddenly and with great in-
tensity, without the usual hissing noise and formation of
bubbles, this phenomenon extending all over the surface of
the copper. An immense number of small globules are vio-
lently thrown against the arched roof, the whole metal bath
is in movement as though caused by rain; the slag is thus
moved back and roundish pieces of it frequently rotate, whilst
its formation is diminished.
As the black copper contains on an average 1 per cent of
206
COPPER.
sulphur, the total quantity in the charge amounts to 100 lbs.,
yielding 200 lbs. of sulphurous acid, or 118,174 cubic
metres = 3,746 cubic feet at 1000° C. If this phenomenon
continues for two hours 31°21 cubic feet of sulphurous acid
are evolved every minute. But as the sparkling with great
intensity only lasts a short time, it may be supposed that
at the last, four times that amount of sulphurous acid,
equal to 120 cubic feet, escape. If the surface of the
metal bath measures 40 square feet it would follow that
I square foot evolves 3 cubic feet of gas every minute, or
I square inch o‘6 cubic foot every second, and as the evolu-
tion of gas takes place at short intervals, 12 cubic feet per
square inch escapes every second. The intensity of the
sparkling may be thus explained:-It lasts for half or three-
quarters of an hour and gradually ceases; larger bubbles
appear, and the hissing noise changes into a boiling bubbling
noise. The slag is thinner, and flows from the working door,
beginning at the period of the boiling (Braten) (1h.)
The metal bath undulates excessively where the air has
freest access; when copper stripes resulting from the ingots.
are added to the fused metal they cause around them a
foaming until they are fused. After about two hours (3h.)
the metal bath becomes quieter and attains an even surface,
but the last particles of sulphur are still retained with great
pertinacity. If Cu₂O is present in excess, sulphurous acid
will still be formed, but it becomes absorbed, and escapes
only at the cooling of the metal, then causing the rising of
the copper. Over refined or dry copper appears to possess
the quality of keeping sulphurous acid absorbed ; a very thin
liquid slag is thus formed; it is removed, whilst coal dust is
thrown upon it. At this period the slag must be removed
as quickly as possible, shutting out the oxidation air, in order
to prevent the evolution of gas. A sample now taken with
a ladle sparkles on cooling; it then rises, ejecting small glo-
bules, and after its surface chills crater-like risings are formed
upon it. The more nearly the copper approaches fineness
the more it rises; this must be observed by frequent samples.
When judging the samples the surface of the metal in the
ladle and the crust sticking to the lower part of the ladle
REFINING BLACK COPPER AT MANSFELD.
207
must be noticed. The surface is first black, and reddish
yellow spots appear later. At the rising of the copper
crater-like elevations are formed, and remain isolated, some
copper flows out of them later, and runs down on to the
ladle, forming worm-like concretions upon the surface of the
sample, and wire-like deposits below. The crust on the
lower outside part of the ladle is first covered with large
holes which are arranged net-like. The finer the copper
the smaller these holes are, and they gradually diminish
until at last only feathery designs remain. This indicates
the fineness, and (5h.) for the purpose of rendering the
copper compact, which operation is termed "dichtpolen,'
d. The Stirring with Wooden Poles now begins. The
mechanical expulsion of the sulphurous acid and the reduc-
tion of the Cu₂O are aimed at. By the reduction of sub-
oxide of copper the copper loses the capacity of absorbing
sulphurous acid. While the air channels are kept open
one end of a green birch pole, from 18 to 20 feet long, and
4 or 5 inches thick, is dipped into the metal bath through
the working door, the thinner end being supported by a fork.
A strong ebullition and a continuous formation of slag are
thus caused; this slag is removed whilst cinders are thrown
upon it. At this reaction Ni, Zn, and Pb chiefly are scori-
fied, as the slag of this period is composed as follows:-
Cu₂O
NiO, COO, MnO
ZnO
PbO
FeO
A1,03
CaO
MgO
Sio
60*64
11.83
10 16
9°36
2.45
0'21
I'03
0'97
3°52
:
The amount of oxygen in the acid stands in proportion to
that of the bases as I 67. This slag contains less Cu₂O
than the former, as this oxide is partly reduced by the
wooden pole.
After some time a sample taken still shows some rising
by the formation of a convex surface, and, when cooling, by
208
COPPER.
a black spot (5h. 10m.) Later samples do not show the
rising superficially, and they must then be judged by the
nature of their fracture. For this purpose the sample is
cooled in water, put into a vice, cut with a small chisel, and
broken. Samples taken at this period are not homogeneous
in their fracture, the upper part of the fracture being even,
while the lower part is trough-like, and the limit between
both parts lies deeper the more Cu₂O the copper contains ;
as the process proceeds the limit appears higher until at
last it disappears. After employing the pole for a quarter
of an hour (5h. 15m.) the limit is seen very distinctly; the
fracture is fine grained on the upper part, with violet coloured
spots, and on the lower part more radiated. From the limit
upwards to the surface small holes are seen of the size of a
pin's head with a bright lustre, and sometimes filled with a
network; they decrease in size as the process proceeds.
5h. 20m.), and finally disappear. As soon as no more holes
are observed by means of a lens (5h. 30m.) the wooden pole.
is taken out. Though the sample now appears dense the
copper still contains some traces of sulphur and Cu₂O, as
the absorbed sulphurous acid seems to prevent a decom-
position of Cu₂S by Cu,O. It is therefore necessary to
oxidise twice more, stirring with wooden poles after each
time.
Whilst at Mansfeld the wooden poles are first employed
when the copper rises most, at Duisburg they are used when
the rising has almost ceased; but this seems to prolong the
process, as at Mansfeld 5 tons of black copper are refined in
the time required at Duisburg for refining 50 cwts. of purer
black copper. At Mansfeld a difference is made between
toughening the copper and rendering it merely compact. The
latter operation is performed by poling the copper with an
admission of air. At Duisburg a tough copper is produced
at once, air being excluded; this method does not allow a
strict control of the process, and a porous copper is easily
produced; a copper may be porous and yet moderately tough.
The above-mentioned limit at Mansfeld, after the first oxi-
dising and poling, is not observable in the examples at Duis-
burg, where the poling commences later when the copper is
REFINING BLACK COPPER AT MANSFEld.
209
already very pure.
show this limit.
The Mansfeld samples also cease to
e. Oxidising (6h).-At this period the limit sinks very low
and is sometimes hardly visible, owing to the fresh formation
of Cu₂O; the fracture appears crystalline with a violet colour;
if the limit re-appears the part beneath it is of a radiated
structure.
f. Second Stirring with Poles (Dichtpolen) (6h. 15m.)—
After some time the copper is stirred again, the limit rises
and is clearly to be seen, the upper part is crystalline and
violet-coloured; upon the limit appears small bright holes,
and the texture beneath it is interwoven and radiated. The
limit rises more and more (6h. 20m.), the fracture above it
becomes more finely crystallised, the violet colour gradually
disappears, and the porous parts are observable only by
means of a lens. Later on (6h. 40m.) the limit comes near the
top, the upper part is composed of very fine crystals without
violet spots, but with bright points instead, the lower part
becomes lighter, and the limit finally disappears (6h. 48m.),
whilst the greater part of the fracture is very finely crystal-
lised, intermixed with bright, gold-like points, but with a
bluish shade on both sides. At 6h. 53m., the texture becomes
finely granular, showing a silky aspect. Sometimes the
sample now shows larger or smaller segregations of a
brilliant golden yellow colour and of a pin-like interwoven
texture, appearing distinctly upon the rose-coloured copper,
which is also dense; this appearance is said to be caused by
traces of nickel and sulphur.
g. Second Oxidisation (7h.).-These segregations are
quite as injurious as the brilliant cavities, and must be
destroyed by a renewed oxidising and a subsequent poling.
The metal is raked and oxidised, thus combining with sub-
oxide, but slowly, owing to its greater purity; the sample
shows no modification in the first quarter of an hour
(7h. 15m.), but the segregations soon reappear, and later on
the copper combines with Cu₂O, the segregations become
scarcer, but the limit characteristic of a greater amount of
Cu₂O does not appear.
VOL II.
P
210
COPPER.
h. Third Poling (Dichtpolen) (7h. 25m.)—At the begin-
ning of this manipulation the samples are slightly modified
(7h. 35.); as soon as they show a finely granular texture,
many golden spots of a silky lustre, and a fine rose colour,
the pole is removed, the air-channels are closed, the metal
bath is covered with soft charcoal; the working-door is also
closed, and the metal bath is allowed to remain quiet for an
hour in order to raise it to a high temperature.
i. The Tough Poling is now commenced (8h. 35m.) in
order to remove the last particles of Cu₂O. The samples
are taken in this period with a ladle 1 inch wide, and are
more difficult to estimate than those of the former polings.
They are tough, difficult to break, and only in rare cases
show a limit; as a rule their texture is homogeneous, the
colour becomes lighter, the texture fine and fibrous, and the
silky lustre increases. If the poling is carried on too long
the copper combines with coal (for Dick's views on this point
see page 81) and a border re-appears, above which the
mass is finely fibrous, of a rich silky lustre, which turns
grey when a greater amount of carbon is taken into combi-
nation; when this is the case the air channel and charging
door are opened.
3
k. The Ladling of the Copper.-As soon as the copper
is tough (8h. 55m.), containing no more Cu₂O, four workmen
wearing linen sleeves coated with loam-water ladle out
the copper, by means of ladles 6 or 8 inches wide and
4 inches deep, made of wrought iron plates of an inch
thick. During the operation, experiments of forging the
copper are made by hammering one part of a sample to a
thin plate, and the other to a square rod, which, when bent,
must show no cracks. As the copper oxidises easily during
the ladling, samples are taken from time to time, and, if
necessary, it is stirred with wooden poles till it has again
become normal. Samples containing a larger amount of
carbon are said to become hard, brittle, and slaty. As
finally (11h. 30m.), a little copper remains on the hearth
which cannot be treated with the poles, it is kept separate
and employed for the manufacture of brass. The furnace is
now cooled by opening the doors and air channels, and after
PROCESS AT SCHMOLLNITZCH
2II
.
the necessary repairs have been made (3h.) a new charge is
given.
5 tons of black copper, containing 94 per cent of copper,
yield about 85.8 per cent of fine copper, and 1975 cwts. of
residues containing from 35 to 65 per cent of copper, con-
suming about 1,220 cubic feet of hard and 85 cubic feet of
soft charcoal.
Tough copper, of the following composition (Raffinad), is
of excellent quality, and is suitable for the manufacture of
sheets for percussion caps; according to Nauwerk, it con-
tains,-
Cu
Pb
99*64
0'13
Ni
Ag
Ο ΙΟ
0'02
The resulting furnace ends are heated together with the
smoke until they cake, and afterwards mixed with 50 per cent
of slate slag, from 1 to 3 per cent of fluor spar, per cent of
iron turnings, and 2 or 3 per cent of furnace hearth, and
melted in a cupola furnace with an open eye. The resulting
black copper is refined, but with some modifications of the
method of refining good black copper.
Process at the Schmöllnitz Copper Works in Upper Hungary.
The ores here treated consist of slate ores free from silver,
fallow ores containing about 10 per cent of copper and of
an ounce of silver, and roasted copper pyrites.
1. Raw Smelting.-48 per cent of raw fallow ores, 42
per cent of roasted slate ores, and 10 per cent of argentiferous
furnace ends were melted in 1860; the mixture contained
on an average 9 per cent of copper, and o'055 per cent of
silver. The melting was effected with an addition of 2.8 per
cent of limestone and 52 per cent of argentiferous copper slag,
yielding 37 per cent of raw matt containing about 25 per
cent of copper and 1 ounce of silver, besides 7 per cent of
raw speiss containing 29 lbs. of copper and 3 ounces of
silver. 5 tons of cupriferous substances (without copper
slags) were melted in 12 hours, at a consumption of 42 per
cent of coal.
P 2
212
COPPER.
The smelting furnace has nearly the same dimensions as
the furnace used for melting the Gelf ores. (Page 102).
2. Roasting the New Matt.-This matt is roasted
differently, according to whether it is to be submitted to a
precipitating smelting or to a black smelting. For a pre-
cipitating smelting it is roasted in four or five fires for 36 days,
consuming o‘65 cubic feet of wood and o˚44 cubic feet of
coal per cwt. of roasting mass; for the black copper smelting
it is roasted in 12 or 15 fires for 80 or 90 days, consuming
2.60 cubic feet of wood and 15 cubic foot of coal per cwt.
of roasting mass.
3. Concentration Smelting.-94 per cent of roasted raw
matt, containing 15 per cent of copper and of an ounce of
silver, with 6 per cent of siliceous ore, were smelted in
1859, together with fallow ores, containing 10 lbs. of copper
and 5-8ths. of an ounce of silver, and with an addition of
13 per cent of quartz. 155 cwts. of cupriferous substances
were melted in 12 hours at a consumption of 30 per cent
of coal, yielding 58 per cent of concentrated (rich) matt,
containing from 24 to 30 lbs. of copper and I or 1 ounce of
silver, besides o˚58 per cent of rich speiss, containing 39 lbs. of
copper and 5 or 6 ounces of silver. The rich matt is
roasted in 12 fires for 70 days, consuming 2'16 cubic feet of
wood, and 2.20 cubic feet of coal per cwt. of roasting mass.
Raw and rich speiss are both desilverised by Augustin's
method.
4. Black Copper Smelting.-
a. With raw matt.-In 1860 were melted
85 per cent of raw matt, containing 23 lbs. of Cu and
1 ounces of Ag;
10 per cent of rich matt, containing 65 lbs. of Cu and
1 ounces of Ag;
5 per cent of furnace ends, containing 10 lbs. of Cu
and ounce of Ag; and
16 per cent of quartz.
115 cwts. of cupriferous substances were smelted in 12
hours at a consumption of 45 per cent of coal, yielding 25
per cent of black copper, containing from 78 to 82 per cent
of copper and about 44 ounces of silver, and 11 per cent of rich
SMELTING PROCESSES IN LOWER HUNGARY.
213
matt, containing 65 or 70 lbs. of copper and 2 ounces of
silver.
b. With Rich Matt.-In 1859 were melted-
83 per cent of concentrated matt, containing 28 lbs.
of copper and 13 ounces of Ag;
8
17 per cent of black copper matt, containing 65 lbs.
of copper, and 1 ounces of Ag; and
12 per cent of quartz.
127 cwts. of cupriferous substances were smelted in 12
hours, at a consumption of 37 per cent of coal, yielding—
30 per cent of black copper, containing 84 lbs. of
copper, and 33 ounces of silver;
9 per cent of matt, containing 65 lbs. of copper,
and 13 ounces of silver.
Formerly the black copper was desilverised by amalgama-
tion, but Augustin's method is now used.
The residues are treated for the extraction of their con-
tained copper.
In 1860 were smelted in a blast reverberatory furnace-
65 per cent of residues, containing 58 per cent of
copper (resulting from Augustin's process);
24 per cent of matt, containing 42 per cent of copper;
II per cent of concentrated matt, containing 68 per
cent of copper; and
5 per cent of quartz ;
producing-
38 per cent of black copper with 95 to 97 per cent
of copper; and
43 per cent of slag with 30 to 40 per cent of copper;
at a consumption per cwt. of smelting mass of 5'6 cubic
feet of wood, equal to 28 per cent of charcoal.
The last resulting slags and furnace ends are worked up
by separate processes for the production of refined copper.
Smelting Works in Lower Hungary (Schemnitz, Kremnitz,
Neusohl, Zsarnoviz, Tajova ).—Copper matt obtained in the
smelting works at Schemnitz, Kremnitz, and Neusohl, from
the desilverising of ores (page 117) by means of lead, con-
tains from 40 to 50 per cent of copper, from 5 to 8 per cent
214
COPPER.
of lead, and from about o'150 to 0*260 per cent of silver; it
is worked up in the smelting works at Tajova. The matt is
roasted ten or twelve times and smelted in a moderately high
cupola furnace for the production of plumbiferous black
copper and a pure matt rich in copper. Both products are
desilverised by Augustin's method.
The residues of the extraction of silver are treated in
reverberatory furnaces for the production of black copper in
the same manner as in Upper Hungary (page 212). The
black copper is refined in reverberatory furnaces.
In 1860 the average loss of copper in smelting black
copper and matt amounted to o˚4 per cent, and to o°46 per
cent in the refining process.
Freiberg Smelting Works.-Copper matt resulting from the
lead smelting process (vol. i., p. 136), and containing 30 per
cent of lead, is smelted in cupola furnaces after being roasted.
The smelting products are:-
1. Plumbiferous Black Copper, which is smelted in
cupola furnaces together with roasted copper ores, thus
yielding-
a. Raw lead.
b. Copper matt, which is concentrated in reverberatory
furnaces.
c. Speiss, yielding when refined saleable speiss, copper
matt for concentration, and slags for the lead smelting pro-
cess.
2. Slags, which are given to the lead smelting.
3. Matt, which is concentrated with a view to its desil-
verisation in an English reverberatory furnace with an oval
hearth 13 feet long and 8 feet broad.
A charge, consisting of 1 ton of matt four or six times roasted,
from 2 to 6 cwts. of rich copper slags, and from 2 to 4 per
cent of quartz or siliceous copper ores, or raw slags and
heavy spar, is smelted in four or five hours, then raked, and
the slag removed as soon as the matt has settled to the
bottom. Leaving this matt in the furnace, a second charge
is given and the process repeated; the matt of both charges
SMELTING PROCESS AT BOSTON.
215
is now tapped off into sand moulds, after having previously
removed the slags. Four or five smelting operations may
be effected in 24 hours, two smeltings yielding from 30 to
34 cwts. of concentrated matt containing from 60 to 70 per
cent of copper.
Part of the concentrated matt is submitted to an oxidising
roasting, and lixiviated with dilute sulphuric acid. The
lixivium is boiled down for the production of mixed copper
and iron vitriol, and the argentiferous residue is given to the
lead smeltings. Another part of the concentrated matt is
submitted to Augustin's process, the resulting residues of
which are smelted in reverberatory furnaces, together with
matt produced from black copper slags and quartz, or with
cement copper free from silver, with iron pyrites, or with roasted
ores rich in copper and free from silver; one charge, for
instance, may consist of 12 cwts. of residues and 7 or 8 cwts.
of copper matt, or of 15 cwts. of residues and 3 or 4 cwts. of
iron pyrites. Six or seven of these charges are smelted in
24 hours, and during this time the raw copper is usually
tapped off three times, yielding each time 15 or 16 cwts.
The resulting slags, containing from 8 to II per cent of
copper, are smelted with 35 or 40 per cent of iron pyrites,
producing matt and slag containing 3 per cent of copper,
which is used as a flux in the lead smelting. The raw copper
produced contains from 94 to 97 per cent of copper; this is
refined in Grünthal.
Augustin's process has of late been abandoned here.
Copper Works at Boston in America.-These works treat
chiefly the less oxidised sulphuretted ores of North and South
America, containing on an average 20 per cent of copper.
The smelting is effected in cupola furnaces, as the price of
fuel (anthracite) is rather high.
The purest and richest pyrites are roasted by themselves
and smelted for the production of a copper of good quality,
being mixed at the black copper smelting with very pure oxi-
dised ores.
Ores containing antimony and arsenic are mixed
with much iron pyrites and roasted in heaps of 100 or 150
tons for two months, 100 tons of ore consuming about
77 cubic metres of wood, at a cost of £4 13s. per ton.
216
COPPER.
The roasted ore is melted in the following proportions:
Roasted ore.
Impure raw ore
Roasting bottoms and furnace soot
Black copper and refining slag
Oyster shells
I'000 ton.
0'033
""
0'050
0*330
""
Ο ΙΟΟ
1*513 tons.
Cupola furnaces 1.65 metres high above the tuyere, o 65
metre wide, and 0.78 metre deep, constructed as sump
furnaces, are used for smelting; anthracite is used as fuel.
The smelting is rapidly effected, thus economising fuel, but
preventing a perfect removal of antimony and arsenic. If
the operation is in a normal state more than 10 tons of mix-
ture are smelted in 24 hours; the matt is tapped off three or
four times in 12 hours, the blast being removed. The opera-
tions of a furnace last from 14 to 16 days. 6'2 tons of ore
or 17 tons of mixture are on an average smelted in 24 hours,
consuming 12 tons of anthracite, and 0.3 tons of coal for
producing the blast. In 15 days 90 tons of roasted ore and
3 tons of raw ore are smelted, producing 52 tons of matt
containing 34 or 35 per cent of copper, at an expense of about
14s, per ton of ore. The slags contain 52 per cent of silica
and only 035 per cent of copper; they are richer when
working richer and purer ores, in which case also some black
copper will be produced as well as the raw matt.
The raw matt produced from pure ores is roasted twice,
and that from impure ores three times, consuming o‘08 ton
of wood per ton of ore, and costing 1s. 94d. per ton.
A charge of mixture for smelting black copper consists of-
Roasted matt
Ore slags.
Roasted concentrated matt .
Black copper slags
Oxidised ores
Furnace deposits
I'00 ton.
0'45
0.06 ""
0'14 ,,
0°05
1'78 tons.
About 120 charges or 120 tons of roasted matt are smelted
in the ore furnace in 15 days, producing 56 or 57 tons of
PROCESS AT THE COPPER WORKS ON THE RHONE. 217
black copper containing 90 or 92 per cent of copper, and from
7 to 8 tons of matt, at an expense of about 8s. 6d. per ton of
roasted matt, and at a consumption of 17.5 tons of anthracite.
The black copper is refined in reverberatory furnaces in
charges of 8 tons. The purer sorts require 10 hours for
smelting; the impure sorts from 15 to 16 hours. They are
subjected to an oxidising treatment for six hours and then
poled in the English manner, so that an operation is finished
in from 24 to 30 hours. From 2.85 to 3 tons of coal are
consumed in 24 hours, and from 78 to 83 per cent of fine
copper is produced from the black copper, or 20 per cent from
the ores. The cost of refining is about 20s. per ton of black
copper, and about 4s. 9d. per ton of ore.
The cost of working 1 ton of ore by the German method
amounts to £1 6s., and to £3 1S. when following the English
method. Whilst the cost for labour is nearly the same in both
processes, the reverberatory furnace requires three times the
quantity of fuel, producing richer slags, and removing anti-
mony and arsenic less completely.
Copper Works on the Rhone.-These works smelt pyritic
and oxidised ores from Spain, Asia Minor, North and West
Africa, Peru, Corsica, &c.; they treat pyritic ores separate
from oxidised ores, and also cupriferous smelting and arti-
ficial products by themselves.
I. Pyritic Ores.-These are ground and roasted in charges
of I ton or I ton 4 cwts., in reverberatory furnaces with a
hearth 45 metres long and 3'5 metres broad; one charge is
roasted in twelve hours, at a consumption of 6 cwts. of coal.
The roasting mass, mixed so as to contain 15 or 17 per cent
of copper, is smelted for the production of raw matt, in sump
furnaces 4 or 6 metres high, o'6 metre wide and o'8 metre
deep. 2 or 3 tons of admixture are smelted in 24 hours,
consuming from 35 to 40 per cent of coke, calculated on the
weight of the smelted ore. The raw matt is roasted in
charges of 1 ton in 12 hours, at a consumption of 4 cwts. of
coal and 4 cwts. of brown coal, and the roasting mass,
containing from 25 to 30 per cent of copper, is smelted in
the ore furnace with an addition of rich slags, furnace
bottoms, smoke, &c. ; 3 tons of this mixture are smelted in
218
COPPER.
24 hours, consuming 28 or 30 lbs. of coal per 100 lbs. of
mixture. The resulting concentrated matt, containing from
45 to 50 per cent of copper, is roasted, and smelted for the
production of black copper containing 60 or 65 per cent of
copper, which is then submitted to an oxidising smelting
(purifying or first refining) in reverberatory furnaces in charges
of 2 or 3 tons for from 24 to 36 hours. The black copper
is alternately heated and cooled by opening the working
doors, thus oxidising the foreign substances; the whole mass.
is now fused, the slag removed, and the copper cast into bars.
The fuel consumed is nearly equal in weight to the black
copper treated. The bars of copper (rosette copper) are
refined in the same furnace by quickly melting 2 or 2 tons
of them, then covering the surface with small coal, poling,
and skimming off. A strong heat is now given, and the
copper is cast into moulds if a sample taken shows a flesh
red colour, a fibrous silky fracture, and a sufficient degree of
malleability. An operation takes from 8 to 12 hours, and
I ton of rosette copper consumes a ton of charcoal and
50 or 60 lbs. of pine wood. The hearth of the refining
furnace is 3 metres long, 17 metres broad, and has an in-
clination of 0'2 metre from the fire-bridge to its deepest
point.
2. Fallow Ores are submitted to the following operations:
-Raw smelting and roasting; smelting in cupola furnaces;
roasting and smelting in reverberatory furnaces for the pro-
duction of black copper; refining of the black copper.
3. Copper Lead Matt, containing from 25 to 30 per cent
of lead and antimony and from 11 to 14 per cent of copper,
is smelted in reverberatory furnaces in quantities of 1 ton
together with slags and metallic iron; the mass is strongly
stirred and tapped off after 12 hours. It yields some lead
and copper matt, which is roasted in reverberatory furnaces.
in quantities of 2 tons whilst alternately raising and lowering
the temperature, thus facilitating the volatilisaton of anti-
mony. The roasting mass, containing from 25 to 40 per
cent of copper, is smelted in reverberatory furnaces; the slag
is frequently removed, and towards the end of the process.
5 per cent of coal and 6 or 7 per cent of lime are added. The
PROCESS AT THE COPPER WORKS ON THE RHONE. 219
resulting black copper is refined. Io cwts. of roasting mass
are charged every 6 hours, and the black copper is tapped
off every 24 hours.
4. Old Copper, Black Copper (purchased), and Rosette
Copper are refined at once, 2 tons in 8 or 10 hours consuming
6 cwts. of coal.
5. Cannon Bronze containing 90 per cent of copper and
10 per cent of tin, is smelted in reverberatory furnaces,
and continually stirred, thus extracting the tin as an oxide,
which is skimmed off. The copper is ladled out when suffi-
ciently pure. An addition of common salt facilitates the
separation, volatilising a great deal of tin as chloride.
Owing to the high price of coal the treatment of ores by
the English process would be more expensive than the appli-
cation of the English and German methods combined.
Bogoslowsk's Copper Works in Siberia.-The ores treated
consist of calcareous pyrites and siliceous oxidised ores,
usually containing from 1 to 7 per cent, seldom as much
as 20 per cent of copper.
I
I
36 parts of sulphuretted ores and 34 parts of oxidised
ores, mixed with So parts of diorite, 20 parts of limestone,
14 parts of raw slag, 11 parts of black copper slag, and
I of furnace deposits, are smelted in cupola furnaces, 14 feet
high, provided with 4 tuyeres. (The new furnaces of Von
Rachette, with 24 tuyeres, possess great advantages with
regard to the consumption of fuel, the extent of production,
&c.). 144 cwts. of mixture are smelted in 24 hours, producing
20 cwts. 80 lbs. of raw matt containing from 30 to 40 per
cent of copper, and from 35 to 50 per cent of iron. The
raw matt is roasted with wood in quantities of 50 tons for
two or three days; it is then smelted in blast reverberatory
furnaces in 6 or 8 hours. The blast is now put in motion
whilst the mass is stirred with moist wooden poles; the slag
containing 67.3 per cent of silicates and 23 per cent of sul-
phides, is skimmed off and given to the raw smelting. A
further quantity of matt is now charged till the hearth con-
tains 5 tons, which are treated as before described whilst
adding rich slags from former operations. The mass is kept
in a fused state for three or four days, when solidified copper
220
COPPER.
matt will appear on its surface; the slag is now skimmed off
and an evolution of sulphurous acid will take place, causing a
boiling and spirting of the mass. The furnace is now allowed
to cool, and 10 cwts. of copper matt and 2 cwts of rich slag
are added. The temperature is again raised and the pro-
cess repeated until the hearth contains 184 cwts. of matt.
After having removed the slag a strong firing is given; after
some time the furnace is cooled and afterwards again heated,
and if the new slag forms a red crust a sample is taken out
with an iron ladle. The black copper produced is tapped
off, if the sample taken does not stick to the iron ladle after
immersion in water, and if its surface is convex and not
concave.
The poorer slags, containing from 2 to 10 per cent of
copper, are given to the raw smelting, the richer ones to the
black copper smelting. The black copper is refined in the
same furnace.
The total production of copper in Russia amounts to about
5,000 tons per annum.
PROCESSES
EMPLOYED IN SMELTING
OXIDISED ORES.
Oxidised copper ores are occasionally smelted by themselves,
but more usually given in admixture when treating sulphur-
etted ores and intermediate products. When treating them
by themselves for the production of black copper a small
addition of iron pyrites is made, sometimes, also, strong
bases, such as protoxide of iron or lime, are added to prevent
the scorification of too much copper. By this means, on
the other hand, the formation of iron deposits is facilitated.
Rich ores may be smelted in reverberatory furnaces if
coal is to be had cheap; poor ores must be treated in cupola
furnaces (page 11) in order to save fuel.
If the cupola
treatment is not profitable, the wet way is sometimes
advantageously resorted to, particularly when treating ores
with a siliceous gangue.
Oxidised ores are at once reduced to black copper
when intended for treatment in reverberatory furnaces. The
SMELTING OXIDISED ORES.
221
ores, in admixture with suitable fluxes and some charcoal
(or preferably with non-caking coal), and with slags rich in
copper, and residues, &c., are gradually smelted with a
reducing flame, then stirred, and again heated for the proper
separation of the different products. The slag is then
removed from the black copper. If the mixture contains a
sufficient quantity of oxide of iron and the process is carried
on long enough most of the iron enters the slags, whilst
part of the reduced metallic iron extracts scorified copper,
forming silicate of iron. Iron is more effective for this
extraction than the intermixed coal, which rises too easily
to the surface of the fused mass, whilst iron remains more
intimately mixed with it. A greater excess of oxide of iron
and too long a reducing action give rise to the formation
of a ferruginous black copper, which is difficult to refine,
whilst too little oxide of iron causes the formation of rich
slags and of a purer black copper; the black copper is re-
fined in the usual manner. The slags are either added to
the ore smelting, or, if obtained in larger quantity from poor
ores, they are smelted in cupola furnaces.
As cupola furnaces possess a stronger and more con-
tinuous reducing action than reverberatory furnaces, the
latter seem to be chiefly adapted for smelting copper ores
rich in iron, although, on the other hand, they cause more
expense for fuel, labour, &c.
The cupola furnaces employed for smelting oxidised ores
are modified in their dimensions and construction, according
to the amount of copper and iron which the ores contain.
As the fuel employed also has a reducing action, more is
consumed for smelting oxidised ores than when treating
sulphuretted ores.
Illustrations of Smelting Oxidised Copper Ores.
Former Process at Chessy* for Smelting Rich Ores,† such as
malachite, red copper ore, &c., associated with calamine,
MARGERIN, in Ann. de min., 2 sér., vii., 293. GUENIVAU und TRIBAUT, in
KARSTEN'S Archiv. 1, R. vi., 310, 375; xviii., 183. LAMPAD., Fortschr., 1839,
P. 137. BERTHEIR'S anal. met. Chemie, deutch v. KERSTEN, ii., 404. RIVOT,
Métallurgie du Cuivre, 1859, p. 75.
*
↑ Erzvorkommen: COTTA's Erzlagerstätten, ii., 421.
222
COPPER.
hydrated peroxide of iron, and silicate of alumina; they
are smelted with coke in a small sump furnace 5 feet high,
in admixture with 20 per cent of lime and 50 per cent of
slags. The mixture, containing from 27 to 30 per cent of
copper, produces in 24 hours 14 cwts. of black copper, con-
taining about 90 per cent of copper, and some little matt
caused by the sulphur contained in the coke.
The black copper is refined in oval blast reverberatory
furnaces, in quantities of 60 cwts., in 16 or 17 hours, at a
consumption of 36 cwts. of coal, and yielding about 50 cwts.
of refined copper.
At Bouc,* on the Rhone, a similar process is carried on
for smelting oxidised ores from Spain, Asia Minor, and
Africa.
Treatment of Poor Ores at Permt.-These orest are mala-
chite, vanadiate of copper, oxide of copper, hydrated silicate
of copper, sometimes native copper, and seldom copper
pyrites, disseminated in sandstone, and containing on an
average 3 per cent of copper. As they are chiefly associated
with ferruginous loam and quartz, they are smelted in ad-
mixture with 30 per cent of dolomite, rich refinery slags,
and 20 or 25 per cent of slags of the same process. The
smelting is effected in cupola furnaces of medium height.
From 3.684 to 4°502 tons of ore are smelted in 24 hours,
and 1 ton of ore with 0.3 ton of dolomite, consumes o‘7 ton of
coal, and yields o'032 ton of black copper, containing o'024
ton of fine copper. The cost per ton of ore for smelting and
materials is only IIS. 4d., owing to the low state of wages
and the cheapness of fuel.||
As the greater part of the iron reduced by the process
becomes carbonised it separates from the copper as cast iron,
while iron free from carbon alloys with the copper.
* B. u. h. Ztg., 1859, p. 243.
CHOUBINE, Ann. de min., 1843. Bgwkfd., vii., 431; ix., 159; xiv. 605
xix., 773.
Rivor, Métallurgie du Cuivre, 1859, p. 86. PERCY, Metallurgy,
i., 443. LEONHARDT, Hüttenerzeugnisse, 1858, p. 108.
+ COTTA, Erzlagerstätten, ii., 548.
|| Oesterr. Ztschr., 1858, No. 51.
TREATMENT OF POOR ORES AT PERM.
223
The metallic products are tapped off every 24 hours, while
most of the slag, being a bi-silicate containing 54 or 56 per
cent of silica, and poor in copper, runs out of the fore
hearth. The slag remaining in the hearth being impure is
removed in one piece, and the cupriferous raw iron is wetted
with water, and formed into cakes, which are removed and
broken. The black copper is ladled into moulds. By
increasing the dimensions of the furnaces a larger produc-
tion is obtained and less fuel consumed.
The cupriferous raw iron contains 3 per cent of copper
during the regular process, but is richer at the end of the
operation; according to Choubine it contains-
Fe.
Cu.
Vd.
Al .
Mg.
Si
C
75'97
12 64
I'99
0.89
0.78
2'51
3°03
It is submitted to an oxidising smelting on a small hearth,
yielding black copper and a slag which consists chiefly of
6FeO,SiO,, containing per cent of copper; 992 lbs. of
cupriferous iron yield in 24 hours 147 lbs. of black copper
and II cwts. of slags. The slags are smelted in the ore fur-
naces together with impure ore and refinery slags, for the
production of cupriferous iron deposits containing less than
30 per cent of copper and slags of the composition of
3RO,SiO3. The iron deposits are treated as cupriferous
raw iron, and, according to Choubine, they are composed as
follows:-
Fe.
Cu.
Vd.
Al .
Si
C
Slag
76.30
19'90
0'12
0'43
0.83
0'73
3'33
According to Rivot, the cupriferous raw iron is re-melted
in cupola furnaces two metres high; the fused mass produced
224
COPPER.
in 24 hours from 1 ton of raw iron at a consumption of
Io cwts. of coal, is kept quiet for an hour, and then the
separated black copper is tapped off into an outside basin,
whilst the raw iron is made to run into foundry pans. The
black copper contains about 80 per cent of copper, and is
ferruginous; the raw iron is employed for foundry purposes.
The cost of smelting I ton of cupriferous raw iron is
7s. 6d. The annual production, from 155 tons of cupri-
ferous raw iron, is about 9 tons of black copper and 145 tons
of foundry iron.
The purifying of the black copper is effected in the furnaces
represented by Figs. 58 and 59. a is the tuyere; b, tapping
FIG. 58.

A
g
FIG. 59.
е
d
www.
a.
Ъ
་་་
C
-D

C
B
C
TREATMENT OF POOR ORES AT PERM.
225
holes; c, outside basins; d, working door; e, opening for
charging the fuel; f, grate; g, hearth formed of a mixture of
clay and coal dust, provided with a slight sloping towards b.
64 cwts. of black copper are heated for about six hours,
when the melting begins, and on continued firing they are
melted in about six hours more, the blast being put on
as soon as the copper becomes red-hot. After all the
copper is melted, the blast is put off, and 8 cwts. of pyritic
ores with a siliceous flux are mixed with the slag on the sur-
face of the metal, and strongly heated for about three hours,
till they fuse. Thus the oxidised copper contained in the
slags becomes sulphurised, and the sulphide of copper
combines with the copper, while the sulphide of iron enters
into combination with the slags. The slags are removed
from the metal bath; the blast is conducted upon the
copper, and the slag again removed; the colour and nature
of the slag, as well as the escape of sulphurous acid, indicate
the approaching fineness of the copper, which is determined
by samples taken. About six hours after putting on the blast
again the refined copper is tapped off into the outside basins
and ladled from them into cast-iron moulds. One ton of
black copper, consuming 1937 tons of wood and o*125 ton
of pyrites, yields o'840 ton of refined copper, at a cost
of 6s. 5d.
The toughening of the refined copper is performed in small
hearths; about 7 cwts. are refined in three or four hours. A
little chloride of mercury is occasionally added to render the
copper more dense. One ton of refined copper yields
o'92 ton of tough copper, at a consumption of o'685 ton of
charcoal, and at a cost of 9s. 2d.
One ton of ore yields o'0262 ton of tough copper and
0*023 ton of foundry iron, consuming 0*7315 ton of charcoal
and o'066 ton of wood, at a total cost of 12s. Though
the consumption of fuel and flux is considerable, and much
labour is required, the cost is very low, as materials and
labour are cheap.*
Metallproduction am Ural in 1859. B. u. h. Ztg., 1862, pp. 157, 362.
VOL. II.
Q
226
COPPER.
SMELTING NATIVE COPPER.
Of all copper ores native copper is the most seldom smelted
by itself. If a large production is possible, and good coal
cheap, the employment of reverberatory furnaces is advisable,
as they yield tough copper at once; in other cases cupola
furnaces may be more profitable. When smelting in rever-
beratory furnaces, the copper contained in the slags swim-
ming on the metal bath, may be partly extracted by mixing
the slags with coal; but in this way a great deal of iron is
also likely to be reduced and alloy the copper. The most effec-
tive plan is to place some iron bars horizontally into the slags
without allowing them to come into contact with the copper,
thus causing a reduction of the scorified suboxide of copper
by means of the iron, without letting the iron enter the copper.
The copper works in Detroit** smelt the native copper which
is found on the southern shore of Lake Superior in the
United States.
In extracting the native copper from the huge veins of that
region a formidable obstacle is to be encountered in the size
and solidity of the masses in which the metal occurs. A
hundred tons are not unfrequently found in a single piece,
and this must of course be divided into manageable blocks
before it can be raised, transported to the shore, and shipped.
These blocks are much larger than formerly, and are some-
times 8,000 or 9,000 lbs. in weight. The usual plan adopted
is to dislodge the mass from the wall by heavy charges of
powder, but this agent is inadequate to a further division, since
the tamping, however excellent, is blown from the hole in
the same manner as a ball is driven from a cannon. Long
slender steel chisels, having a cutting edge about a quarter
of an inch in length, are managed by two men, one to direct
the tool while the other strikes with a heavy hammer. By
this means narrow channels are cut, chip after chip, through
the mass, a process requiring much skill, and at best ex-
ceedingly slow, thirty months of constant work being some-
times required to cut a single mass.
*
Rivor, Métallurgie du Cuivre, 1859, p. 48. B. u. h. Ztg., 1856, p. 341 ;
1859, p. 455. LAMBORN, Metallurgy of Copper, 1860, p. 172.
SMELTING NATIVE COPPER.
227
The product, in its various forms of massive copper, barrel
work, or stamp work, is transported by lake, river, and rail-
way to Detroit, Cleaveland, and Pittsburg, to be smelted.
The advantage secured by this transport is found in the cir-
cumstance that it can be reduced in the immediate vicinity
of the vast coal deposits of Ohio and Pennsylvania, and
fuel can thus be obtained at the works in large quantities at
a low price. Some years ago, however, a smelting works
was started in the copper region, on the site of an old furnace
near the Bruce mine; it is supplied with fuel brought as a
return load by vessels which carry ore and metal into
the coal regions near Lake Erie. Beside these there are
three furnaces in the Union, located respectively at Balti-
more, Boston, and New Haven, which smelt the produce of
Lake Superior as well as of other home mines, and also
import ores from South America, the Pacific islands, &c.
At Detroit, the larger pieces of ore are smelted in charges
of from 4 to 5 tons, employing reverberatory furnaces and a
reducing flame. After about 16 hours the liquid slag formed,
containing about 4 per cent of copper as suboxide, and from
3 to 6 per cent in the form of grains, is removed, and air is
admitted in the furnace in order to form a sufficient quantity
of suboxide in the copper.
The suboxide of copper soon oxidises the carbon contained
in the copper; it is formed in excess, and must be reduced
in the usual manner by means of wooden poles.
One charge is finished in 24 hours and yields 20 per cent
of slags and 76 per cent of copper; or 5 tons of ore yield
3'5 tons of copper, at a consumption of 2 tons of coal.
I ton of ore, yielding o‘76 ton of copper, and o‘2 ton of slags,
costs £I IS. Id. for smelting.
Smaller fragments of native copper nearly liberated from
the gangue by means of pounding stamps are either smelted
together with the large pieces or by themselves, and in the
same manner as the former. The smelting takes a little
more time, the copper combines with more carbon, less slag
results, and the refining and toughening of a charge of
8 tons take 2 days longer and consume about 5 tons of
coal. I ton yields o'g ton of tough copper and o'11 ton of
Q 2
228
COPPER.
slags, costing £1 9s. 10d. As the small fragments frequently
contain a large amount of silver which may be profitably
extracted, it is advisable to smelt them separately.
The very small dressing products are smelted in charges of
3 tons in 7 hours, the slags are removed, and a new charge
smelted again and separated from the formed slags. The
copper of both charges is then refined and toughened. The
process is finished in 24 hours, and I ton of ore yields o˚5
ton of copper and o˚5 ton of slags, consuming o'5 ton of coal,
at a smelting cost of £1 2s. 4d. The smelting cost of the
preceding three sorts of ore per ton of refined copper amounts
respectively to £1 16s. Id., £1 13s. 3d., and £2 4s. 9d.
The slags resulting from the former processes in reverbe-
ratory furnaces are smelted with anthracite, without any
addition of fluxes, in cylindrical cupola furnaces, 3 metres
high and I metre wide, for the production of black copper.
The furnace has three tuyeres, and a pressure of blast is em-
ployed corresponding to the density of the fuel. IO tons of
slags are smelted in 24 hours, producing 1 ton of black
copper, containing from 5 to 8 per cent of iron at a con-
sumption of 275 tons of anthracite and o'4 ton of charcoal,
causing an expense of £8. The resulting slags are thrown
away.
The black copper is refined in charges of 4 tons in
24 hours, yielding 75 per cent of tough copper and 30 per
cent of slags, which are added to the smelting in cupola
furnaces.
The smelting cost of 1 ton of slag, producing o'085 ton of
tough copper, amounts to 18s. 4d., whilst o*282 ton of anthra-
cite, 0*042 ton of charcoal, and o*112 ton of coal are con-
sumed. Including the yield of the slags of the reverberatory
furnaces I ton of native copper yields o'77 ton of tough
copper, at a consumption of 0'522 ton of coal, o'025 ton of
wood and charcoal, and o'056 ton of anthracite, and at a
cost of £1 4s. 7d. I ton of schlich yields o˚55 ton of tough
copper, consuming o‘556 ton of coal, o‘051 ton of wood and
charcoal, and o'141 ton of anthracite, and causing an expense
of £1 IIS. 6d. To these sums the general cost must be
added, which is frequently considerable.
EXTRACTION OF COPPER IN THE WET WAY.
229
At the French copper works,* Corro-Corro ores from
South America are smelted and treated similarly.
At Grünthal in Saxony, native copper is worked in cupola
furnaces. The ores consist of Corro-Corro ores (page 3) and
of cuprobarillas which are sold in canvas bags holding 75 lbs.
weight, under the name of sand ores. These ores seem to be
the product of dressing operations, and consist of native
copper containing a little malachite, red copper ore, azurite,
and atacamite, associated with quartz and felspar, and con-
taining from 73 to 75 per cent of copper. 50 cwts. of these
ores are smelted, for the production of black copper, in ad-
mixture with 15 cwts. of very siliceous old slags of Katharinen-
berg in Bohemia, which are rich in earths but poor in copper
(per cent), o'85 cwts. of fluor spar, and 6 cwts. of slags
of the same process. The slags thus resulting are so pasty
that they must be removed from the slag-gutter; this pasti-
ness is probably owing to the great amount of silica in the
old slags.
The smelting is effected in sump furnaces, 6 feet high.
The resulting black copper is tapped off every 1 hours into
cast-iron moulds, forming ingots 18 inches in diameter, from
I to 2 inches thick, and from 1 to 1 cwts. in weight.
41*25 cwts. of sand ores are smelted in 24 hours in 253 charges
each of 23 lbs. and 12 lbs. of charcoal. The ore yields
70 per cent of black copper containing 95 per cent of fine
copper, and slags containing 6 or 8 per cent of copper, which
is chiefly inclosed mechanically. The slags are partly
returned to the ore smelting, and partly kept for a separate
smelting. The black copper is refined in reverberatory
furnaces, yielding 90 per cent of fine copper, including that
resulting from the slags. The refinery furnace has the
capacity of 45 or 50 cwts. and is heated by lignite or coal of
inferior quality.
EXTRACTION OF COPPER IN THE WET WAY.
This mode of extraction is chosen, if the dry way is less
economical and profitable, owing to the poorness of the ores
Rivor, Métallurgie du Cuivre, 1859. p. 36.
230
COPPER.
and to other unfavourable circumstances with regard to
foreign intermixtures. The ores best fitted for the process
in the wet way are poor oxidised ores, associated with siliceous.
substances, which can only be smelted at a considerable con-
sumption of fuel, and which would also cause, mechanically,
great loss of metal in the dressing and in the smelting by
scorification. Sometimes, also, poor sulphuretted ores may
be treated advantageously in the wet way, if they are either
intimately admixed with other sulphides, or so finely dis-
seminated in the gangue, as to be impossible to dress and
unprofitable to smelt, owing to the enormous consumption of
fuel and the loss by scorification to which the required compli-
cated processes in the dry way would give rise. These poor
ores may yet be smelted advantageously, provided fuel is
cheap, if the admixed earths are present in such proportions
as to give a good slag without any further addition (Mansfeld
copper slate), or if they contain quartz and iron pyrites
(Fahlun, Röraas), as the pyrites when roasted yields the
base for the silica. If these ores contain an excess of quartz
they are only profitable when treated in the wet way. It is
of essential influence for the ores not to contain much sub-
stance soluble in acids, such as lime or sparry iron ore, and
therefore oxidised ores with insoluble gangue, and containing
per cent of copper, may be treated with greater advantage
than ores containing 2 per cent of copper and a small per
centage of lime, as for instance cupriferous sand stone of
the Eifel* and at Twistet in Waldeck.
The problem of extracting copper by means of acids from
calcareous ores is not yet perfectly solved; experiments
made for this purpose aim at the employment of cheap re-
agents for its extraction, such as sulphate and chloride of
iron, &c., or the removal of the lime by means of acids
previous to the treatment of the ores.
Bischoff and Porth|| have made suggestions on this point.
worthy of notice. Sometimes the wet way is employed for
*
+
JUNG, in B. u. h. Ztg., 1862, p. 229.
† B. u. h. Ztg., 1859, p. 412; 1860, p. 27; 1862, p. 191,
Oesterr. Ztschr., 1860, No. 5. B. u. h. Ztg., 1860, p. 419.
|| STANM, neuste Erfindungen, 1862, No. 1.
HYDRO-METALLURGICAL EXTRACTION OF COPPER.
231
treating rich ores and products, mostly if silver is to be
extracted at the same time.
The following are the chief operations of the hydro-
metallurgical extraction of copper:-
The Transformation of the Copper into a Soluble State.
The materials employed for this purpose are-
1. Water, if the ores contain sulphate of copper, which
is formed, for instance, by allowing iron pyrites containing
These ores will
copper pyrites to become weather worn.*
become the more easily decayed the more sulphide of iron
they contain and the less compact they are. Copper pyrites
is decomposed with more difficulty.
This decaying takes place chiefly in the mines at Schmöll-
nitz,† Wicklow,‡ Maidenpek,|| Riotinto,§ Isle of Anglesea,¶
Rammelsberg,** &c., by the natural action of warmth,
humidity, and air; sometimes they are allowed to decay in the
open air, as at Schmöllnitz,++ Riotinto,‡‡ and Rammelsberg.
When decaying, an admixture is formed of iron pyrites,
copper pyrites, arsenio-sulphide of cobalt and nickel (cobalt
ores of Siegen), the sulphide of iron is next decomposed :—
FeS₂+70 FeO,SO, + SO,; and
2
Fe₂S₂ + 110 = 2FeO,SO3 + SO3•
3
An admission of air transforms the protoxide of iron into
a basic salt of the peroxide, and affects the chloride of iron
more easily than protosulphate of iron. Chloride of calcium
is therefore sometimes added to the decaying ore :—
2FeO,SO₂+0= Fe₂O¸‚SO₂+SO3.
* HAUCH, in Oesterr. Ztschr., 1860, p. 590. STROMEYER, in B. u. h. Ztg.,
1856, p. 219. JUNG, in B. u. h. Ztg., 1861, p. 394.
+ HAUCH, in Oesterr. Ztschr., 1853, p. 119; 1860, p. 273. B. u. h. Ztg.,
1861, p. 326.
**
B. u. h. Ztg., 1858, p. 25.
|| Ibid., 1857, P. 3.
Ibid., 1861, pp. 287, 302.
DUMAS, Angew. Chemie, iv., 230.
KERL, der Commun., Unterharz, 1853, p. 21.
++ Oesterr. Ztschr., 1860, p. 274.
++ B. u. h. Ztg., 1861, p. 303.
Ibid., 1861, p. 394.
232
COPPER.
The formation of sulphate of copper is then facilitated by
the free SO, :—
Cu₂S+50+SO3=2CuO,SO3.
Finally, CoS2, NiS2, CoAs₂, and NiAs, become oxidised also
by the action of the free sulphuric acid, forming sulphates
and arseniates:-
(Co,Ni)S₂+70=(Co,Ni)O,SO3+SO3.
(Co,Ni)As₂+11O=(Ni,Co)O,AsO5+AsO5.
The latter salts are transformed by basic sulphate of iron
thus:-
(Co, Ni)O,AsO5+2Fe2O3,SO3 = (Co,Ni)O,SO3+
Fe₂O3,AsO5+ Fe₂O3.
Upon these reactions W. Jung has based a mode of ex-
tracting in the wet way copper, cobalt, and nickel from the
cobalt ores of Siegen.
If it is important to extract the copper more perfectly and
quickly, the ores are roasted and afterwards lixiviated.
Ores and matts are roasted either in a pulverised state in
reverberatory furnaces (Bankart's and Bischof's method*),
in mounds at Schmöllnitzt and Mühlbach,‡ in open heaps.
at Schmöllnitz || and Riotinto, § sometimes at a kernel roasting.
(Argodo¶ and Wicklow**); cupola furnaces, as employed at
Linz by Gossage's method,++ are seldom used. Markus‡‡
recommends that copper slags containing sulphide of copper
should be roasted and lixiviated. Guerngross treated copper-
lead matt of the Smjeffssky's smelting works;|||| Newton§§
suggested a similar mode for auriferous and argentiferous
**
B. u. h. Ztg., 1852. p. 354; 1856, p. 219; 1862, p. 328. PERCY, Metallurgy,
Oesterr. Ztschr., 1860, 310.
i., 447.
† Oesterr. Ztschr., 1861, pp. 41, 57.
Ibid., 1861, pp. 41, 51.
|| Ibid., 1860, p. 327.
B. u. h. Ztg., 1861, p. 289.
Oesterr. Ztschr., 1860, p. 334. B. u. h. Ztg., 1856, pp. 210, 227; 1860,
p. 439; 1862, p. 129.
Bgwkfd., xix., Nos. 12,
** B. u. h. Ztg., 1858, p. 25.
tt Ibid., 1860, p. 256.
ㅏㅏ
​Oesterr. Ztschr., 1857, p. 323.
B. u. h. Ztg., 1852, p. 503.
AS Ibid., 1845, p. 1972.
17.
i
HYDRO-METALLURGICAL EXTRACTION OF COPPER. 233
copper matt.
In Cornwall* copper is extracted by means
of iron from the water used for washing roasted pyritic tin
ores. Daehnet transforms the oxide of copper contained in
roasted pyrites into sulphates by heating it with sulphate of
iron.
As it is very difficult to produce by the roasting process
sulphates only, and as a great deal of oxide of copper inso-
luble in water will be formed if the ores do not contain
sufficient iron pyrites, or if the temperature has been some-
what too high, it is in most cases better to roast the sul-
phuretted ores as perfectly as possible, and to treat the
formed oxide of copper with acids or other solvents which
usually but slightly attack the sulphides still present.
The use of one or the other acid chiefly depends on the
local price.
2. Dilute Muriatic Acid or Sulphuric Acid.-Sulphuric
acid is seldom employed in a concentrated state, as it may
then form insoluble basic sulphates of iron, which will con-
taminate the cement copper. Dilute sulphuric acid has
the advantage of easily dissolving the oxide of copper con-
tained in roasted sulphuretted ores, but it dissolves with
difficulty the peroxide of iron, and this again is more readily
dissolved in muriatic acid; also a saleable by-product,
iron-vitriol, results from this treatment. According to Mit-
scherlich, an excess of concentrated sulphuric acid also
strongly attacks peroxide of iron. Muriatic acid is more
generally employed than sulphuric acid, as it is usually
cheaper, and even frequently a troublesome by-product of
the soda manufacture.
a. Oxidised Ores are treated with sulphuric acid (for
instance, in Marseilles ||) for the conversion of pure malachite
into copper vitriol. Muriatic acid is employed at the copper
works at Linz on the Rhine, at Stadtbergen¶ in Westphalia,
*
B. u. h. Ztg., 1859, p. 207; 1862, p. 145.
+ Polyt. Centr., 1861, p. 1020.
B. u. h. Ztg., 1961, p. 364.
|| Ibid., 1859, p. 243.
RIVOT'S
§ Berggeist, 1858, pp. 566, 662. B. u. h. Ztg., 1860, p. 191.
Kupferhüttenkunde, Deutch v. HARTMANN, 1860, pp. 275, 280, 284, 301.
¶ Bgwkfd., xix., 232, 262.
234
COPPER.
at Twiste in Waldeck, at Friedrich-Wilhelms-Hüttet near
Commern, and at Alderley Edge,*‡ for the treatment of
arseniate, phosphate, carbonate, and pure oxide of copper.
Calcareous ores containing o'9 per cent of copper and 4 per
cent of lime cannot be profitably treated by this process.
Red copper ore is best exposed for some time to the atmo-
spheric air after being soaked with acid, by which treatment
a readily soluble salt of oxide of copper will be formed.
Siliceous malachite and phosphate of copper are with more
difficulty attacked by dilute sulphuric and muriatic acids
than other copper salts, and ores containing these combina-
tions yield residues rich in copper. When employing muri-
atic acid the solubility of the oxide of copper is facilitated
by an addition of common salt, thus forming easily soluble
double soda salts.
b. Roasted Ores and Products are more frequently treated
with sulphuric acid, in order to retain the greater part
of the formed peroxide of iron in the residues. For instance,
the treatment of the residues of the black copper amal-
gamation with sulphuric acid by Lewis and Roberts's ||
method at Schmöllnitz§ leaves oxide of iron, antimonial
oxides, and noble metals in the residue; a similar treatment
of roasted argentiferous copper concentration matt is adopted
in Freiberg.
Ores are treated with muriatic acid at Linz, where they
are roasted by Escalle's process,** which consists in roasting
them first by themselves, and afterwards with an addition of
muriatic acid.tt By treating the ores roasted in this manner,
chloride of copper is formed, and chlorides of antimony and
* B. u. h. Ztg., 1859, p. 412; 1860, p. 37, 111; 1862, p. 191. Berggeist,
1859, No. 48; 1861, No. 55.
† B. u. h. Ztg., 1862, pp. 140, 229.
Schles. Wochenschrift, 1861, No. 2. Oesterr. Ztschr., 1860, p. 30. B. u. h.
Ztg., 1862, p. 18.
|| B. u. h. Ztg., 1858, p. 217. Oesterr. Ztschr., 1858, p. 248. Berggeist,
1858, No. II.
§ Oesterr. Ztschr., 1859, p. 331; 1860, p. 351.
¶ LORENZ, Stammbaum zu den Freiberger Hüttenprocessen, 1861. LAM-
BORN, Metallurgy of Copper, 1860, p. 197.
** B. u. h. Ztg., 1859, p. 243. PERCY, Metallurgy, i., 450.
Oesterr. Ztschr., 1860, p. 341.
HYDRO-METALLURGICAL EXTRACTION OF COPPER. 235
arsenic volatilised. The extracted copper is then precipitated
by sulphide of calcium.
Triplier's method,* i. e., roasting of the fallow ore with or
without the presence of steam, extraction of the copper
by means of muriatic acid, &c., is sometimes used, also
Spence's method,* i. e., roasting the ore in a reverberatory
furnace,† and treatment of the roasting mass with muriatic
acid in which soda-saltpetre is dissolved.
Undecomposed sulphides in the roasting mass are not acted
upon by dilute acids, but if they are moistened in a finely
pulverised state with concentrated muriatic acid, and exposed
for some weeks to the atmospheric air, the sulphide becomes
perfectly oxidised. Sulphuric acid reacts much more slowly.||
At Schemnitz§ the copper is extracted by dilute sulphuric
acid at the same time when desilverising raw matt by
Ziervogel's method.
c. Black and Refined Copper.-Granulated copper is treated
with dilute sulphuric acid with an admission of air
at Mansfeld¶T for the extraction of nickel, and at Oker,**
Fahlun, and Moldova for the extraction of gold and silver.
Sometimes also old copper is converted into copper vitriol
by this method,tt and sometimes the copper is roasted with
2 or 3 per cent of common salt‡‡ previous to its solution in
sulphuric acid.
3. Sulphurous and Muriatic Acid Gas together with
Steam.—According to Clement's method, || || the muriatic acid
gas produced, when employing cupriferous iron pyrites in the
manufacture of sulphuric acid, and using the sulphuric acid
from the same manufactory for the production of soda, is
* B. u. h. Ztg., 1852, P. 544.
B. u. h. Ztg., 1862, p. 148.
+
Oesterr. Ztschr., 1857, p. 279.
DINGLER'S Polyt. Journ., Bd. 164, p. 449.
B. u. h. Ztg., 1862, p. 175.
§ Bericht uber d. zweite Versammlung v. Berg. u. Hüttenmännern in Wien.,
1862, p. 90.
B. u. h. Ztg., 1859, p. 371; 1861, pp. 67, 471.
Ibid., 1859, p. 165.
++ Ibid., 1860, p. 39.
tt Oesterr. Ztschr., 1857, pp. 282, 289.
++
Berggeist, 1858, p. 203. Polyt. Centr., 1857, No. 15.
236
COPPER.
made to react upon the roasted pyrites together with steam.
More often oxidised ores, as in Stadtbergen,* or roasted
pyritic ores, as formerly in Linz,t are acted upon direct
by sulphurous acid gas and steam. Birkmeyer‡ makes sul-
phurous acid gas obtained by roasting iron pyrites react
upon finely ground raw copper pyrites.
At Stadtbergen|| and Linz muriatic acid is at present em-
ployed instead of sulphurous acid gas, and with the following
advantages:-Simpler and cheaper apparatus, independence
of working poor copper ores without the simultaneous working
of other sulphuretted ores, and the possibility of carrying out
all required operations at an ordinary temperature. On the
other hand, roasted ores are more perfectly decomposed by
sulphuric acid gas than by muriatic acid at a common tem-
perature, and the resulting lixiviums contain iron vitriol as
a saleable by-product.
4. Acid Fluids containing Sulphate of Iron.-According
to Stromeyer, § sulphate of iron transforms metallic, oxidised,
and sulphuretted copper into sulphate:
Fe₂O3,3SO3+Cu=2FeO,SO,+CuO,SO¸.
Fe₂O3,3SO₂+Cu₂O=2CuO,SO3+Fe₂O3,SO3.
Fe₂O3,3SO₂+CuO=CuO,SO₂+Fe₂O3,2SO3.
2(Fe₂O3,3SO3)+Cu₂S=4FeO,SO3+2CuO,SO3+S.
Sulphates of iron and alumina also have a solvent action
upon oxide of copper :-
2FeO,SO¸+CuO+0=CuO,SO3+Fe2O3,SO3.
Al₂O3,3SO3+3CuO+3HO=3CuO,SO3+Al2O3,3HO.
5. Chlorination without employing Dilute Muriatic
Acid. Although muriatic acid is the cheapest, its transport
is very inconvenient; experiments have therefore been made
to replace it by other substances, dry or liquid, which contain
* Ann. d. min., 1842, livr. i., p. 1842. Bgwkfd., vi., 417. B. u. h. Ztg.,
1856, p. 218.
Berggeist, 1858, p. 566; 1859, No. 4. B. u. h. Ztg., 1859, pp. 107, 223,
438; 1860, pp. 27, 191. Oesterr. Ztschr., 1857, p. 277; 1860, p. 351.
Bgwkfd., xii., 669.
Ibid., xix., 232, 262.
§ B. u. h. Ztg., 1856, p. 226.
Oesterr. Ztschr., 1860, 293.
CHLORINATION IN THE WET WAY.
237
chlorine in a more concentrated form. These substances
are common salt, chloride and perchloride of iron, which
also allow the extraction of calcareous ores and of ores con-
taining iron spar.
H. Meyer* determined the reaction of chloride and per-
chloride of iron upon the basic substances with which oxi-
dised copper ores are usually associated, as well as the cir-
cumstances which take place at the per-oxidation of the
chloride of iron in the presence of air, and either with or
without chloride of copper, and the loss of soluble muriatic
acid by the separation of basic salts.
Perchloride of iron reacting upon carbonate of lime forms
chloride of calcium and hydrated peroxide of iron containing
some muriatic acid, and it decomposes carbonate of copper,
air being excluded, as follows:-
4FeCl+3(2CuO,CO₂)=2Cu₂C1+2CuCl+2Fe₂O3+3CO₂.
When oxidating chloride of iron by the action of the air,
6FeC1+30 will be transformed into 2Fe,Cl,+Fe₂O¸•
3
One equivalent of perchloride of iron decomposes three
equivalents of carbonate of lime :-
Fe,Cl,+3CaO,CO₂ = 3CaCl + Fe₂O₂+3CO₂,
and it is transformed with carbonate of copper thus:—
3(2CuO,CO₂)+aq.+2Fe₂Cl¸=6CuCl+2Fe₂O3+3CO₂+aq.
Chlorination in the Wet Way.
a. By Chloride and Perchloride of Iron.
Muehlhent thoroughly roasts pyritic copper ores con-
taining iron spar, and from 1 to 4 per cent of copper, in
reverberatory furnaces. At the end of the process he adds
some small coal, charcoal, or saw-dust, to reduce the
oxide of copper to metallic copper, and he extracts the copper
by means of chloride of iron obtained by precipitating the
copper by means of iron, after having separated basic iron
salts from the lixivium. Gossage treats roasted copper
pyrites poor in copper first with water and afterwards with
chloride of iron and muriatic acid.
* B. u. h. Ztg., 1862, p. 182.
† Berggeist, 1860, No. 3. B. u. h. Ztg., 1860, F. 439.
+
B. u. h. Ztg., 1860, p. 256.
238
COPPER.
b. By a Solution of Common Salt.
Richardson* moistens finely ground pyrites with 10 per
cent of common salt in solution or with sea-water, spreads
out the mass, and frequently turns it over without allowing
the temperature to fall below 27°, and also preventing its
rising too high. After the mass has become dry, salt solu-
tion is again added, and finally the mass is lixiviated.
Sal-ammoniact reacts like common salt.
Chlorination in the Dry Way by Roasting with
Common Salt.
Common salt was formerly used, for instance, by Orschall,
for the extraction of copper from its ores. Longmaid!
roasted Cornish pyrites with common salt, and with this
solution extracted copper and silver. Maumené roasts.
argentiferous and cupriferous pyrites with rock salt, extracts
the soluble copper salts by means of water, and the chloride
of silver in the residue by means of salt solution. Both the
silver and copper are precipitated from their solutions by
iron; the precipitate is then calcined, and the oxide of copper
extracted by sulphuric acid. The chlorination method is
patented in England and North America by Haehners, and
it has been recently introduced chiefly by Becchi and Haupt T
into the copper works of Capanne Vecchia, near Massa Mari-
tima, in Tuscany.
Copper pyrites containing about 2 per cent of copper and
siliceous gangue is roasted in heaps, afterwards ground and
roasted in a reverberatory furnace for about three hours, and
then from 2 to 8 per cent of common salt is added, according
to the amount of copper, and roasted for about twelve minutes
more. The oxide of copper is thus supposed, without being
decomposed by the salt, to be converted in the presence of
* B. u. h. Ztg., 1861, p. 8.
† DINGLER'S Polyt. Journ., Bd. 164, p. 185.
TUNNER'S Jahrb., 1852, p. 157.
|| Bgwkfd., xv., 688.
§ B. u. h. Ztg., 1858, p. 89.
¶ Ibid., 1857, pp. 34, 88, 182. PERCY, Metallurgy, i., 451.
CHLORINATION IN THE DRY WAY.
239
steam chiefly into oxychloride, which is extracted by dilute
sulphuric or muriatic acid.
Petitgaud* denies this theory, and supposes a direct trans-
formation of oxide of copper and common salt into chloride
of copper and soda, the soda combining with quartz. Grunert
suggests that only a small amount of chlorides and chiefly
sulphates are formed by this process; the sulphates may be
formed by a well conducted roasting without the addition of
salt. Bischoff entertains almost the same opinion, and
recommends Bankart's method (page 232) which he had fol-
lowed at Toplice,|| near Laak, in Upper Krain, with an addi-
tion of iron pyrites. Gurlts advocates Becchi's method in
opposition to Bischof's, but for the formation of chloride of
copper he does not consider it necessary to carry on the
roasting very far, as otherwise a prevailing amount of inso-
luble subchloride of copper is formed. Plattner has shown
that at an imperfect roasting Cu₂S and Fe₂S are formed,
which, with an addition of common salt and admission
of air, are transformed into metallic chlorides as, first, sul-
phurous acid, and then sulphuric acid are formed. The
sulphuric acid decomposes the common salt, liberating the
chlorine which transforms the remaining sulphides into
chlorides. If, however, the roasting is carried on as far as
Gruner's method requires, so as to form as much sulphates
as possible, then, according to Gurlt, so much chloride of
copper will be volatilised at the subsequent chlorination that
but little insoluble subchloride of copper will remain. Hen-
derson has proposed to volatilise all the copper as chloride,
and to condense the chloride.
Schaffner's experiments made at Eppichnellen in Thuringia
concerning the chlorination of copper ores have thrown some
light on these partly opposite theories, and have nearly
upset Gurlt's views of the formation of subchloride of copper
*B. u. h. Ztg., 1858, p. 90.
† Ibid., p. 325. Berggeist, 1858, No. 35.
++
Oesterr. Ztschr., 1860, Nos. 14, 15. Berggeist, 1860, No. 40.
|| Berggeist, 1860, No. 40. B. u. h. Ztg., 1861, p. 39; 1862, p. 129.
Berggeist, Nos. 3, 31, 47, 48. B. u. h. Ztg., 1860, p. 438; 1861, p. 39.
PERCY, Metallurgy, i., 451.
240
COPPER.
and the volatilisation of much chloride of copper if too many
sulphates are formed.
If, according to Becchi's method, the ore was roasted so
far that on lixiviation it showed no more sulphates, and
common salt was then added, 3 lbs. to every 1lb. of copper
contained in the ore, and roasted for a short time longer,
the copper might without any particular difficulty be
lixiviated so perfectly that the residues only contained
o'2 per cent of copper. A considerable improvement seemed
to be the formation of as much sulphates as possible by the
first roasting at a low temperature, and afterwards an addi-
tion of common salt, causing the chlorination to be per-
formed in a much shorter time than in the former instance.
In order to ascertain the proper time for the addition of
salt, samples were taken from time to time and lixiviated with
an addition of ammonia. As soon as the ammonia did not
produce a precipitate of iron, proving that the sulphate of
iron was decomposed, and a blue solution of copper was pro-
duced instead, the moment for adding salt was considered
to have arrived. In this way a slight volatilisation of chloride
of copper took place, but not nearly to the degree that Gurlt
suggests, and this chloride of copper could be readily trans-
formed into copper vitriol and muriatic acid by conducting
air, sulphurous acid, and steam into the condensers. The
loss of copper only amounted to from 2 to 472 per cent
of the copper contained in the ore, to which also the me-
chanical loss of the pulverised ore by the draught contributed.
If the roasted ore was removed in a dry state from the
furnace into the lixiviation vessels, a basic copper salt
appeared upon its surface as a green coating, soluble only
in a great excess of acid; but if the roasting mass was
previously moistened with hydrochloric acid the copper was
readily dissolved, and more readily the longer the ore
remained in the moistened state. The basic salt contained—
Cu.
Cl
HO.
O
58
12.49
15
11.82
corresponding to CuCl + 4CuO + 5aq, and appeared in
AGENTS USED FOR CHLORINATION.
241
small quantity when the chlorination had been judiciously
conducted; it increased at too high a roasting temperature.
On the other hand the experiments did not produce sul-
phates only without an addition of salt, even if 50 per cent
of iron pyrites was added. At best, only half the copper
contained in the ore was extracted. According to Phillips,
iron pyrites from Wicklow is roasted by itself at St. Helen's,
the chloride of copper formed by common salt is extracted
by muriatic acid, and chloride of silver is afterwards ex-
tracted by a solution of common salt. Gold is separated from
the precipitated silver.
Chlorination in the Dry and Wet Way. This mode
lately adopted by Schaffner and Meyer's suggestion with
ores of Starkenbach proves that it allows a more perfect
extraction of copper than Becchi's method.
The following agents for chlorination are used:-
a. Concentrated Muriatic Acid.-Ground oxidised ores
are moistened with concentrated muriatic acid and
exposed to the atmosphere for about a fortnight, being fre-
quently turned. Thus a great part of the copper chlorinates
and the copper glance present chiefly oxidises, rendering the
copper extractable. The decomposed mass is now heated
and continually raked, until the neutral and basic chlorides
of aluminium and iron have become nearly decomposed.
The hydrochloric acid and chlorine gases thus evolved react
whilst permeating the roasting mass, energetically chlorin-
ating the remaining oxidised and sulphuretted copper, and
the jelly-like basic iron salts, preventing the subsequent
extraction, become decomposed. If the temperature is too
high, the chloride of copper becomes transformed into
subchloride insoluble in water, and oxychloride of copper.
B. Lixiviums containing Chloride and Perchloride of
Iron. The suitably divided oxidised ore, or cupriferous resi-
due of the extraction, is moistened with a lixivium composed.
of 3 parts of chloride of iron to 1 of copper, and exposed
to the atmosphere for two or three weeks, thus causing a
partial chlorination of the copper. The greater part of the oxide
of copper is transformed into chloride, but another part which
remains insoluble is not extracted even by a considerable
VOL. II.
R
242
COPPER.
excess of chloride of iron; it may, however, be extracted
by dilute muriatic acid. But before employing muriatic
acid the ore is heated until the basic iron salts lose their
water and the greater part of their acid, and are thus
rendered insoluble in dilute muriatic acid, when the result is
a solution of copper poor in iron. If the temperature is too
high the chloride of copper is also decomposed.
y. Chloride of Magnesium, and Lixiviation.—According
to Cobley* the ore is formed into a paste with a solution
of salt, dried and roasted at a red heat, thus forming soluble
chloride of copper, from which the copper is then extracted
by magnesite, or by magnesia produced by heating chloride
of magnesium.
8. Ammonia.-It has been tried for dissolving carbonate of
copper contained in calcareous ores by Barruel,† Von Hauer,‡
at the Friedrichs-Wilhelms-Hütte|| near Commern, &c.
Attempts have been made to regain the ammonia by distil-
ling the cupriferous solution. This method has given no
favourable results, as the powdered ore absorbs much am-
monia and the distillation causes great loss.§
e. A Solution of Sulphite and Hyposulphite of Soda for
calcareous ores was proposed by Stromeyer as a substitute for
ammonia. Upon heating carbonate of copper with one of
these reagents the oxide is supposed to become transformed
into suboxide by the reaction of the sulphurous acid, thus
forming a soluble double salt of hyposulphite of copper and
soda, from which the copper may be precipitated by sulphide
of sodium.
According to Bischof,** this method is deficient in several
points; the reagent for extraction is not a cheap one and
soon loses its solvent power by decomposition; a cupriferous
precipitate, probably of subsulphite of copper, is formed,
Polyt. Centr., 1861, p. 1306.
+ B. u. h. Ztg., 1852, p. 799. Bgwkfd., xviii., 463.
+
‡ Jahrb. der. K. K. geolog. Reichsanstalt, 1852, No. 4, p. 98.
Minister. Preuss. Ztschr., 1861, Lief 2. B. u. h. Ztg., 1862, pp. 140, 230.
§ B. u. h. Ztg., 1860, pp. 111, 419.
¶ Bgwkfd., xxii., No. 30. B. u. h. Ztg., 1860, p. 111.
** Oesterr. Ztschr., 1860, No. 5. B. u. h. Ztg., 1860, p. 419.
PURIFICATION OF LIXIVIUMS.
243
and is not available at the extraction; the solution proceeds
slowly if much lime is present, the conclusion of the reaction
can only be guessed at, and the final product is only sulphide
of copper.
The Lixiviation of the Copper Salts.-The extraction
of the copper salts formed is effected by solvents in wooden
vessels, either by stirring the cupriferous substance for some
time with the reagent, and after allowing the mass to settle,
separating the liquid by drawing off with a syphon, or by filtra-
tion.* In order to obtain sufficiently saturated lixiviums a
process of continuous extraction is used, thus making the
same liquid react several times upon the substance to be
extracted. Whether the lixivium is more or less saturated
may be observed by an experienced eye by putting a drop of
solution upon a clean bright iron plate, when an unsaturated
solution will give a lighter coloured precipitate than a
saturated solution. The greener the lixiviums are the more
acid they contain, and the browner the less acid. In the
former case more iron is consumed. If the ores contain
lead it will be dissolved or enter the lixivium mechanically
as an insoluble salt, and will be obtained later as plum-
biferous copper.
The cupriferous lixiviums are heated sometimes, as at
Oker, for the production of copper-vitriol; at Freiberg, for
producing mixed vitriol; or at Mansfeld, for copper nickel
vitriol; but they are usually cemented, and occasionally
previously purified.
Purification of the Lixiviums before Precipitating from
Iron, &c.—If solutions containing iron vitriolt are kept
a longer time in contact with atmospheric air, soluble
neutral, and insoluble basic salts of hydrated peroxide of
iron are formed, of a different composition and colour according
to the degree of dilution of the solution :—
10FeO,SO3+50=3(Fe₂O3,3SO3)+2(Fe₂O3)SO3
6FeO, SO3+30=2Fe2O3,3SO3+ Fe₂03.
* Oesterr. Ztschr., 1859, p. 332; 1860, p. 342. Heeren über die Operation
des Auslaugens, in Mittheil. des Gewerbe-Ver. f. d. Königr. Hannover, 1862,
Hft. 1, u. 2.
† Oester. Ztschr., 1860, p. 292.
R 2
244
COPPER.
As the solution is more diluted, more basic salts are
separated, and at a lower temperature. Heating the solution
facilitates the oxidation of the iron vitriol.
The more neutral the per-sulphate of iron the lixivium
contains, the more iron will be consumed in the precipitating
operation. In order to lessen this consumption the following
remedies are used :—
1. An Addition of thin Milk of Lime for precipitating
iron in such quantity that the solution still remains some-
what acid, otherwise copper will also be precipitated.
The same result may be obtained by slowly conducting the
lixivium over limestone.
2. Organic Substances* (turf, saw-dust) for converting the
proto-salts into sub-salts; according to Stromeyer, the effect
is very slow, and but little iron for precipitating is saved, as
the quantity of sulphuric acid is not lessened.
When treating ores containing arseniate of copper,
arsenic acid enters the solution and is precipitated by iron
together with the copper; to avoid this, the ferruginous
copper solutions are heated with iron solutions obtained by
the precipitating operation, and thus arseniate of iron is
separated, whilst free hydrochloric acid is formed (Alderley
Edge, page 255).
According to Tripliert antimony may be separated, when
treating fallow ore with muriatic acid, by heating the solution.
until dry and again dissolving it in water; the antimony
will thus remain behind. An addition of lime also precipi-
tates iron and antimony.
Precipitation of the Copper (Cementation).—Iron, sul-
phuretted hydrogen gas, and milk of lime are chiefly used as
reagents for precipitation of the copper.
Copper may be precipitated by iron from its solutions in
muriatic acid and sulphurous acid, but not from nitric acid,
owing to its slight affinity to oxygen, and the further extrac-
tion of copper is caused by a galvanic action, to which the
contact of iron with copper gives rise. The most effective
* DINGLER'S Polyt. Journ., cvii., 446; cxi., 271.
B. u. h. Ztg., 1856, p. 210.
PRECIPITATION OF THE COPPER.
245
reagent is wrought iron, yielding chiefly a coarse granular
precipitate, but economical considerations frequently necessi-
tate the employment of pig iron, and the grey reacts more
quickly than the white pig iron. The adhesion of copper to
the separated graphite causes a greater mechanical loss at
the washing of the cement copper than when using wrought
iron. The copper precipitates upon grey pig iron more in
the form of schlich, and upon white iron as coherent plates.
The precipitation is facilitated by employing a higher tem-
perature, a large surface of the precipitating iron, by letting
the liquid fall upon the iron in thin streams, by frequently
cleaning the iron from precipitated copper, &c. Cement
water flowing quietly over large pieces of iron is only slightly
decopperised, but much more powerfully decopperised if it
falls in drops from one piece of iron to another.
Aast recommends spongy iron as a very powerful agent
for precipitating copper; it is also employed by Bronac and
Deherrypon for decomposing metallic sulphides in the dry
way.' At Moldavia|| iron deposits from the copper furnaces
are used as the precipitating agent, and at Kronstadt§ in
Siebenbürgen cupriferous pig iron produced by smelting ferru-
ginous copper slags with lime and coal powder is used.
At Alderley Edge iron chips, the residues of the tin-plate
manufacture, and old tin-plates previously freed from lead
and tin by a treatment with caustic soda, &c., are used for
precipitating the copper.
To determine whether a solution still contains copper, a
piece of clean, bright iron is dipped into it, or better, some
drops of the liquid are put upon a platinum plate, and into
them a piece of zinc, which will separate the smallest amount
of copper, appearing as a fine red coloured layer on the
platinum.
If the solution has been too concentrated (above 30° B.)
Bericht über die 2te Versammlung von
1862, p. 92.
Oesterr. Ztschr., 1860, p. 285.
Berg- und Hüttenmännern in Wien,
-+-
B. u. h. Ztg., 1862, p. 24.
Ibid., 1861, p. 292.
LAMPAD. Fortschr., 1839, p. 137.
STUNNER'S Bericht über d. Londoner Industrie-Austellung in 1862. Wien,
1863, p. 114.
246
COPPER.
it will with difficulty become clear after the precipitating
operation.
According to theory, 100 parts of copper may be precipi-
tated by 88 parts of iron, but in practice from 200 to 300
parts or more are used. This is occasioned by the free acid
present, as well as the neutral sulphate of iron, both of which
dissolve iron :—
2(Fe₂O3,3SO3)+5Fe+5HO=5FeO,SO3+2Fe₂O,,SO3+5H.
The cement copper thus becomes contaminated by the
precipitated basic iron salt. It is therefore important to de-
compose the oxide of iron by lime previous to the precipi-
tating with iron, and to counteract its formation by an oxi-
dation of the protosulphate of iron during the precipitating
operation, by excluding atmospheric air as much as possible
during the precipitating. Owing to the higher temperature,
the precipitation is performed more quickly in summer than in
winter, some iron is therefore saved in the summer, as the
action of the air then takes place in a shorter time. Dechaud,*
Sola, Becquerel,‡ Kessler, || Fitzgerald,§ and others have
suggested the employment of galvanic electricity for keeping
the copper free from iron.
Experiments at Agordo¶ prove that much of the iron is
wasted during the operation of precipitating copper, and the
basic salt when precipitated from a hot concentrated solution
is composed thus:-
(Fe₂O₂+Al₂O3+3ZnO),SO, +aq.
while that from a cold diluted solution consists of-
3(3Fe₂O3,3Al2O3),SO3+(2Fe₂O3+2Al2O3),SO3+aq.
Sulphuretted hydrogen is employed by Sinding,** at
Foldal in Norway, for the working of poor ores which could
not be profitably treated by means of iron. The pyrites is
submitted to a kernel roasting; the ore crust is extracted
+
* Bgwkfd., X., 140. B. u. h. Ztg., 1856, p. 205.
+ B. u. h. Ztg., 1853, p. 729.
‡ Ibid., 1856, p. 204. ERDMANN'S Journ. f. pr. Chemie, Bd. 62, p. 369.
|| Oesterr. Ztschr., 1858, p. 413.
§ DINGLER'S Polyt. Journ., Bd. 164, p. 185.
¶ Oesterr. Ztschr., 1862, No. 24.
** B. u. h. Ztg., 1856, pp. 210, 217; 1860, pp. 439, 487; 1862, p. 129. Oesterr.
Ztschr., 1860, p. 364.
PRECIPITATION OF THE COPPER.
247
with water, and the clear cupriferous solution obtained put
into a pierced vessel standing above the precipitating chamber.
The solution drops into the chamber like fine rain, whilst
the sulphuretted hydrogen gas is also conducted into it, thus
precipitating sulphide of copper. The sulphuretted hydrogen
gas is produced most cheaply in the following manner.
Gases produced in a gas generator from raw fuel are con-
ducted from below, through a layer of glowing fuel, across
Whilst
a fire-bridge into a vessel containing iron pyrites.
the burning carbonic oxide heats the pyrites to the re-
quired temperature, the sulphur combines partly with the
hydrogen that had been liberated from steam, and chiefly
with the hydrogen of the carburetted hydrogen, forming sul-
phuretted hydrogen, while carbon is separated.
This method not only yields a very pure copper, but it
also allows the extraction of a greater amount of copper.
One disadvantage of the method is that the sulphuretted
hydrogen gas may injure the health of the workmen.*
Lime water or lixivium of ashes (chiefly carbonate of
potash) yields ferruginous precipitates poor in copper and
containing gypsum; their smelting for the production of
copper is expensive.†
Other reagents for precipitation, such as sulphide of
sodium,‡ hydrated sulphide of iron,|| hyposulphite of iron,§
carbonic oxide,¶ sulphide of calcium,** &c., have no practical
value. Carbon also sometimes separates iron from solutions,
and a separation of copper occasionally takes place without
the employment of a reagent, probably in consequence of the
presence of suboxide of copper, which becomes transformed
into oxide and metallic copper.tt
DÖBEREINER, die Lehre von den giftigen und explosiven Stoffen, Dessau,
1858, p. 64.
B. u. h. Ztg., 1862, p. 232.
† Ibid., 1856, p. 218; 1860, pp. 111, 419.
|| Ibid., 1856, p. 217; 1860, p. 256. Oesterr. Ztschr., 1860, p. 365.
§ B. u. h. Ztg., 1856, p. 218. Oesterr. Ztschr., 1860, p. 365.
**
¶ WAGNER'S Jahrbuch, 1858, p. 73.
Polyt. Centr., 1862, p. 1026.
++ DINGLER'S Polyt. Journ., xvi., 261.
POGGENDORF's Ann., iii., 195.
SCHWEIGGER'S neues Journ., xvi., 372; xvii., 325; xxviii., 40.
248
COPPER.
Treatment of the Products of Precipitation.-These are
either cement copper or sulphide, oxychloride of copper, &c.
The cement copper is a mixture of metallic copper con-
taining basic iron salts, ore particles, particles of raw iron,
graphite, silica, sometimes antimony, arsenic, arseniate of
iron, &c., and the amount of copper is in consequence very
variable. The cement water of native solutions is usually
purer than that of artificial solutions; it deposits less basic
salts owing to the presence of free sulphuric acid, and
such deposits are carried away by the flowing of the liquid.
The cement copper is usually washed and sifted to separate
it from iron and basic salts, thus yielding chiefly three sorts.
of finely divided copper, namely, sand containing 90 per cent
or more of copper, slime containing from 50 to 70 per cent,
and slime containing 20 or 25 per cent. As a mechanical loss
is unavoidable at this dressing, it is often preferable to smelt
a more impure and somewhat poorer cement copper without
washing it, together with sulphuretted ores or matt.
The cement copper is converted into—
a. Copper Vitriol, by calcining and treating with dilute
sulphuric acid, when the greater part of the peroxide of iron
remains in the residue; sometimes no calcination is employed.
Persulphate of iron which has entered the solution does not
crystallise together with the sulphate of copper; the sulphate
of copper may be separated from iron in various ways.*
b. Metallic Copper.-A mere re-melting or a direct re-
fining of the cement copper does not yield a sufficiently pure
product, owing to the amount of basic sulphates and some-
times of antimony and arsenic present; also some matt will
always result. For this reason the purest cement copper
only is refined direct, or it is previously melted for the pro-
duction of black copper, which is then refined,t as, for in-
stance, in Schmöllnitz, in air reverberatory furnaces.
Impure cement schlich, which by itself would yield a very
impure black copper, is given in admixture to the matt
smeltings, and the more impure sorts to the ore smelting.
**
†
Oesterr., Ztschr., 1860, p. 397.
Ibid., 1860, p. 413. RIVOT, Kupferhüttenkunde, Deutch v. HARTMANN,
1860., p. 299. Oesterr. Ztg., 1860, p. 390.
PRECIPITATION OF THE COPPER.
249
In Riotinto* it has been found advantageous to roast the
cement copper after having formed it into lumps.
c. Sulphide of Copper, obtained by Sinding's method
(page 246), containing free sulphur, is filtered, pressed, and
afterwards smelted in cupola furnaces for the production of
rich copper matt. Weltzt advises that the free sulphur
should be obtained by distillation, and the residue used after
roasting, either for the production of copper vitriol, or melted
in reverberatory furnaces for the production of black copper.
d. Oxychloride of Copper and Hydrated Oxide, forming
the precipitate obtained by Becchi's process by means of
lime. According to Petitgand it is composed of-
CuO
CuCl
CaO,SO,
Fe₂O, and Al2O3
ZnO
3
Sand and silica
26.8
4'4
38.8
4.6
2.6
6'0
and is smelted in low cupola furnaces for the production of
black copper and matt; the employment of reverberatory
furnaces would be more advisable.
The remaining liquid from which the copper has been pre-
cipitated, containing muriatic acid (and, when exposed for a
longer time to the atmospheric air, much chloride of iron),
is used for the extraction of oxidised copper ores, as well as
the sulphate solution, after having extracted its iron vitriol.
The liquids rich in basic salts of sulphate of iron are fre-
quently made to circulate in sumps in order to make the
salts profitable as ochre, or when heated, as English red
(colcothar).
At Alderley Edge (page 255) a liquid containing chloride
and perchloride of iron, a small amount of chloride of cobalt,
and a large amount of chloride of manganese, is concentrated
in iron pans until it has a specific gravity of 14, and is then
poured in the form of a fine rain into a reverberatory furnace
* B. u. h. Ztg., 1861, p. 302.
† Ibid., 1862, p. 133.
+
Oesterr. Ztschr., 1860, p. 365.
250
COPPER.
having a red-hot sole covered with sand. The chloride of
iron is thus transformed into oxide of iron, steam, and hydro-
chloric acid, which is condensed in a chamber filled with
moistened coke; this acid is used for the lixiviation of ores.
EXTRACTION OF CEMENT COPPER FROM NATIVE
SOLUTIONS.
According to the quantity of solution available, the cemen-
tation is performed either at once in the mines (Rammels-
berg), or the liquid is raised to the surface (Schmöllnitz).
At Schmöllnitz* the solution is raised from the mines by
means of pumps; according to Lill von Lilienthal it con-
tains per 1 cubic foot-
Vienna lb,
0333) containing 0.264 Fe.
FeO,SO3
Fe2O3,3ŠO,
CuO,SO,
0*081}
3
다
​ZnO,SO,
0'049
Al2O3,3SO3
0'4.97
CaO,SO3
0*063
MgO,SO,
0*303
3
1.785
0°032 Cu =
0'55
per cent Cu.
The solution, containing on an average o*25 ounce of
copper per cubic foot, is conducted into precipitating vessels,
a, (Figs. 60, 61) 12 feet long, 12 inches broad, and 10 inches
FIG. 60.

I
deep, connected with each other and arranged in terraces.
The precipitating iron consists of pieces of cast iron, b, each
10 inches long, 2 inches broad, and inch thick, which are
put into the vessels and arranged lattice-like. 936 of these
horizontal vessels are employed. The solution loses most
of its copper when it has passed the first 45 or 50 vessels,
HAUCH, in Oesterr. Ztschr., 1860, pp. 276, 382.
EXTRACTION OF CEMENT COPPER.
251
and during its passage the solution also becomes muddy and
deposits basic salts. It is then conducted by a gutter into
FIG. 61.

a
a
a
ს
vertical vessels furnished with iron plates, and the precipita-
tion of the remaining copper is thus facilitated by the impact
of the solution upon the precipitating iron; 30 of these vertical
vessels are provided. The iron plates are cleaned daily, and
the cement copper is removed every fortnight from the first
vessels and every 4 weeks from the lowest. The resulting
schlich is washed on a fine sieve and the residue sorted by
hand.
The precipitating vessels contain about 20,000 cwts. of
iron, having a surface of 131,250 square feet, and about
21,444,480 cubic feet of solution pass through them annually,
yielding schlich with 57 per cent of copper on an average,
from which about 2,500 cwts. of tough copper are produced.
In 1859, 3838.69 cwts. of cement schlich, containing 215330
fine copper, consumed 5598'5 cwts. of wrought and cast iron.
The two workmen employed receive Is. 4'Sd. per cwt. of fine
copper obtained from the schlich.
The remaining liquid contains, per cubic foot,—
FeO,SO3
Fe2O3,3SO3
CuO,SO3
lbs.
I'025
Ο ΟΙΙ
0*005
5}
containing 0 382 Fe.
""
0'002 Cu
= 0*003
per cent Cu.
ZnO,SO,
A1,03,3 SO
0'045
0'448
3
CaO,SO3
ообо
MgO,SO3
0*288
1.882
252
COPPER.
Therefore the amount of iron has increased 7 per cent and
the amount of copper decreased 94 per cent, causing a loss
of 6 per cent.
The poor schlich, containing about 25 per cent of copper,
is melted together with gelf ores (page 101), and the rich
schlich, containing from 40 to 72 per cent, with quartz in a
reverberatory furnace. In 1859, 100 cwts. of schlich, con-
taining 68.8 per cent of copper, yielded about 64°2 per cent
of black copper containing 95'2 per cent of copper, and
32.8 of residues containing 17.1 per cent of copper; the loss
of copper amounted to 3'14 per cent.
At the mine of Mona, at Amlwch, in the Isle of Anglesea,
the cement solution is raised into a large basin in which the
iron ochre becomes deposited, and it is then conducted into
a system of 500 sumps filled with old iron. The ferruginous
copper slime, containing on an average 15 per cent of copper,
is added to the matt smelting. The annual production
amounts to from 12,000 to 15,000 cwts. of slime, yielding
from 1,600 to 2,000 cwts. of cement copper.
This method causes a great loss of iron, as the acid solu-
tion dissolves more iron than the precipitation of the copper
requires; the iron dissolved as protoxide in iron vitriol is
but imperfectly regained, and all the copper is not precipi-
tated.
EXTRACTION OF COPPER FROM OXIDISED ORES.
When treating oxidised ores with insoluble gangue, muri-
atic acid is usually employed, as in Linz, Stadtbergen,
Alderley Edge; sulphuric acid is seldom used, and sometimes,
as in Linz and Starkenbach, mother lye containing chloride
of iron or sulphate of iron when treating somewhat calcareous
Yet the choice between the reagents depends on the
price of materials and other local circumstances.
ores.
Illustrations of this Method of Extraction.
Sterner Smelting Works at Linz on the Rhine.-The ores of
Josephsberg, near Rheinbreitenbach, consisting of copper
glance, variegated copper ore, copper pyrites, malachite,
EXTRACTION OF COPPER FROM OXIDISED ORES.
253
azurite, and native copper in a siliceous gangue, are sub-
mitted to a sorting by hand, thus yielding ores (Scheideerze),
containing 28 or 30 per cent of copper, and poor ores (Laug-
erze), the better sorts containing from 5 to 7 per cent and
the very poorest from 1 to 3 per cent of copper, and consisting
wholly of oxidised ores, the waste ore and the poor ore being
disseminated in quartz.
The poor ores are now roasted in double reverberatory
furnaces, 100 cwts. in 24 hours, consuming 18 cubic feet of
coal; one charge remains 16 hours in the furnace. The
roasting mass is now put in quantities of 240 cwts. into
wooden cases with two bottoms, and which are surrounded
with beaten down clay, for extracting the copper by means
of dilute muriatic acid of 2° or 3° B. The acid is allowed
to remain upon it for a longer or shorter time according to
the weather and the amount of copper contained in the ore.
The time for ores containing from 5 to 7 per cent is about
8 days, and for ores containing from 1 to 3 per cent is about
4 days. The consumption of acid varies accordingly. When
the acid is saturated it is tapped off, and fresh acid is added,
which is poured upon fresh ore for the purpose of its perfect
saturation; finally, the ore is washed with water.
residues of ores containing from 1 to 3 per cent of copper,
are thrown away if they contain from o°2 to 0.3 per cent; and
the residues of ores containing from 5 to 7 per cent of copper
are thrown away if containing o'4 or o'5 per cent.
If they
contain above o˚5 per cent, owing to an imperfect roasting,
they are collected in a heap and allowed to remain for three
months, whilst mother lye containing perchloride of iron is
poured over it from time to time. Thus the perchloride is
transformed into protochloride, causing a solution of the
copper, which is then extracted in a separate vessel by means
of acidified water.
The
Experiments on performing the lixiviation whilst con-
ducting a blast of air under the double bottom of the vessel
gave a favourable result, in so far that the lixiviation became
accelerated and the residues poorer, chiefly when treating
raw ores containing carbonic acid.
The cementation of the copper is performed by means of
254
COPPER.
old iron plates, 100 cwts. of cement copper consuming 120
cwts. of iron. The resulting mother lye containing chloride
of iron is employed for extracting rich residues and for
diluting muriatic acid, but is thrown aside, if of the density
of 30° B., after having again passed over iron.
The cement copper, in a dry state, containing 92 per cent of
copper, and contaminated by iron, per- and protoxide of iron,
ore slime, &c., is put in a wet state on a small hearth for
refining. To prevent the refining taking place at the smelting
of the copper, cupriferous iron deposits are added by smelting
them first in the hearth; in this way the metal is also kept
at a higher temperature. To one charge of 400 lbs. of
cement copper 20 lbs of iron are added, and the charge is
refined in about three hours, consuming 200 lbs. of coke.
The monthly production of the Sterner Smelting Works
amounts to 150 cwts. of rosette copper of very good quality,
three-quarters of which is produced from cement copper.
The production of 100 lbs. of rosette copper requires about
11,500 lbs. of ore, 837 lbs. of muriatic acid, 3s. per cwt.,
and 64 lbs. of iron, causing an expense of £3 12s.; when
treating oxidised ores this expense amounts to £2 148.
At Stadtbergen in Westphalia, and at Linz, similar processes
are performed. Rich ores containing 40 per cent of copper
are melted direct; acidified ores (5,000 tons), with 1 or
2 per cent, are treated by the sulphurous acid gas resulting
from the roasting of zinc blende and iron pyrites; and about
60,000 tons of acidified ores containing per cent of copper
are treated with muriatic acid and mother lye containing
sulphate of iron.
2
The vessels for cementation are divided by a lattice into
two vertical compartments, the interiors of which are pro-
vided with stirring machinery for keeping the solution in
continual motion. In the outside partition, between the
lattice and sides of the vessel, iron plates for precipitating
the copper are placed upon a grate. The cement copper is
washed, yielding purified copper and slime; the copper is
refined in reverberatory furnaces. The refinery slags are
smelted together with the slime for the production of black
copper, which is then refined. The copper resulting from
EXTRACTION OF COPPER FROM OXIDISED ORES.
255
the lixiviation with muriatic acid is refined apart from that
treated with sulphuric acid; the refined copper of the former
operation being somewhat finer than the latter. Ores con-
taining o'75 per cent of copper yield 0.63 per cent, and con-
sume from 420 to 500 lbs. of muriatic acid of 18°B, and 120 lbs.
of iron plate waste per 5 tons of fine copper produced.
The annual production amounts to 500 tons of copper,
450 tons of vitriol, 1000 tons of roasted blende, 150 tons of
litharge, and 300 lbs. of silver.
At Alderley Edge.-The ores consist of mixtures of ar-
seniate, carbonate, phosphate, and pure oxide of copper, with
baryta, disseminated in sandstone. The ore is pounded to
the size of half a cubic inch and put into vessels of wood,
sandstone, or slate, 11 feet long, 8 feet wide, and 4 feet deep,
with a loose bottom upon which brush-wood and straw are
spread out. A vessel contains about 9 tons of ore, and crude
muriatic acid is poured over it until at first it is completely
covered by it. As soon as the acid has disappeared the
surface below the vessel is again filled with the washing water
of other vessels till within two inches from the top, and the
lixivium between the loose and chief bottom is pumped upon
the ore until its specific gravity no longer increases. The
ore from which the copper is not yet completely extracted is
again treated with fresh muriatic acid and washing water;
the resulting lixivium which is not saturated is poured into
fresh ore. The extracted ore is now washed several times
with water, removing each time about six inches of the
ore until all is removed. This extraction takes about
three days, and about 1000 tons of ores may be treated in
16 vessels in one month. The solution of copper containing
arsenic is then treated with a solution of iron in the propor-
tion of 2 parts of the solution of copper to I part of the solu-
tion of iron, in order to purify it from arsenic. The solution
thus rendered acid is again poured upon fresh ore together
with the precipitate of arseniate of iron, and its copper is
afterwards precipitated by old iron plates. The ore is com-
pletely extracted by fresh acid, and the resulting solution
freed from arsenic and put upon fresh ore.
The mother lyes resulting from precipitating the copper
256
COPPER.
are employed for the production of muriatic acid (page 249),
and the cement copper, containing 75 per cent of copper, is
washed and dried, and added to the smelting.
EXTRACTION OF COPPER FROM SULPHURETTED ORES
AND PRODUCTS.
The most simple of these processes consist in exposing
ores to the atmosphere and extracting the sulphates formed.
Where economy in fuel is not so much an object as a quick
and more perfect extraction of copper, the ores are roasted
in heaps, and occasionally in cupola and reverberatory
furnaces, and the copper afterwards extracted by means of
water, liquid, or gaseous acids; sometimes the ores are
chlorinated previous to their extraction. Iron, sulphuretted
hydrogen gas, or lime is used for precipitating the copper.
Illustrations of this Process of Extraction.
At Riotinto* in Spain both native and artificial solutions
are treated.
Iron pyrites, associated with copper pyrites, variegated
copper ore, and chiefly Cu₂S, and containing on an average
3 per cent of copper, is roasted in open heaps for six or
eight months on a foundation of brush-wood 1 metre high.
The roasted ore is then immersed in water in sumps or
vessels; after 24 hours the water is tapped off, and the ore
is treated with fresh water eight or ten times until the
copper is completely extracted. The residues are exposed
to the air to decay and again treated. After being allowed
two or three hours for settling, the solution is put into
sumps for cementation, in which the copper is precipitated
by means of cast-iron, 100 parts of copper consuming
300 parts of iron.
After the solution is freed from its copper fresh solu-
tions are added from three to five times without removing
the cement copper. The mother lye afterwards passes.
several sumps for precipitating copper, thus yielding preci-
pitates containing from 8 to 10 per cent of copper, while the
* B. u. h. Ztg., 1861, p. 289; 1862, p. 301.
EXTRACTION OF COPPER FROM SULPHURETTED ORES. 257
cement copper containing 50 per cent is formed into lumps,
dried in the open air, and heated on the grate of a small
cupola furnace (page 249). The roasted product is sub-
mitted to an oxidising smelting in a small hearth and
refined in a blast reverberatory furnace.
The yield of copper amounts to from 15 to 17 per cent,
and therefore the loss of copper is 50 per cent or more; at
Agordo only 10 per cent is lost.
At Agordo the crusts of roasted ore waste and schlich
(page 103) are lixiviated three times, and afterwards used as
bottoms and covers in the kernel roasting; they are then
lixiviated three times more, and afterwards given back to the
roasting, and the copper again extracted. The residue is
sifted and the coarse powder roasted anew and washed. The
first lixiviation yields a rich solution, but the solutions of the
second and third lixiviations are poor,
To obtain rich solutions wooden vessels with a capacity of
about 400 cubic feet are charged with 285 cwts. of fresh ore.
crusts, and water is added until it covers the ore by several
inches. After 24 hours the lixivium is tapped off into the puri-
fying basins; this contact of the ore with water is twice
repeated. The residues are given to the roasting, and the
second and third lixiviums of 14° or 15° B. are employed
instead of water for lixiviating fresh ore.
Previous to the precipitation the rich solution of from
31 to 40° B. is diluted with the solution of the third lixiviation
in order to facilitate its purification after precipitating.
The roasted residues are treated as the original ore crusts;
after sifting, the resulting residue is again roasted, and the
roasting mass submitted to a continued washing; for this
purpose four vessels stand close to each other. The roasted
mass is put for 24 hours together with water into vessel No. 1,
the solution is tapped off into the purifying basin, and the
roasted mass is removed from No. 1 to No. 2, fresh water being
added. After 24 hours, the solution of No. 2 is pumped
upon
fresh ore in No. 1, while the ore of No. 2 is removed
into No. 3, with fresh water. After 24 hours the solution of
No. 3 is added to No. 2, that in No. I is tapped off, and
the ore of No. 1 is removed into No. 2, and that of No. 3
VOL. II.
S
Ι
258
COPPER.
into No. 4, fresh water being added. The ore of No. 4 is
then put upon an inclined sieve to separate the coarser parts,
while the solution of No. 1 is always tapped off into the
purifying basins. The solutions of the different vessels
mark respectively 16°, 12°, 8°, and 4° B.
According to Rivot, the annual production from 11,400 tons
of ore crusts is 7,000 cubic metres of rich solution of 29° B., con-
taining 98 tons of copper; 4,000 cubic metres of solution of the
second lixiviation of from 23° to 24° B., containing 30 tons of
copper; and finally, 1,000 cubic metres of the last lixiviation
of 18° or 19° B., containing 8 tons of copper; altogether
12,000 cubic metres, containing 136 tons of copper, whilst
the assay determined an amount of 151 tons; the loss amounts
therefore to 10 per cent. The slime, which is thrown away,
does not contain more than 0.0017 or o'о018 of copper.
The precipitation of the copper is effected chiefly in wooden
vessels 13 feet long, 11 feet broad, and 5 feet deep, with a
capacity of 650 cubic feet, and containing about 160 cwts. of
cast-iron which are distributed on benches standing on the
inside of the vessel. In the middle a leaden bell is suspended.
above a grate by means of an iron frame, the upper part of
which is connected with a tube for charging fuel (turf char-
coal), and it has a tube on one of its sides which enters a
leaden vessel provided with three horizontal partition walls.
for equalising the temperature. Out of this leaden vessel
the cooled products of combustion enter a wooden chimney
by means of a leaden tube. That chimney also received
the vapours of the water, as the vessels are kept covered for
excluding air. The decomposition of 500 cubic feet of warm
solution of 60° by means of iron is finished in 20 hours; 48 hours
are allowed for cooling, and then the clear solution, which
has a temperature of 35° C., is tapped off into crystallisation
vessels, in which the iron vitriol crystallises; 2°5 parts of
iron are consumed to I part of fine copper.
In rare cases only are reverberatory furnaces employed for
precipitating the copper.
According to Rivot, for each ton of ore o*027 tons of cast-
iron, 0*014 tons of wood, o'030 tons of charcoal, o'027 tons
of turf are consumed, at a cost of 8.674 francs; in the
EXTRACTION OF COPPER FROM SULPHURETTED ORES. 259
production of 0.1336 ton of kernels, containing o'006 ton of
copper, and o‘0225 ton of cement copper, containing o'009 ton
of copper. The poor and rich cement schlichs are smelted
together with the sulphuretted ores.*
Of late the pyrites schlich is treated in the following
manner :- It is mixed with mother lye of 25° B., containing
sulphate of iron, and formed into a pasty mass, which is then
beaten down in semi-circular brass moulds. By turning the
moulds the pieces of schlich are made to fall upon a wooden
board, and are dried for 3 or 4 weeks on roasting heaps, and
afterwards roasted in heaps without forming kernels. They
are then lixiviated with cold water in vessels provided with
three pierced intermediate bottoms. The first strong solution
of 30° B. is cemented; the succeeding weaker solutions are
employed for the subsequent lixiviations; pyrites containing
135 per cent of copper yields residues containing o'2 per
cent, which are thrown away; the loss of copper amounts
to 14 per cent.†
Mrazek and Eschka have further investigated this process,
which may be of importance to establishments using expen-
sive fuel, particularly if it should succeed in transforming
the sulphide of copper contained in richer pyrites into sul-
phate. In Freiberg the ore pieces formed showed no con-
sistance, probably owing to deficiency of lime.
At Foldal in Norway, cupriferous solutions, containing
2 or 3 per cent of copper, are obtained by allowing the crusts
obtained from the kernel roasting of rich pyrites to decay in
the open air; or solutions containing o'5 to 15 per cent of
copper after roasting raw pyrites for 5 or 6 months. For the
purpose of roasting the pyrites is divided into pieces of the
size of the fist, which are then roasted in open heaps of
from 3,000 to 4,000 cubic feet, and afterwards exposed to the
air to form sulphates. The solutions obtained by watering
the heaps are allowed to settle in large sumps and then con-
ducted into reservoirs standing above chambers, which are
filled with wooden sticks. The solution drops down between
Oesterr. Ztschr., 1860, p. 389.
† Ibid., 1861, Nos. 41, 50.
S 2
260
COPPER.
these sticks, whilst sulphuretted hydrogen gas enters the
lower part of the chambers and circulates through them
several times, thus precipitating the copper as sulphide.
Sinding's method of producing sulphuretted hydrogen gas
is as follows:-The gas generator, A (vide Figs. 62, 63, 64), is
provided with a grate, a, a lid, b, and two tuyeres, c; burning
coal is put upon the grate, which is then filled with raw fuel
such as wood or turf, and sometimes coal; the blast is put
in motion as soon as the fire comes up to the tuyeres on
FIG. 62.
FIG. 63.


N
B
h
h
A
C
α
A
d
n
B
g
2
account of the natural draught caused by the ash pit and
the channel, d, with which the furnace is provided for purifi-
cation. The lid is now hermetically closed, and the produced
g
FIG. 64.

g
B
7
A
B
g
9
gases, together with steam which is then decomposed by the
burning coal, pass through the grate, a, into the gas channel,
h, on to the fire-bridge, f; here air for combustion is supplied
PRODUCTION OF SULPHURETTED HYDROGEN GAS. 261
by means of the channel, i, from which it passes through
various openings, k. On passing a red-hot iron into the
furnace, B, by the opening, e, the carbonic oxide will chiefly
burn, the iron pyrites in в becoming heated will volatilise
sulphur, which then combines with the free hydrogen and
that contained in the carburetted hydrogen, forming sulphur-
etted hydrogen; this, as well as the products of combustion,
passes beneath 1, through the channels m and n, into the
precipitating chambers. From time to time the roasted
pyrites is removed by g, and fresh pyrites is charged by e.
This roasted pyrites decays easily, forming a coppery solution.
The precipitated sulphide of copper is conducted, together
with the decuprified solution, into a zigzag channel in which
the sulphide of copper deposits, while the decuprified solu-
tion is made to run off or once more added to the precipita-
tion apparatus. The residue from the channel is put into a
vessel for settling, the mother lye is decanted, and the
sulphide of copper, as a thick paste, put upon large wooden
sieves which are covered with cloth; it is there somewhat
washed, and the liquid passing through the sieve is made to
pass over pieces of iron. The residue is then dried; it contains
from 15 to 20 per cent of copper, besides free sulphur, water,
powdered pyrites and soot, and is liable to spontaneous
combustion if kept in large quantities, but this may be avoided
by firmly pressing it.
The product has been treated experimentally in different
ways. It has been tried to lessen its volume by pressing,
but without good result; roasting proved difficult and en-
tailed loss. Smelting in cupola and reverberatory furnaces
also gave unfavourable results. At present it is smelted in
cupola furnaces for the production of matt, after being first
distilled to extract the free sulphur. As this product is diffi-
cult to roast, dead roasted raw matt of the ore smelting is
added when smelting it, thus facilitating the process and pre-
venting great loss of copper. A partial smelting in rever-
beratory furnaces would certainly be of some advantage.
According to Rivot, the cost of producing 1 ton (1000 kilos.)
of copper at Agordo is £37 18s.; at Foldal the cost is only
£20 15s. 8d.
262
COPPER.
Becchi and Haupt's Process at Capannevecchie, near Massa
Maritima.-Copper pyrites disseminated in a quartz vein in
Purassic limestone, associated with a little blende and iron
pyrites, and containing about 2 per cent of copper, is roasted
for about a fortnight in open heaps containing 250 tons. The
roasting mass is then divided by pounding and grinding,
sifted, and re-roasted in charges of about 2 tons in double
reverberatory furnaces at a strong fire for about three hours,
whilst it is frequently raked. When no more sulphurous
acid is evolved the temperature is lowered to a dark red
heat, and from 2 to 8 per cent of common salt is added,
according to the amount of copper in the ore. It is raked
with this addition for about ten minutes, and afterwards.
lixiviated.
The roasting mass is moistened in wooden lixiviation
vessels provided with a filtering apparatus, then lixiviated
with acidified water, and the copper precipitated with lime.
The precipitate is smelted in a low cupola furnace together
with poor copper slags for the production of matt; the matt
is roasted and treated for black copper, which is then refined.
One cwt. of fine copper is produced from about 27 cwts.
of ore, at a cost of £4 16s.
Bischof's Method.-Calcareous ores are burned in a lime
furnace, and washed to extract the caustic lime formed. They
are afterwards smelted with iron pyrites for the production
of raw matt. The raw matt is pulverised and roasted in a
muffle furnace at a dull red heat; the sulphates formed are
extracted by means of water, and the residue is treated with
dilute sulphuric acid in order to extract all the copper and
silver, provided too much antimony and arsenic are not pre-
sent. If antimony and arsenic are present in larger quantity
the residue must be re-roasted with coal and iron pyrites, or
with raw matt, and the copper extracted by means of the
solution obtained before. A second roasting is not required
if silver is present.
The copper is precipitated by means of iron, the cement
copper then washed with a solution of copper vitriol, in order
to remove metallic iron; it is next treated with water, and
finally with an alkaline solution in order to extract basic
ANNUAL PRODUCTION OF COPPER.
263
sulphates of iron. The product so purified is roasted again
to volatilise antimony and arsenic, and is afterwards smelted.
The annual production of copper is about as follows:-
Great Britain
Russia.
Austria.
Prussia.
Sweden.
Spain
Belgium
Saxony.
•
Tons.
15,000
5,500
3,000
2,000
2,000
2,000
1,000
$35
CHAPTER II.
IRON.
WHILST the commercial value of metals generally increases,
pari passu, with their purity, pure iron may be said to have
no commercial existence. On the other hand, this metal
forms several combinations with carbon which are most ex-
tensively applied for technical purposes. Regarding their
chemical composition, there is no exact line of demarcation
between those combinations, and their diversity is merely to
be found in the peculiar nature of each of the combinations,
as well as in the different ways of producing them.
According to these variable properties iron is classified
into :-
I. Cast-iron or Pig-iron, the combination richest in
carbon. It is not ductile and cannot be welded; it is more
or less hard, and melts at a high temperature (1400° to
1600° C.) According as the metal is most adapted for the
use of founders or forge-masters, it is called forge or foundry
pig.
2. Wrought or Malleable Iron.-This is soft, malleable,
and capable of being welded when at a white heat. It melts
only at the very highest temperatures (above 2000° C.), and
when heated and suddenly cooled it retains its softness. Its
carbon never exceeds o°25 per cent. It is usually produced by
the conversion of pig-iron, and in rare cases is obtained direct
ANNUAL PRODUCTION OF PIG-IRON.
265
from the ores. The varieties of malleable iron are distin-
guished by different names, having reference rather to form
and destination than to differences in composition.
3. Steel. This variety of iron contains more carbon than
malleable iron, and less than pig-iron. It is both fusible
(1850° C.) and malleable, and may be welded at a lower tem-
perature than wrought iron, and therefore requires greater
compressing power. Its distinguishing property is its power
of being hardened or softened at pleasure by the process
termed tempering. Steel is produced from either cast or
wrought-iron, and seldom direct from ores.
According to the varieties of iron its metallurgy may be
classified as follows:-The production of cast-iron; the pro-
duction of wrought-iron; and the production of steel.
Its great importance will be obvious when we review the
annual production of iron, the value of which exceeds that
of all other metals together.*
The manifold uses of iron prove its necessity, and the in-
dustrial development of a nation seems to stand in direct
proportion to its consumption of iron; and the prosperity of
a nation to a certain degree depends on the sufficiency,
cheapness, and quality of the iron it produces.
In December, 1855, Mr. Blackwell estimated the annual
production of pig-iron in the whole world to be as follows:-
Great Britain
France.
United States in America
Prussia.
Austria.
Belgium
Russia.
Sweden
Various German States
Other countries .
Tons.
3,000,000
750,000
750,000
300,000
250,000
200,000
200,000
150,000
100,000
300,000
6,000,000
A cube formed of this quantity of iron would measure
393 feet in its sides, and a cylinder with a diameter of
B. u. h. Ztg., 1857, p. 181.
266
IRON.
600 feet would be 100 feet high, and transformed into rails,
there would be sufficient to pass twice round the earth.
Hewitt calculated the consumption of iron in 1855 for
each inhabitant of the following countries to be-
Consumption.
England
United States in America
Belgium
France.
Germany
Sweden and Norway
Switzerland
Austria.
Russia .
:
Spain
lbs.
144
117
70
60
50
30
22
15
ΙΟ
5
Mr. Hunt,† in his very valuable "Mineral Statistics," gives
the total quantity of pig-iron produced in Great Britain in
the year 1864,—4,767,951 tons, and in 1865,—4,819,254 tons
-to be as follows:
1864.
1865.
Tons.
Tons.
Northumberland
55,467
49,290
Durham.
466,980
476,767
Yorkshire, North Riding
West Riding
409,106
486,421
II2,093 123,233
Derbyshire
174,743
189,364
Lancashire.
Cumberland
•
195,460
204,925
141,033
107,430
Shropshire.
130,666
117,343
North Staffordshire
•
217,996
206,268
South Staffordshire and Worcestershire
628,793
692,627
Northamptonshire and Lincolnshire
22,823
25,728
Gloucestershire, Wilts, and Somerset
65,312
65,471
North Wales
51,108
51,874
South Wales
988,729
916,909
Scotland.
1,158,750 1,163,478
4,767,951 4,819,254
* Preuss. Handelsarchiv., 1857, Bd. i., 433.
1861, p. 39.
† Dr. URE's Dictionary of Arts, &c., ii., p. 695.
B. u. h. Ztg., 1859, p. 260;
DIFFERENT KINDS OF PIG IRON.
267
In 1865, the number of furnaces in blast used to furnish this
astonishing quantity were, in England, 376, distributed over
176 iron works; in Wales, 135, distributed over 49 works; and
in Scotland, 141, over 32. To supply these furnaces there
were raised 9,910,045 tons of ore, the estimated value of
which, at the place of production, was £3,324,804; that of
the pig-iron, at the mean average cost at the place of pro-
duction being £12,048,133. Of the iron stone, 1,384,500 tons
were argillaceous carbonate from the coal measures of
Staffordshire and Worcestershire; nearly 1,500,000 tons from
the coal measures of North and South Wales, and 1,500,000
tons of argillaceous carbonate from Scotland.
PRODUCTION OF PIG-IRON.
Different Kinds of Pig-Iron, its Properties and
Formation.
Iron ores consist chiefly of oxidised iron and slag-forming
components. Upon exposing them in a blast furnace at a
gradually rising temperature to reducing and carbonising
agents, the oxidised iron will be reduced to the metallic
state. At a certain temperature this iron becomes saturated
with carbon, thus forming different sorts of pig-iron of the
most varying physical and chemical behaviour; this partly
depends on the more or less high temperature of the furnace
hearth where, on passing before the tuyere, the carbonised
iron melts together with the slags that have been formed.
Certain substances, such as sulphur, phosphorus, silica,
arsenic, manganese, copper, &c., contained either in the ore.
flux, or fuel, will likewise influence the nature of the pig-iron,
as they become more or less reduced according to the
temperature of the hearth, and the state of their original
combination; these substances then combine with the pig-
iron.
According to external appearances, the pig-iron may be
divided into the following kinds :-
1. White Pig-Iron.-It is usually obtained by smelting
ores rich in manganese which are easy both to reduce and to
fuse.
Upon smelting very pure ores almost free from
268
IRON.
sulphur and phosphorus and rich in manganese, and employing
a temperature just high enough to completely saturate the
reduced iron with carbon (the temperature for smelting and
for reducing must be nearly the same), a distinct variety of pig-
iron termed spiegeleisen is formed. Upon keeping the tempera-
ture of the furnace so low as to prevent the iron from becoming
saturated with carbon, different kinds of white iron are
formed, containing less carbon than spiegeleisen. Their
structure is radiated crystalline, granular crystalline, &c.,
according to their degree of carbonisation, which depends on
the temperature employed.
The common white iron from the ordinary process. is pro-
duced from less pure ores than those used for the production
of spiegeleisen, and at a somewhat higher temperature.
Other kinds of white iron are intentionally produced in
some smelting works by either suddenly chilling grey iron,
or by partially decarbonising it in the hearth by means of
oxidising agents (protoxide of iron, carbonic acid); in some
cases these kinds are produced from ores rich in sulphur and
phosphorus.
The varieties of white iron are more easily fused (1400° or
1500° C.) than grey iron, but they chill sooner; some kinds.
when melted are thinly liquid, others thickly liquid; they
dissolve in concentrated muriatic acid, emitting an odour of
carburetted hydrogen, usually leaving a brown residue, and
sometimes even some graphitic carbon. When suddenly
cooled, they have a clear sound, and though very hard may
be easily broken; when suddenly chilled they cannot be
attacked by the hardest file, and the lighter their colour the
harder they are. These qualities generally render white
iron unfit for casting purposes, while some of its varieties
are in great request for conversion into malleable iron.
White iron produced from pure ores at a comparatively
low temperature is only slightly contaminated by foreign
substances, and contains carbon in chemical combination,
which easily oxidises. Some kinds of white iron pass before
melting through various statés of a pasty or semi-liquid
condition, in which they remain for a longer or shorter time,
thus facilitating the decarbonisation, whilst other sorts,
SPIEGELEISEN.
269
spiegeleisen, &c., become first perfectly liquid, and then
quickly assume a pasty condition. The former sorts are
more easily converted into malleable iron, but at a greater
loss of iron by oxidation; they are therefore usually treated
in hearths, whilst the latter sorts are treated in hearths and
also in puddling furnaces for the production of wrought-
iron and steel. The decarbonisation takes place quicker in
puddling furnaces than in hearths, owing to the greater con-
tact with oxidising agents. Those sorts attaining different
states of semi-fluid condition immediately upon melting do
not pass through a state which would allow of welding.
Light grey iron containing phosphorus is usually unfitted
for the production of malleable iron, but it is suitable to
certain castings, as it becomes very thinly liquid when
melted. The white iron obtained by chilling grey iron con-
tains all the impurities of the grey iron, and in most cases is
unfit both for castings and for conversion into wrought-iron.
Spiegeleisen is probably the only iron forming with carbon
a chemical combination corresponding to the formula Fe̟C,
and it is undecided whether another combination of iron
with carbon in the proportion of equivalents exists, but its
existence is assumed by V. Mayerhofer* and Gurlt;† and
though their suppositions are of no practical value, they may
assist in explaining the theory of blast furnaces.
The kinds of white iron of which we have spoken possess
the following properties:
a. Spiegeleisen, the so-called specular pig, contains from
45 to 55 per cent of carbon in chemical combination, and
up to per cent of silicon. It is silver white in colour, of a
bright metallic lustre, and very hard and brittle, so that it
may be pulverised. It scratches glass, and cannot be attacked
by a file, and forms lamellar crystals frequently several inches
in size, but the form of these crystals is not yet distinctly
defined.
Whilst Rammelsberg|| states the plates to be mono-klino-
* LEOBENER and PRZIBRAMER, Jahrb., 1861, Bd. 10, p. 432.
↑ B. u. h. Ztg., 1855, p. 400.
+
Ibid., 1852, p. 274. LEONHARD, Hüttenerzeugnisse, 1858, p. 247.
|| LEONHARD, Chem. Metallurgie, 1850 p. 63.
270
IRON.
metrical prisms with angles of 112°, 116, and 130-131°; these
angles measure 120°, according to Mitcherlich, and from
128° to 129°, according to Gurlt. Karsten,* Fuchs,† and
Hausmann‡ have suggested that the plates do not belong
to the isometric system, but are merely rudiments of crystals
which intersect each other at the most various and perfectly
irregular angles.
The composition of spiegeleisen is shown by the following
analyses :-
Fe.
I.
82.860
II.
III.
IV.
V.
93'364 88.30
89.72
89.80
Mn
10*707 3°204 4'50
4'50
4°24
Ni
0.016
Co
trace
Zn 0*30
Cu
0'066
0.18
Al
0'077
Ti
0*006
Mg
0'045
Ca
0'091
K
0.063
Na
trace
Li
trace
As
0'007
Sb
0'004
Р
0*059
S
0'014
N
0*014
Si
0'15
0*08
0*08 trace
0'002
trace
0'20
0*56
0'37
5°14 5'41
C
O
o'997 0*640
4*323 2*250 5'48
0*665
No. I is spiegeleisen produced from the spathic ore of
Stahlberg, near Müsen, worked by the Müsen and Cologne
Mining Company, analysed by Fresenius in 1862. 0'475 O is
combined in SiO2 of the intermingled slag, and o'190 O is com-
bined in the bases of that slag. The spathic ore contained—
FeO,CO₂
MnO, CO₂
CaO,CO,
2
MgO,CO₂
Sandy residue
Moisture
74 47
17 08
I'34
5'75
I'08
0°09
*
LEONHARD, Eisenhüttenkunde, 3, Aufl. i., 181.
†DINGL., Bd. 164, p. 348.
+
+
Beiträge zur met. Krystallkunde, 1850, p. 7.
SPIEGELEISEN.
271
No. 2 is spiegeleisen from New Jersey* produced with
charcoal from Franklinite, containing 0'240 per cent of
Al₂O, and 0*170 per cent of CaO.
3
No. 3 the same smelted with anthracite.
No. 4 spiegeleisen from Hammhütte (Sayn-Altenkirchen),
analysed by Karsten, and, according to Gurlt, consisting of—
(Fe+Mn),C+Mn,Si+ Mn¸P+MngS.
No. 5 spiegeleisen from Lohhütte, analysed by Karsten;
according to Gurlt, it consists of-
Fe₁C+ FeSi.
Karsten,† Gurlt,‡ Von Mayrhofer, and others have sug-
gested that the composition of spiegeleisen is Fe,C, containing
94916 Fe and 5'084 C; their suggestion is probably correct,
as it is indicated by many analyses as well as other known
combinations of iron with electro-negative substances in the
proportion of 4 equivalents to 1. A higher amount of carbon
sometimes contained in spiegeleisen may be derived from
manganese, which is mostly found in it and replaces the iron,
but combines with a larger quantity of carbon than an equal
weight of iron, owing to its lower equivalent. Spiegeleisen
may therefore be the richer in carbon the more manganese it
contains, but it never exceeds the limit of 5'93 per cent.
Spiegeleisen with the largest amount of carbon represents,
according to Rammelsberg, the formula-
Fe¸C + FeC,
which would be equal to 5'77 per cent of carbon.
Buchner considers spiegeleisen to be a combination
of pure iron with some carburetted iron of unknown compo-
sition, and thinks the suggestion of a chemical formula
unjustified by the facts.
Spiegeleisen has a specific gravity of from 7·6 to 7·66, and
melts at about 1,500 or 1,600° C., becoming perfectly fluid;
it solidifies without previously attaining a pasty or semi-fluid
* B. u. h. Ztg., 1860, p. 465.
† KARSTEN, Eisenhüttenkunde, 3. Aufl., i., 181.
B. u. h. Ztg., 1855, p. 400. Bgwkfd., xviii., 325. Berggeist, 1860, No. 19.
GURLT, pyrogenet. Mineral., p. 38.
LEOB. Jahrb., 1861, Bd. x., pp. 350, 432.
§ Bgwkfd., Bd. 21, No. 5. B. u. h. Ztg., 1858, No. S.
272
IRON.
condition and without graphite separating from it. Whether
chilled slowly or suddenly it still retains its texture, colour,
hardness, and brittleness.
If, on the other hand, spiegeleisen is heated above its
melting point (1600° C.), and made to cool slowly, graphite
will separate, converting the speigeleisen into grey iron,
and lessening its specific gravity. When suddenly cooled
this over-heated spiegeleisen retains its white colour, as the
graphite is most finely distributed in the whole mass of iron,
so that it cannot be distinguished by the naked eye, but its
presence must be ascertained by a chemical analysis. Grey
iron slowly cooled after melting remains grey; it becomes
white and decreases in density when cooled suddenly.
The conversion of spiegeleisen into grey iron by means of
a high temperature and slow cooling is explained by Gurlt.
He supposes that the tetracarbide was transformed by the
influence of the high temperature into a lower carbide and
free carbon--
as sulphide of iron
and free sulphur-
2 Fe C
may be
FeS₂
=
FesC + C,
converted into a lower sulphide
FeS + S.
The condition of the separated carbon in liquid grey iron
is not exactly known. Karsten, Mayrhofer, Schafhäutl and
others suggest that carbon was chemically combined with
the liquid iron and separated from it upon chilling, as solidi-
fied iron cannot dissolve so much carbon as liquid iron; in
explanation of this it is stated that grey cast-iron rendered
white by suddenly chilling contains all its carbon in chemical
combination. Caron* has proved by analyses that hardened
steel upon dissolving in acids yields less graphitic residue
than steel of the same kind before it is hardened, from which
we may conclude that, upon hardening, steel combines
chemically with some carbon.
Grey iron intended for conversion into malleable iron is
sometimes intentionally converted into white iron by a sudden
chilling, in order to convert the carbon into a combined
*DINGL., Bd. 168, p. 36.
SPIEGELEISEN.
273
state, thus avoiding the presence of graphitic carbon, which
is difficult to burn.
Kuhlmann has analysed samples of grey iron taken from
the same ladle; the first sample was taken whilst the iron
was liquid, and when dissolved in acid it yielded a consider-
able residue of graphite; the second sample was taken after
having suddenly chilled the iron, and yielded only a small
quantity of graphite.
Gurlt* and others do not consider graphite to be in che-
mical combination in the liquid iron, but contained in chilled
white iron in a very fine state of distribution, which may be
proved by analysis. Eyfertht concludes from experiments and
from various phenomena of the process in iron blast furnaces
that graphite is already contained in the liquid grey iron.
More exact chemical analyses of chilled white iron will be
required to decide this point, as some of the analyses at
present known show a small amount of graphite, whilst
others do not show any. It is probable that the conversion
of white iron into grey iron is analogous to phenomena
observed in other cases, in which a chemical combination is
partially decomposed and one of its components is separated
from the liquid combination owing to a tendency to crystal-
lisation. This separation of graphite may be facilitated or
prevented by the presence of other foreign substances, such
as sulphur, phosphorus, silicon, manganese, &c.
Karsten states that the combined carbon in liquid
spiegeleisen may be converted into graphite not only by
variation of temperature but also by adding to spiegeleisen
sulphur, phosphorus, or silicon; these substances then form
chemical combinations with the iron, thus separating part of
the combined carbon.
Spiegeleisen is produced when smelting porous iron ores,
chiefly spathic ores free from sulphur and phosphorus, and
containing a sufficient amount of manganese to render the
ores easy to fuse, and when employing a temperature just
VOL. II.
* B. u. h. Ztg., 1855, p. 403.
† Ibid., 1861, p. 142.
KARSTEN, Eishenhüttenkunde, 3 Aufl., i., 427.
T
274
IRON.
high enough completely to saturate the iron with carbon and
prevent the formation of grey iron; grey iron will be formed
if the carbonised iron passes into the hearth at too high a
temperature. Fuel free from sulphur, usually charcoal, is
employed. The charges of ore must be large enough to
cause a smelting of the ore a little above the tuyere, otherwise
the spiegeleisen would be exposed for a longer time to the
decarbonising action of carbonic acid. Ores containing
sulphur and phosphorus, as we have before mentioned,
prevent the perfect carbonisation of the iron and give rise to
the formation of white iron, poorer in carbon, not usually
having the texture of spiegeleisen, but radiated or even
granular in texture and bluish white in colour. These
modifications of white iron may also be formed by dissolving
iron free from carbon in liquid spiegeleisen.
The temperature required for the formation of spiegeleisen
is sufficiently high to reduce some of the silica contained in
the ore mixture, causing the iron to become somewhat
siliceous. Though the mixture frequently contains a large
amount of oxidised manganese, a proportionally small quan-
tity only becomes reduced by the fuel or the carbon of the
pig-iron which has already been formed, as the oxide of
manganese, being more difficult to reduce than oxide of iron,
chiefly enters the slag, rendering it easy to fuse. If the
mixture contains a considerable excess of oxide of manganese,
white iron poorer in carbon, less hard, and more difficult to
fuse, but without an increase of manganese, is formed, owing
probably to a readier formation of slag and a more energetic
reaction of the oxide of manganese upon the carbonised iron
in the hearth of the furnace. The amount of manganese in
spiegeleisen does not usually exceed 7 per cent, and only
increases up to 22 per cent under peculiar circumstances,
such as a higher temperature in the hearth, stronger pressure
of carbonising gases and vapours, basic slags from which
protoxide of manganese is more easily reduced, &c. The
value of spiegeleisen for certain applications is fixed by its
per centage of manganese.
Spiegeleisen is fit for conversion into malleable iron and
steel, in hearth fineries as well as in puddling furnaces; but
SPIEGELEISEN.
275
as it becomes very liquid upon melting, it takes more time to
fine than the porous variety of white cast-iron, termed
"luckige floss," a circumstance which is favourable when
working for the production of steel, as it facilitates the inter-
ruption of the oxidation at the proper time.
As spiegeleisen is expensive and difficult to treat it is not
usually worked by itself, but is used as an addition to
improve the qualities of other sorts of cast-iron, chiefly in
the production of superior sorts of wrought-iron. It is
principally the manganese* of the spiegeleisen that has a
purifying action upon the inferior sorts of cast-iron, as it
combines with the impurities (sulphur, phosphorus, silicon,
&c.), and separates with them on the surface of the liquid iron;
an addition of Prussian spiegeleisen rich in manganese acts
in this manner when producing Bessemer steel from English
cast-iron, which always contains a certain amount of sulphur.
Caront has proved by direct experiments that an addition of
metallic manganese (93 Mn, 1'0 Fe, 5′5 C, 0'5 Si) removes
sulphur and silicon from cast-iron when re-melting it, and
also when converting cast-iron into wrought-iron and steel.
An addition of o'005 of manganese to steel combines with
the free carbon of the steel and improves its quality; a larger
addition renders the steel brittle and hard.
When converting cast-iron containing phosphorus into
wrought-iron in open fires or hearth fineries—an operation
which is termed by Dr. Percy the finery process—and giving
an addition of spiegeleisen, phosphate of manganese is said
to be formed, which cannot be reduced even at the highest
temperature, whilst phosphate of iron is converted into
phosphide of iron. When refining cast-iron containing man-
ganese, part of the manganese oxidises before the carbon is
separated, and the protoxide of manganese formed also causes
the oxidation of silicon; silicate of manganese will then result,
having the property of combining with the sulphides present,
such as sulphides of iron, manganèse, &c., and these com-
LEOBEN., Jahrb., 1862, xi., pp. 291, 295.
1862, p. 320. Preuss. Ztschr., 1856, iii., 268.
†DINGL., Bd. 168, p. 380.
B. u. h. Ztg., 1861, p. 159;
Oesterr. Ztschr., 1861, p. 374.
T 2
276
IRON.
pounds then enter the slag. Sulphide of manganese also
combines direct with protoxide of manganese; therefore
the favourable influence of manganese does not consist
in its combination with iron, as was formerly supposed,
for this combination has, on the contrary, an injurious
effect.
Certain kinds of white iron, also of a crystalline lamellar
texture, but containing certain compounds which interfere
with its conversion into malleable iron, must not be con-
founded with spiegeleisen of good quality. According to
Scheerer,* some spiegeleisen contains a certain amount of
silicide of arsenic, tersulphide of arsenic, and sesquisulphide
of aluminium; this does not prevent the formation of Fe,C.
Huenet and Karsten state that spiegeleisen containing
sulphur may be produced by fusing grey iron with sulphur,
thus forming sulphide of iron, and separating carbon at the
same time, the carbon serving to saturate the undecomposed
carbide of iron, and the spiegeleisen results when the mixture
is quickly cooled. It is obvious that the formation of the latter
spiegeleisen is influenced by circumstances other than those
which lead to the formation of the real spiegeleisen, and this
is proved by the behaviour of grey iron fused with sulphur
when slowly cooled, when a grey iron rich in graphite results.
Varieties of spiegeleisen are sometimes produced from impure
iron refinery|| and puddling slags.
b. Fine Grained, Flowery Pig-Iron (Blumige Floss)
and Porous White Pig-Iron (Luckige Floss).-The
former contains 4 or 5 per cent, and the latter 3 or 4 per
cent, of carbon in chemical combination, and up to o'5 per
cent of silicon; their specific gravity is from 75 to 76.
These varieties of iron are usually silver white. The
fracture of flowery pig-iron shows either small indistinct
crystals or a bluish colour, and a radiated, fibrous, or flowery
texture.
* SCHEERER's Métallurgie, ii., 54.
+ DINGL., Bd., 85, p. 374.
KARSTEN'S Eisenhüttenkunde, 3 Aufl., i., 427.
Oesterr. Ztschr., 1861, p. 37.
FLOWERY AND POROUS WHITE PIG-IRON.
277
Their composition is shown by the following analysis :-
I.
II.
III.
Fe
94.68
92.II
95'715
Graphite
C, combined
0'770
3·83
5'05
I'750
S
0'02
trace
0.261
P
0'04
0°429
Si
0'41
0.83
O'939
Mn
0*98 2'00
0*166
Cu
0'020
Mg
Ca
Ο ΟΙ
0°07
No. I is porous white iron from Eisenerz, analysed by
Wildtermann.
No. 2, radiated white iron from Vordernberg, analysed by
Mayrhofer.
No. 3, radiated white pig-iron from Gittelde (Hartz),
analysed by Fr. Werlisch.
These kinds of white pig-iron are produced from pure ores
which are easy to fuse and to reduce, when not employing a
temperature high enough completely to saturate the reduced
iron. The produced pig-iron is poorer in carbon and more
difficult to fuse the lower the temperature employed in its
production; but it is also more difficult to conduct-the pro-
cess is so difficult, indeed, that usually the porous white iron
can only be produced with interruptions, as otherwise the
blast furnace would become disarranged. The lowering of
the temperature is effected by more or less high charges and
by a less strength of blast. This iron is less hard than
spiegeleisen, and the blisters contained in the porous white
are of different sizes. If the temperature is still further
lowered the resulting product usually contains up to 3 per cent
of carbon, has a lamellar grain, showing iridescent colours
inside the blisters, and is almost malleable and capable of
being welded. Flowery and porous white iron form thick
liquids upon fusing, the latter more so than the former, so
that the silicide of manganese, gases, &c., formed cannot be
278
IRON.
completely separated on the surface, and the gases then give
rise to the formation of blisters. These gases, passing through
the rather pasty surface of the thick liquid iron, carry iron
along with them, which burns in the atmospheric air with
scintillation.
These kinds of white iron consume for their production
from 10 to 15 per cent less fuel than white iron from the
regular process (gaares Eisen), and are more quickly converted
into malleable iron, owing to their small amount of carbon
and their property of forming a thick liquid upon melting.
On the other hand, their nature makes it more difficult to
fix the limit of decarbonisation for the purpose of producing
steel, which may be easily done when working spiegeleisen.
They cause a greater loss of iron when treated in puddling
furnaces than when treated in hearth furnaces. When
heating these kinds of iron with substances containing
oxygen, such as peroxides of iron, manganese, &c., they
are easily transformed into malleable cast-iron.
Two modifications standing between spiegeleisen and the
flowery white pig-iron, namely, a radiated crystalline white
iron and a granular crystalline iron, may be produced from
the same admixture of iron ore at suitable temperatures.
The former modification resembles spiegeleisen as regards.
its contents of carbon, and the latter is more like flowery
pig-iron in that respect. Both sorts are very hard, and, upon
melting, form sufficiently thin liquids to allow a profitable
treatment for the production of wrought-iron and steel, both
in puddling furnaces and in open fires or hearth furnaces.
Mayrhofer represents the composition of the different sorts
of white iron by the following formulæ ; however, the amount
of carbon appears to be somewhat too low.
Spiegeleisen
Radiated crystalline
Fe4 C containing 5*10-5°33 per cent of C.
white pig-iron .
Fe, C
""
Grained
Fes C
""
3*45-3'62
2.61-2'74
""
وو
Flowery
Porous
Fe, C
Fe,₂C
2*10-2*20
""
""
1'75-1.84
WHITE PIG-IRON OF THE REGULAR PROCESS.
279
c. White Pig-Iron Produced by the Regular Process.—
We give some analyses showing the composition of this sort
of iron :—
Carbon, graphitic`
I.
II.
III.
IV.
V.
0°583 0.887
combined 2'90
I'91 2.217
2'451
3°50
Sulphur
I'II 0'015
2.516
trace
Phosphorus. O'II
o'913 0*27
Silicon.
0'77
I'OI 0.951
I'124 0.62
Manganese
I'73
2.715
Aluminium
0'27 0*06
Calcium
Magnesium
0'51
0'48
Nitrogen.
•
Sand
Iron
0'72
0'502
• 93'90
89.863 95*32
No. 1 is white iron produced from magnetic iron ore.
No. 2, white iron from Creusot, analysed by Schafhäutl.
No. 3, white iron from Lancashire, produced from red hæma-
tite, analysed by Miller. No. 4, white iron from Gartsherrie,
analysed by Gurlt. No. 5, white iron produced from puddling
cinders by Lang and Frey's method.
White iron containing a small quantity of sulphur and
phosphorus (from traces up to 2 per cent), and consequently
a smaller amount of carbon (from 1 to 4 per cent, usually
from 2.5 to 3.5 per cent), is produced, from pure iron ores
when using as fuel coke containing sulphur; from iron ores
containing some little sulphur and phosphorus ; or from ores
which are more difficult to reduce when keeping a moderate
temperature in the furnace hearth, as when producing spie-
geleisen. The resulting slags are poor in iron, as nearly all
the oxide of iron is reduced. The amount of manganese
which the ore mixture always contains renders the mixture
easier to fuse, and has a purifying reaction upon the pig-iron,
fitting it for the production of a wrought-iron of good quality.
The amount of silicon in this pig-iron is not much larger
than that in spiegeleisen, and, at all events, is smaller than
that contained in the mottled and grey sorts of pig-iron,
280
IRON.
seldom exceeding o'5 per cent, and I per cent at the very
highest.
White iron poor in carbon and containing phosphorus and
sulphur and a considerable amount of silicon, is produced
when smelting ores with a larger amount of sulphur and
phosphorus, even at a higher temperature in the furnace
hearth; the appearance of the slag is the same as in the
regular process. Such a white iron produced with coke, at
Firmy, contained 4'1 per cent of silicon, o 3 per cent of sulphur,
1°4 per cent of carbon, and 2.3 per cent of phosphorus.
White iron produced from puddling cinders* is either fit
or unfit for the production of wrought-iron according to the
purity of the cinders.
Pig-iron containing silicon and phosphorus assumes a light
bluish colour, and a dirty dark yellow or brown tint when
containing sulphur. It assumes a steel grey colour if its
amount of carbon is smaller, when the crystalline or flowery
texture will also become granular. The lustre of the white
iron decreases as it grows darker. Similar tints may be
formed by variable quantitative combinations of those sub-
stances, so that even an experienced eye cannot judge the
nature of a white iron by its external appearance; moreover,
it requires a knowledge of the mode in which the iron was
produced. The common white ron sometimes has a con-
choidal, and rarely a granular fracture, whilst the flowery
iron has a splintery fracture.
d. White Iron Produced by an Irregular Process
(Grelles Roheisen).-An irregular process of an iron blast
furnace, which in German is termed Rohgang, and consists
in an imperfect reduction of the oxide of iron, may be brought
about by different circumstances, and is indicated by a
ferruginous, thinly liquid slag. This oxidised iron in the
slag renders the ore mixture easy to fuse, thus facilitating
the formation of white iron, and also has an oxidising re-
action upon the carbonised iron, causing the formation of
white iron poor in carbon, and more or less thickly liquid.
* B. u. h. Ztg., 1862, p. 320.
CHILLED WHITE PIG-IRON.
281
When produced from pure iron ore, this iron may be fit for
the production of wrought-iron, but this process cannot be
carried on for any length of time without corroding the furnace
walling and seriously disarranging the process. Though pro-
duced at a proportionally low temperature, this iron usually
has a larger amount of silicon than the kinds of white iron
produced by a regular process, because less carbonised iron
combines more readily with silicon, sulphur, and phosphorus
than iron with a larger amount of carbon. This is proved
by the following analyses of iron from Königshütte on the
Hartz, by Kuhlmann :-
Foundry iron produced by the
regular process
The same from the same ore
mixture produced by the ir-
regular process
Forge iron produced by the
regular process
Forge iron produced from the
same ore mixture by the ir-
regular process
C. Graphite. Si. S.
P.
0.98 1.87 3'44 0048 0*163
2*02 077 3.82 0.081 0'213
0'55 244 315 0'009 0'067
I'97 0'93 4'09 0'019 0'093
When treating iron ores rich in sulphur and phosphorus
under such circumstances, a very inferior iron, unfit for forging,
is produced; it sometimes remains liquid, owing to its amount
of sulphur and phosphorus, and not on account of its con-
tained carbon..
e. Chilled White Iron.-This iron results from grey iron
when suddenly chilled; it has a radiated texture, and is
whitish grey or silver white in colour. According as the
grey iron has been produced at a more or less high tempera-
ture, the chilled iron differs from the varieties of original
white iron in its great hardness, which exceeds that of the
hardest steel, its lower specific gravity, its property of
liberating some graphite when dissolved in acids, and its
resumption of the grey colour when re-melted and allowed
to cool slowly. The chilled white iron also has a lower
specific gravity than the original grey iron, probably owing
to the circumstance that the grey iron expanded by the high
282
IRON.
temperature is not able to contract perfectly upon chilling,
and its volume therefore remains permanently enlarged.
This artificial expansion of the white iron causes a great
molecular tension, which is rendered sensible in its consider-
able hardness; tempering the iron destroys that tension and
causes the iron to lose its hardness. A molecular change* takes
place in the grey iron on sudden chilling, as is the case in
the hardening of steel, and according to Caron,t part of the
graphitic carbon becomes transformed into combined carbon.
According to Deville, grey iron contains crystallised carbon,
and chilled white iron amorphous carbon. If the presence
of silicon is disregarded, chilled white iron produced from
pure iron ores is better fitted for conversion into wrought-
iron than the original grey iron, which, owing to the graphite
it contains, can only be converted with great difficulty. But
if the grey iron contains much sulphur and phosphorus, it
is better to work it direct without chilling it, as the longer
time required for its conversion into malleable iron affords
more opportunity for the removal of the foreign substances;
the resulting malleable iron will be of a better quality than
when operating with white iron, which is quickly converted
into wrought-iron.
When casting grey or mottled iron in damp sand moulds
or in iron moulds (chills), the castings become superficially
hard and white, and for some purposes (hard rollers, railway
wheels) this is desirable. Upon re-heating these castings to
a strong red heat, and slowly cooling them (tempering), they
become softer, owing to a re-arrangement of the molecules,
and possibly to a partial separation of carbon.
All kinds of pig-iron become white when suddenly chilled,
and iron poor in carbon most readily so. Phosphorus, sul-
phur, and manganese, if contained in larger quantity in the
grey iron, facilitate its conversion into white iron; manganese
renders the ore mixture more fusible, and sulphur and
phosphorus lower the melting point of pig-iron.
Grey Iron. The following analyses show the composition
of some sorts of grey and mottled pig-iron :
**DINGL., Bd. 168, p. 36.
B. u. h. Ztg., 1863, p. 304.
GREY PIG-IRON.
283
A. Iron with a large amount of Carbon.
Fe
Graphite
I.
II.
III.
IV.
V.
92*22 88.983
2.II
I'99 2 45 3156 3574
C, combined 2.17 2.78 2'00
I'347 0.858
trace o'18
1267 0002
I'23 0.25
0.842 0*602
0'09 0'71
2°23 2°721
2.23
2.721 2.301
S.
P.
Si.
Mn
Al.
trace 0'21 2°401 0*066
trace 0°40
No. 1 is mottled iron from Liezen, analysed by Buchner.
No. 2 is mottled iron from Königshütte (Hartz), produced
with charcoal and cold blast; according to Gurlt its formula
is Fe,C+FesC+ FeSi+Fe,P. No. 3, Scotch pig-iron pro-
duced with hot blast, analysed by Schaffhäutl. No. 4, dark
grey iron from Gartsherrie; according to Gurlt, FesC. No. 5,
grey iron from Rothehütte (Hartz), produced with charcoal.
B. Grey and mottled Pig-Iron with less Carbon.
I.
II.
III.
IV.
V.
Fe
Graphite
C, combined
S.
P.
Si
Mn
Ca
Na
K.
96*484 96*44 94°10 95°23 89°314
1634 078 1'92 0'49 2.110
0.561 0.83 1'87 1'77 1793
0113 0°27 trace trace 1*480
0*156 0*16
1*052 1'52
0'21 O'12 I'171
1°30
1°30 0°31 2.165
I'12 0'34 1'596
0°05 ΟΙΟ
0.16 019
trace
No. 1, grey iron from Hasslinghausen (Westphalia), pro-
duced with coke, analysed by Lürmann. No. 2, mottled
iron from the same place and analysed by the same. Nos.
3 and 4, pig-iron from South Staffordshire, (No. 3 produced
with cold blast, No. 4 with hot blast), analysed by Wrightson.
No. 5, mottled iron from Gartsherrie; according to Gurlt it
is an admixture of Fe§C and Fe,C.
Grey iron is generally produced if the iron, reduced from the
ore in the furnace shaft (and after being perfectly carbonised,
so as to have the composition of spiegeleisen), is exposed in
284
IRON.
the hearth to a higher temperature than was required for its
reduction. According to Gurlt, part of the chemically com-
bined carbon will then separate as graphite,* and mix with
the lower carbonised white iron when slowly cooling, thus
forming grey iron. The higher the temperature to which
the liquid pig-iron is exposed in the hearth, the more
graphite will be separated, and the darker will be the re-
sulting iron; and the nearer this temperature approaches the
fusion point of pig-iron, the lighter will be its colour, as less
graphite will separate. Upon keeping this grey graphitic
iron, after fusion, for a longer time in a liquid state at a tem-
perature not much exceeding the melting point of grey iron
(1700-1750° C.), the graphite will gradually dissolve and the
iron assume its original constitution. The graphite has an
essential influence upon the state of aggregation and the
texture of the iron, chiefly by its form and state of division
in the mass. The chemical composition of the varieties of
grey iron does not, therefore, always correspond to the
physical properties of the iron.
Owing to the higher temperature required in the furnace
hearth for the formation of grey iron, other components of
the ore mixture, such as calcium, magnesium, aluminium,
and chiefly silicon, will also be reduced in larger quantities;
the grey iron is therefore richer in them than white iron,
which is produced at a lower temperature.
This larger
amount of foreign substances seems to facilitate the separa-
tion of graphite previous to the solidification of the iron,
and some of them, as, for instance, silicon, seem to combine
chemically with the graphite.
When producing white iron instead of grey from a mixture
difficult to fuse and employing a high temperature, this
phenomenon may be attributed to different circumstances.
A sudden lowering of temperature may have taken place,
causing a chilling of the pig-iron, or the unreduced iron may
have come into contact with the carbonised iron in the
hearth and reacted upon it, or the admixture may be too rich
in sulphur, phosphorus, or manganese. A small amount of
* KARSTEN'S Archiv., 1 R., viii., 43 ; xii., 91; xiii., 232; xv., 177; xvii., 118.
GREY PIG-IRON.
285
sulphur may be kept more perfectly from combining with
grey iron, than with white iron, owing to the higher tempera-
ture, and an amount of sulphur in grey pig-iron may be
better extracted by converting the pig-iron into malleable
iron, than such an amount in white iron, as the white iron
is converted too rapidly into malleable iron. It may there-
fore be advisable to produce a purer grey iron from ores
which are not sufficiently pure, regardless of the greater
consumption of fuel, instead of producing a more impure
white or mottled iron at a small consumption of fuel.
Gurlt considers (page 272) the normal, granular, grey iron
to be an octocarbide with intermingled graphite, FesC+C,
which results by heating spiegeleisen, Fe,C, to a certain
temperature above its melting point, and cooling it slowly-
2Fe,C=FesC+C.
Gurlt also presumes that the crystals sometimes formed in
cavities of castings are octocarbide. Tunner considers
these crystals to be pure iron, and his theory has been con-
firmed by an analysis made by Richter, in so far as to prove
that the crystals are not octocarbide; the crystals contained—
Si .
Al .
S.
As.
P.
C, combined
Graphite
Mn
Fe.
0*972
0*565
0*008
0'005
0'021
o'967
2'122
Ι'ΟΙΟ
•
94'330
If the temperature is not sufficiently high to convert the
tetracarbide into octocarbide, less graphite will be separated
and a mixture of both carbides formed, as exists in the
varieties of mottled iron (3Fe,C=Fe,C+ FesC+C). The
mottled iron may have a darker or lighter shade, and shows
grey or black points of different size on a white or whitish
grey ground.*
*B. u. h. Ztg., 1857, p. 366.
286
IRON.
Mayrhofer considers the different sorts of pig-iron to be
composed as follows :-
P.C. of
Combined
Carbon.
2.61-2'74
2°10-2˚20
Mottled granular pig-iron. Fe, C+C containing 3'45-3*62
Mottled radiated pig-iron.
Granular grey iron.
Darkish grey iron
Feg C+nC
Fe₁C+nC
Fe₁₂C+nC
175-184
Though his suppositions are partly hypothetical they
nevertheless furnish a more probable explanation of the pro-
duction of the different sorts of pig-iron than any formerly
entertained.
White iron always contains more carbon than the grey
iron resulting from the same tapping of the hearth of a fur-
nace, for the reason that grey iron upon slowly cooling
liberates some graphite on its surface, whilst the carbon is
uniformly disseminated throughout the whole mass of the
white iron.
The varieties of light grey, and mottled fine granular pig-
iron, produced from purer iron ore at a lower temperature,
contain from 1 to 1 per cent of chemically combined carbon,
from 1 to 2 per cent of graphite, from to 1 per cent of
silicon, and traces of sulphur only, supposing that the iron
is not of too light a colour and too dull a lustre. Iron of a
similar appearance results, at a higher temperature from ores
containing sulphur and phosphorus; this iron is sometimes
poorer in carbon, and may contain from 1 to 3 per cent of
silicon, and as much as 2 per cent, or even more, of sulphur
and phosphorus; phosphorus renders the iron brittle if pre-
sent in larger quantity.
1
The common grey granular pig-iron produced from a purer
ore mixture at the regular process, and at a somewhat higher
temperature than that applied to the production of light grey
iron, contains from 2 to 3 per cent of graphite, from o'5 to
to 24 per cent of combined carbon, from o˚5 to 2 per cent of
silicon, and only small amounts of sulphur and phosphorus.
If produced from ores containing sulphur and phosphorus,
and at a still higher temperature, it contains nearly the
same amount of combined carbon and graphite, but is richer in
silicon (1 to 3 per cent), sulphur, and phosphorus. It contains
THE PROPERTIES OF GREY IRON.
287
less sulphur than the light grey iron produced from the same
ore mixture, if other circumstances do not interfere.
Dark (blackish) grey pig-iron, produced from purer ores at
a very high temperature, gives up upon cooling a considerable
amount of graphite (3 or 4 per cent), whilst it retains from
I to 1 per cent of carbon in chemical combination. Its
amount of silicon is from 1 to 2 per cent; if this amount
increases the carbon will decrease (Scotch pig-iron).*
Blackish-grey iron, produced from an ore mixture containing
sulphur and phosphorus, contains less of these substances
than common grey iron produced from the same mixture;
a larger amount of sulphur and phosphorus limits the
amount of carbon in the iron. The more or less dark colour
of this iron is derived not only from the absolute amount of
graphite but also from the size and arrangement of the
laminæ of graphite.
Grey and white iron sometimes occur distinctly separated
in the same piece.
The Properties of Grey Pig-Iron.
Texture and Strength.-Grey iron may vary in colour
from lightish grey to blackish grey, and the darker it is the
more lustre it possesses. Dark, dull, and hard iron, without
lustre, is usually produced from an ore mixture too poor in
silica and therefore rich in earthy bases (calcium, magnesium,
aluminium); whilst grey iron of a light colour and little
lustre yields friable castings and inferior wrought-iron,
owing to the presence of phosphorus and silicon.
The texture of grey iron varies between coarse grained
iron and compact iron; the lighter varieties have the finest
grains if containing an excess of graphite the grains are
coarse and irregular. The pig-iron is better, the darker,
brighter, and more serrated its fracture; a slight lustre, an
ashey grey colour, and a scaly texture indicate an inferior
iron. If the surface of the iron is artificially crystallised the
crystals which the different sorts of grey iron form exactly
correspond to the interior aggregation of the iron-which may
be observed by means of a microscope-and the aggregation
* B. u. h. Ztg., 1862, p. 323.
288
IRON.
again corresponds to a certain strength. At Ilsenburg
(Hartz), Schott* applies these observations practically to
judging the varieties of pig-iron produced from different ore
mixtures.
Crystals in cavities of pig-iron are hard and difficult to
fuse; their composition is shown on page 285.
The purer sorts of iron are less hard the more grey the iron
is; these varieties may be drilled, filed, and sawn. Impure
light grey iron, especially if containing much phosphorus,
sulphur, and earthy bases, is usually hard, but if con-
taining a large amount of silicon it is sometimes soft and
friable.
The tensile strength of grey pig-iron is greater than that
of white pig-iron in good qualities by about one-third, the
transverse strength is twice as great as that of wrought-iron,
and the crushing strength is less than that of white pig-iron
and greater than that of wrought-iron.
According to Guettier* the following appearances are
to be noted on the fractures of different classes of foundry
iron :-
A moderately large grain of slight lustre, mottled with
fine patches and with a tendency to whiteness, indicates the
highest degree of resistance.
A smaller grain, but likewise dull, and with a mottled
grey base, marks the quality of metal best suited to resist
tensile strain.
A somewhat fibrous grain, terminating in fine, jagged,
pyramidal points on the fractured surface, with a close,
regular base, indicates great transverse strength.
A fine-grained grey iron, bordering upon mottled iron
when the fracture is small-grained and even, and not in flat
broad plates, is the best for resisting compression.
The varieties presenting least resistance are those that are
full of graphite, of a blackish grey colour, and large brilliant
grain, or of an irregular grain upon a shining base, and the
mottled white kinds in which there is no granular structure.
apparent.
* Bauerman's Treatise on the Metallurgy of Iron; London, 1868; p. 236.
THE PROPERTIES OF GREY IRON.
289
Repeatedly re-melting pig-iron causes a chemical change
in its composition and in its texture, and greatly influences
its transverse strength and resistance to crushing. Guettier
found that No. 1 Scotch pig reached its maximum strength
after the eighth melting. Fairbairn* found that the same
point was reached with No. 3 pig (Eglinton) after twelve
meltings.
Besides the carbon, an amount of foreign substances in
iron, such as sulphur, phosphorus, silicon, &c., influences the
strength of the iron; the latter substances usually lessen its.
strength, therefore pig-iron produced with hot blast being
richer in silicon is less strong than iron produced with cold.
blast; whilst, on the other hand, a small amount of sulphur
may increase the strength of iron, rendering it poorer in
graphite and denser in the castings, as, for instance, in the
Swedish gun iron.
The Specific Gravity of the pig-iron depends on the
quality and quantity of the carbon and of the other foreign
substances present. The specific gravity of the grey iron
(7.1) is less than that of the white iron (76). Charcoal pig-
iron has a greater specific gravity than coke pig-iron of equal
texture and colour, and the specific gravity increases upon
re-melting the iron. Grey iron containing phosphorus has
the lowest specific gravity. A small amount of sulphur and
mechanical pressure seem to increase the specific gravity.
The blackish grey iron possesses a low specific gravity, the
mottled iron the highest specific gravity.
When tempered, grey iron becomes softer, just as white
iron (page 282), but also friable. If repeatedly tempered the
increased volume of the iron becomes permanent (about
1-24th of the original length of the iron), and, according to
Erman and Herter,† the degree of this expansion seems to
increase as the amount of graphite in the iron increases, and
to decrease the more combined carbon the iron contains.
This behaviour of iron is practically applied to castings,
*
Report of the British Association for the year 1853.
+ B. u. h. Ztg., 1856, p. 152. Bgwkfd., xix., 205.
DINGL., XXxiii., 76, Oesterr. Ztschr., 1855, p. 211.
pp. 57, 189.
VOL. II.
U
B. u. h. Ztg., 1855,
290
IRON.
such as cannon-balls which have been cast too small. The
required volume is obtained by repeatedly heating the
castings. Chilled white iron becomes soft and grey upon
tempering, and resembles the grained light grey iron.
The Melting Temperature of the grey pig-iron is usually
between 1600° and 1700° C. Before reaching this tempera-
ture the grey iron, for a short time, is in the state required
for welding, and this circumstance is sometimes made use of
to correct defects in casting.
The liquid mottled iron shows a silver-white colour, and
if the iron is very hot, hues are formed on its surface by more
or less distinct and evanescent figures; these figures are fre-
quently seen until the iron solidifies; the iron afterwards
shows larger or smaller dark points on its surface, and the
founder judges the iron from this appearance. The mottled
iron is very thinly liquid, fills the moulds well, and upon
slow solidification it forms dense and compact castings which
are superficially even, fine grained, of a more or less uniform
texture, light in colour, and sufficiently soft to allow a further
mechanical treatment, such as boring, &c.
The liquid grey iron produced by the normal process
varies, according to the degree of its temperature, between
milk white and silver white, and is more thickly liquid the
more graphite it contains. It solidifies without showing
hues and figures as observed in the mottled iron, or they are
very slight, and upon slowly cooling it has a uniformly even
surface, with occasionally some graphite upon it which has
separated from the iron. This iron is grey in colour, has a
coarse, brilliant grain, and is softer than the mottled iron.
The blackish-grey sorts of pig-iron form very thick liquids,
and, upon solidifying, usually show parallel striæ on their
surface, and the borders of the pig rise over the centre; their
fracture is blackish grey, of a peculiar coarse grain, and very
brilliant, with pre-eminently brilliant grains.
Mottled iron, very light in colour, so as to approach white
iron, shows, when fused, a reddish colour; it is pasty and
forms a thick liquid, and the spots on its surface become
larger when cooling. In white iron resulting from an
irregular process these spots are formed by blisters; even
THE PROPERTIES OF GREY IRON.
291
larger laminæ of silicide of iron and oxides of iron are fre-
quently separated. The surface of the solidified light grey iron
with a fine grained fracture without lustre is convex.
The hues and spots on the surface of liquid iron are
formed by the separation of foreign substances which are
partly oxidised and set in motion as the superficially cooled
parts of the pig-iron sink to the bottom, to make room for
hotter iron which rises. These segregations also form the
larger or smaller spots and blisters on the solidifying iron,
and they are formed in larger quantity if the iron is very
impure or produced by an irregular process, in which case
the iron is unfit for castings. Pig-iron containing sulphur
usually shows beautiful hues (Altenau Iron Works, Hartz),
and the design of the figures seems to stand in intimate
connexion with the nature of the iron, so as to allow certain
qualities of the iron to be ascertained by those figures.*
I
The pig-iron occupies a smaller space after cooling than
when in a liquid state; it contracts in such a manner that
at the commencement of its solidification it first expands so'
as to be capable of filling up the smallest cavities and
depressions of a mould, but after solidifying it contracts.
Grey pig-iron contracts by about 1 per cent, and white pig-
iron from 2 to 2 per cent; white, thickly liquid pig-iron
resulting from an irregular process expands instead of con-
tracting. When making models for castings we must take
into consideration the contractiont of the iron, and make the
models correspondingly larger. In the iron foundries of the
Hartz the contraction of grey and mottled iron is considered
to be inch to a foot, and as strongly mottled iron contracts
somewhat more than grey iron, an extra allowance of about
inch to 12 feet is given for contraction when making long
castings. To prevent the castings from cracking in conse-
quence of the contraction, they must be uniformly cooled on
their thicker and thinner points. Large castings sometimes
burst suddenly without any visible cause, frequently with a
*
Bgwkfd., vi., 241.
† Bgwkfd., xii., 447. DINGL., Bd. 132, p. 392. B. u. h. Ztg., 1854, p. 101.
Mittheil. d. Hannov. Gew.-Ver., 1853., Hft. 4; 1854., Hft. 1.
U 2
292
IRON.
detonation, after they have been in use for some time, or
when lying without being used; this bursting is, perhaps,
due to a quick or non-uniform cooling, which causes a tension
in the casting.
The grey pig-iron oxidises more easily, and is more readily
attacked by acids than is white iron; its solubility in acids
stands in inverse proportion to its amount of carbon. At
a common temperature dilute muriatic and sulphuric acids
will hardly attack spiegeleisen, and grey iron only slowly,
whilst dilute nitric acid has a somewhat stronger action;
the dilute acids react violently at boiling heat. Concentrated
muriatic and sulphuric acids dissolve iron in a shorter time,
chiefly at a higher temperature, whilst concentrated nitric
acid scarcely acts upon white pig-iron; but it completely
attacks the grey iron. Hydrogen gas is evolved when using
dilute sulphuric and muriatic acids, and combines with the
chemically-combined carbon, sulphur, phosphorus, &c., thus
causing the peculiar odour. Hahn's investigations showed
that, upon dissolving pig-iron, several combinations of car-
buretted hydrogen of distinct composition (chiefly Can H2n),
both liquid and in the form of gas, are produced. The
residue contains all the graphite and part of the chemically-
combined carbon, as well as the other various impurities of
the iron.
The analyses by Schafhäutl* of the residues obtained by
dissolving white and grey iron in muriatic acid will explain
the nature of the different sorts of pig-iron better than its
physical properties.
Upon dissolving white pig-iron, a brown residue remains;
this residue is a distinct chemical combination of iron,
silicon, carbon, hydrogen, and occasionally nitrogen, and has
the form of brilliant scales when seen by means of a micro-
scope. Boiling caustic potash reacts upon it but slightly,
only extracting some silica. Carbon can only be separated
as carburetted hydrogen by a longer treatment with muri-
atic acid, and iron can be extracted by a repeated heating
and subsequent treatment with muriatic acid, thus forming
ERDM., J. f. pr. Ch., xix., 159; xx., 465; xxi., 129; lxvii., 257. DINGL.,
cliii., 349. B. u. h. Ztg., 1861, p. 38.
THE PROPERTIES OF GREY PIG-IRON.
293
silica; this behaviour shows that the residue contains an
intimate combination of SiC and FeSi. Neither caustic
potash nor caustic ammonia evolve hydrogen from the
residue. After removing all soluble parts, a small amount
of graphite sometimes remains.
The residue of grey iron behaves differently; it is of a grey
colour, and evolves hydrogen when treated with the above-
named reagents. Through a microscope it is seen to be a
mixture of graphite and gelatinous silica and oxide of
silicon, and these latter substances may be completely
extracted by means of caustic potash; graphite will then
remain.
As the residue from dissolving grey iron contains all the
silica free from carbon and iron, we may conclude that the
silicon has been in direct combination with the iron.
Schafhäutl supposes the evolution of hydrogen to be caused
by a certain amount of aluminium which is always contained
in the residue of grey iron, but it is more probable that the
presence of oxide of silicon*, which, according to Wöhler,
is always formed upon dissolving pig-iron in muriatic acid,
is the real cause. According to Calvert,† when dissolving
grey iron, all acids except aqua regia form SiH(?), which is
otherwise only observed when dissolving silicide of alu-
minium. The manner in which other bodies contained in
the iron influence the solution of iron by acids has not been
sufficiently investigated. Calvert has analysed the graphitic
residue (a) obtained from a cube of pig-iron (b) upon which
acetic acid had reacted for two years:
C
N
Si.
Fe.
S
P.
Loss
a.
b.
II 020
2'900
2'590
0*790
6'070
0*478
79'960
95'413
0096
0*179
0*059
0'132
0'205
O'IOS
Schafhäutl supposes that graphite does not exist as such
in pig-iron, and is formed when separated in the same
* Ann d. Chem. u. Pharm., liv., 374.
+ Polyt. Centr., 1861, p. 1071. B. u. h. Ztg., 1862, p. 255.
294
IRON.
manner as the residue of white iron, and that it contains
SiC; this is also the opinion of Wehrle. Karsten considers
the graphite to be pure carbon.
Struve* has investigated the residues of pig-iron containing
phosphorus, and found that phosphorus is contained in iron
in different combinations.
The Applications of Grey Pig-Iron.-Its properties of
becoming thinly liquid and of expanding just before solidifying
make it particularly fit for foundry purposes; its softness
also contributes to its utility. The varieties of mottled iron
produced from pure ores yield the strongest, densest, and
most distinct castings, owing to the absence of separated
graphite. If a great softness of the castings is required, or,
if impure iron ores which are difficult to fuse are to be
applied, it is better to produce a purer good grey iron at a
higher temperature, which is then made use of for castings,
either direct or after being re-melted in a cupola or reverbe-
ratory furnace, in order to lessen its amount of graphite.
Blackish grey, thickly liquid pig-iron always requires a re-
melting and mixing with cheaper sorts of iron poor in
graphite.
The varieties of grey iron are generally less adapted for
the production of wrought iron and steel, as they require a
longer time for conversion, owing to their amount of graphite,
which is difficult to oxidise, and to their melting in a thinly
liquid state; as they contain a larger amount of silicon the
loss of iron also increases. Circumstances may, however,
necessitate the use of grey iron in the production of wrought-
iron; for instance, if the same blast furnace has to furnish
the material for foundry and forge purposes at the same time,
or if impure ores or pure iron ores which are difficult to reduce
(magnetic iron ore) are to be smelted, and so require a higher
temperature.
In order better to adapt graphitic pig-iron for conversion
into malleable iron, it is sometimes chilled by being run into
cast-iron moulds. It is sometimes previously treated by an
oxidation (refinery) process partly to free it from impurities.
* ERDM., J. f. pr. Ch., Bd. 79, p. 321. Polyt. Centr., 1860, p. 891.
CLASSIFICATION OF PIG-IRON.
295
Though the foreign substances contained in an impure.
grey pig-iron may be more completely separated than from
equally impure white pig-iron at a larger consumption of fuel,
and at a greater loss of iron (caused by the longer time re-
quired for the conversion into the malleable state), grey iron.
may, on the other hand, be so impure as to be unfit for the
purpose. But the impure grey iron may nevertheless be well
fitted for foundry purposes, its chemical combination being of
less importance for castings than its density and softness.
The different sorts of grey pig-iron are classified according
to their amount of graphite, and also according to their
adaptability to foundry or forge purposes. The gradations
in the scale are mainly dependent upon the colour, the degree
of greyness, the texture or size of the crystalline plates, and
their uniformity and lustre. The largest grained brilliant
and graphitic dark grey iron is known as No. I pig, while
the smaller-grained varieties with diminishing lustre and
colour are distinguished by succeeding numbers up to No. 4•
Beyond this point, when the metal ceases to be grey, the
numerical scale is not used, the remaining qualities being
known as mottled, with, in some instances, a further division
into weak and strong mottled and white, the last being the
lowest. This classification is subjected to slight variations
in different districts, as in the following examples of scales
used in different parts of England :—
Cleveland. . . Nos. I, 2, 3, 4, forge, mottled, white.
Lancashire hæmatite,, I, 2, 3, 4, 5 mottled, white.
The grey numbers, as far as No. 3, are also called foundry
or melting pigs; the lower qualities, which are only fit for
conversion into malleable iron, come into the class of forge
pigs. In Lancashire and Cumberland two extra classes are
made, known as Bessemer iron Nos. I and 2. These command
higher prices than the same numbers in the ordinary scale.*
As the white pig-iron may also be the product of an irre-
gular process, grey iron, which can only be produced by a
regular process, is frequently bought in preference to it.
It has been already mentioned, that certain impurities
contained in pig-iron, either isolated or in combination with
* BAUERMAN, Treatise on Metallurgy, &c., p. 231.
296
IRON.
each other, sometimes have a favourable influence, and that
now and then these impurities are deleterious to the iron when
applied to foundry and forge purposes; the external properties
of the iron do not always allow the extent of this influence
to be detected, but the operator must be assisted by experience
and chemical analyses.
The qualities of pig-iron are modified by the following
substances, which may have been derived from the ores, the
fluxes, or the fuel :-
1. Silicon. Nearly all varieties of pig-iron contain silicon,
which is reduced chiefly in the hearth of blast furnaces, from
the silica of the mixture, or from the ash of the fuel, by the
action of coal and particularly of iron; and more silicon
is reduced the more siliceous the mixture is, the more inti-
mately the silica is mixed with the ore, and the higher the
temperature of smelting. It has been observed in some iron.
works that the amount of silicon in pig-iron does not increase
with the increase of silica in the ore mixture. This behaviour
seems to depend on the amount of alumina present in the
mixture; and this has also been remarked by Abel. The
ore mixture in some iron works of the Upper Hartz has
lately been richer in lime and poorer in alumina, and the
pig-iron contains more silicon than that which was formerly
produced. The sorts of white pig-iron seldom contain more
than per cent, but the grey sorts may contain as much as
3 per cent of silicon and more. Carbon and silicon seem to
replace each other in the pig-iron, causing grey iron con-
taining a large amount of carbon to be poorer in silicon
than the same iron containing a less amount of carbon,
whilst the silicon and carbon in the iron may amount
to from 4 to 8 per cent and more,* according to the tempera-
ture of the furnaces and other circumstances. The varieties
of pig-iron, rich in carbon and silicon, are chiefly pro-
duced when employing a strongly heated blast, and coke,
or anthracite, and they will be richer in silicon the
more siliceous the ore mixture is.† According to Abel,‡
2
+
* B. u. h. Ztg., 1862, p. 323.
Ibid., 1856, p. 306.
‡ Ibid., 1863, p. 296. WAGNER'S Jahrsber., 1860, p. 37.
SILICON IN PIG-IRON.
297
silica is more readily reduced if alumina is not present in
sufficient quantity to neutralise it; this has been before
mentioned. Coke pig-iron is usually richer in silicon than
charcoal iron, and pig-iron produced with hot blast contains.
fromto per cent more silicon than iron produced with
cold blast, supposing the other conditions to be the same.
Grey pig-iron produced at Lerbach (Hartz) with coke and a
proportionally low temperature of hot blast, contained 5°5 per
cent of silicon; iron from Vorwärtshütte in Lower Silesia,
produced from a mixture difficult to fuse, contained 7 per
cent; and some sorts of Scotch iron contain as much as
13 per cent of silicon, whilst the carbon present diminishes.
to I per cent.
Upon dissolving the pig-iron from Lerbach, gelatinous
silica was separated in a considerable quantity; and it would
seem that this great amount of silicon in the iron originated
from the deficiency of alumina and the long continued
reducing action, owing to the very slow progress of the opera-
tion. The presence in the hearth of partially carbonised iron
seems to facilitate the combination with silicon, and for this
reason pig-iron produced by an irregular process poor in
carbon is richer in silicon. If iron, containing the highest
amount of carbon in the furnace hearth, comes into contact
with silica, more or less of this body will be reduced by the
carbon and the iron according to the temperature in the
hearth; the silicide of iron which is formed enters into
combination with the pig-iron which has lost part of its
carbon, and it facilitates the separation of graphite upon
cooling the pig-iron.
If the temperature is sufficiently high, the liquid iron rich
in silicon may still combine with a great deal of carbon,
but, upon cooling, the separation of graphite will be larger.
According to Mayrhofer, six equivalents of carbon replace
one equivalent of silicon. An abundance of slag in the furnace
hearth covering the iron, chiefly if siliceous, may extract
some carbon from the iron, at the same time replacing it by
silicon; a more basic slag is therefore produced when an
abundance of slag is required.
298
IRON.
According to the investigations of Schafhäutl, silicon seems
to be contained in pig-iron in different states of combination,
viz., as carburetted silicon, carbide and silicide of nitrogen,
silicide of iron, and sulphide of silicon, and the influence of
silicon upon pig-iron is different according to whether it is
present in one or the other of the states of combination;
these combinations are also found in the separated graphite.
The frequent occurrence of silicon in ferriferous bears* of
the copper smelting furnaces also indicates the affinity of
silicon to iron.
According to Schafhäutl, silicon exists in pig-iron likewise
in an elementary form, perhaps combined with a little carbon
and sulphur, and it may be retained for some time, even in
malleable and rolled iron until it burns with formation of
silica. Richtert has lately found crystallised silicon in pig-
iron, and it may be that the separated silica found in the
hearth of blast furnaces originates also from free silicon con-
tained in iron.
Dr. Hahn states that FeSi2, containing about 50 per cent of
silicon, is insoluble in all acids, even in hydrofluoric acid,
but it may easily be, decomposed by carbonate of soda; the
substance investigated by Richter was doubtless a similar
combination, as a small amount of iron was found in it.
Silicide of iron, containing 31 per cent of silicon, when in a
finely pulverised state is soluble in boiling muriatic acid, and
most readily in hydrofluoric acid; this silicide is more soluble
the less silicon it contains.
When casting pig-iron in cast-iron moulds (chills), it will
be found that the least silicon is contained in the lower part
of the pig-on the other hand, most of the combined carbon—
whilst the upper part of the pig is richer in silicon and other
substances which may separate; in such cases most of the
manganese will also be found in the upper part. Upon casting
pig-iron in moist sand the lower part of the pig is less
modified; it is therefore advisable to cast good forge pigs in
cast-iron moulds forming thin plates.
* ERDMANN'S J. f. pr. Ch., xvi., 196.
† LEOB., Jahrb., 1862, xi., 289. B. u. h. Ztg., 1862, p. 320.
SILICON IN PIG-IRON.
299
The following plans may be adopted for producing pig-iron
poor in silicon :-Employing low smelting temperatures and
carbonising the iron as perfectly as possible; or if a higher
temperature is required, employing admixtures of lime to
render basic the sufficiently aluminous mixtures from which
silicon is reduced with more difficulty than from siliceous mix-
tures poor in alumina; and, finally, adding manganiferous
fluxes. The latter fluxes partly render the mixture easier
to fuse, and the manganese combines with the silicon and
separates on the surface of the liquid iron.* According to
Lohage,+ manganese as well as aluminium also facilitates the
separation of silicon at the smelting of cast-steel.
Good grey foundry pig-iron may contain as much as 2 per
cent of silicon; if containing a larger amount it is harder
and less strong, but it may be improved by re-melting, when
part of the silicon will be separated; such iron containing
an excess of silicon is fine-grained, of light colour, has but
little lustre and solidifies quickly.
Forge pig-iron suffers more loss of iron by scorification.
the more silicon it contains; it may, nevertheless, be more
advantageous to produce pig-iron rich in silicon and poor in
sulphur when treating impure ores at a high temperature,
than by employing a low temperature to produce iron poorer
in silicon and richer in sulphur, as silicon may be more per-
fectly separated than sulphur. The state of combination of
the silicon essentially influences the behaviour of the iron at
its conversion into malleable iron.‡
Lohage|| suggests that grey forge iron should contain at
least 2 per cent of silicon in order to form a slag which shall
thus preserve the iron from further oxidation; and as the
silicon first oxidises, time is allowed for the separation of sul-
phur and phosphorus; forge iron containing above 1'5 per
cent of silicon is not usually desirable.
Bessemer considers a certain amount of silicon necessary
B. u. h. Ztg., 1862, p. 320. LEOB., Jahrb., 1861, X., 327.
† B. u. h. Ztg., 1861, p. 160.
Ibid., 1861, p. 38; 1862, p. 320.
|| Berggeist, 1860, No. 50.
300
IRON.
in the steel produced by his process in order to obtain solid
castings, but Tunner* thinks this unlikely.
Schafhäutlt suggests that a certain quantity of silicon is.
required for the formation of steel. Carbide of silicon will
be formed, which mixes with the fused carbide of iron and
separates more or less according to whether the mass solidi-
fies more or less quickly. Silicide of carbon being less soluble
in acids than carbide of iron, gives rise to the formation of
the silver white designs of Damaskeened steel upon treating
its surface with acids. Breant attributes these designs to
different carbides of iron which are soluble in various degrees.
Jullien considers the least amount of silicon to be detrimental
to the quality of steel.
These contradictory views necessitate further investigation.
According to Karsten, all good wrought-iron contains at
least 0.05 per cent of silicon, which amount may rise even to
O'I per cent without much prejudice to the strength of the
iron; this, however, will not be the case if the iron contains
0*37 per cent of silicon. Wrought-iron containing a larger
amount of silicon has not a uniform texture, being either
partly fibrous and partly grained, or in grains of different
sizes; a larger amount of silicon renders steel red-short.
According to Schafhäutl, a certain amount of manganese
facilitates the formation of steel in the following manner :-
When converting manganiferous pig-iron into the malleable
state, the carbide of manganese is first oxidised and after-
wards leaves in the residue so much silicon and carbon com-
bined with iron that the mass may be used as steel. If
manganese is absent too much silicon is at first oxidised, and
the steel is thus deprived of an essential component.
Buchner, Richter,|| and Eggertz have proposed newer
and apparently good methods for the quantitative determina-
tion of silicon in pig-iron.
* TUNNER'S Bericht über d. Londoner Industrie-Austellung, 1862; Wien,
1863, p. 77.
+ PRECHTL, techn. Encykl., xv., 366, 371, 385, 402. Bgwkfd., vi., 357.
KARST., Archiv., 1 R., v., 158. DINGL., Bd. 36, p. 245.
+-
Bgwkfd., xxi., 77.
LEOB., Jahrb., 1861, X., 503.
SULPHUR IN PIG-IRON.
301
2. Sulphur.-Smaller or larger amounts of sulphur, either
as sulphides or sulphates, are frequently contained in the
iron ores, fluxes, or fuel. The higher sulphides lose in the
upper parts of the furnace part of their sulphur, which com-
bines with iron if iron is present in a reduced state, and if
the temperature is sufficiently high. If mono-sulphides
produced either from polysulphides or from sulphates come
into contact with iron at the time of its reduction and car-
bonisation they will be decomposed, forming sulphide of iron.
The carbonisation of iron containing sulphur is rendered
difficult, partly on account of its being easily fusible, and
partly by its property of counteracting the absorption of carbon
even at a higher temperature (silicide and phosphide of iron
do not possess this property), therefore no carbonisation
takes place when the iron has passed into the hearth, and a
white iron poor in carbon and more or less rich in sulphur is
produced. Janoyer states that the formation of white iron
is facilitated by the circumstance that sulphide of carbon is
formed from the carbon and sulphur contained in the iron
owing to the higher temperature in the hearth, thus depriving
the pig-iron of carbon and making some of the heat latent
through the volatilisation of the sulphide of carbon.
If sulphide of carbon and sulphide of iron in the hearth
come into contact with the blast in the presence of moisture,
sulphuretted hydrogen gas will be produced, which partly
sulphurises the metallic iron and oxide of iron present in the
furnace above the hearth, and partly decomposes the com-
ponents of the slag, forming chiefly sulphide of calcium.
The sulphur in pig-iron does not seem to be always com-
bined with iron, but sometimes with silicon. Schafhäutlt
mentions an instance, that at the tapping of pig-iron in the
Tividale Iron Works, near Dudley, sulphide of silicon of the
composition Si₂S,, analogous to the oxide of silicon, was sepa-
rated in the form of a whitish yellow, spongy, earthy sub-
stance. According to Mayrhofer;‡ sulphide of silicon sepa-
rates like silicide of manganese at the solidification of iron,
Bgwkfd., xv., 361. B. u. h. Ztg., 1852, No. 20.
† ERDM., J. f. pr. Ch., Bd. 67, p. 257. DINGL., Bd. 153, p. 349.
+ LEOB., Jahrb., 1861, X., 325.
302
IRON.
causing the iron to be poor in silicon and proportionally rich
in combined carbon.
The sulphur is frequently not divided uniformly in the pig-
iron, and in the common pigs collects more in the upper part
than in the lower part.
As an amount of sulphur is more or less prejudicial to the
applicability of iron to almost any purpose, the sulphur must
be removed as much as possible. The remedies to be applied
depend partly on the state of combination in which the sul-
phur is present, and partly on whether the sulphur occurs.
chiefly in the ore, the gangue, the fuel, or the fluxes.
a. Sulphur occurring as sulphide.
a'. In the iron ore (iron pyrites, magnetic pyrites, zinc
blende, copper pyrites, galena). The removal of this sulphur
in a mechanical way (dressing by hand sorting) is first aimed
at, and afterwards by chemical operations (roasting, weather-
ing, and lixiviation). But as these remedies are not suffi-
ciently effective, other methods must be applied at the
smelting process; these are chiefly as follows:—
a. Additions of lime* whilst applying higher tempera-
tures, thus forming sulphide of calcium, which has the
property of dissolving other metallic sulphides (sulphides of
iron, manganese, &c.), and of sending them into the slag.
This reagent is more effective if the sulphur is contained less
in the ore than in the gangue, fuel, and fluxes, as in the
latter case sulphide of iron is not at first formed, but sul-
phide of calcium, which directly enters the slag. Slags
produced at Hattingen from iron pyrites of the coal measure
contained from 4 to 6 per cent of sulphide of calcium.
B. Manganiferous Fluxes.-As we have before mentioned
manganese combines with sulphur, and the sulphide of man-
ganese formed partly separates from the liquid pig-iron and
partly enters the slag together with other metallic sulphides
(Scotch slags of the coal measure iron-stones contain as
much as 8 per cent of sulphide of manganese). Part of the
sulphide of manganese also enters the ferriferous bears,
KARST., Archiv., 1 R., xvi., 180.
SULPHUR IN PIG-IRON.
303
sometimes in a crystallised state. These fluxes at the same
time facilitate the formation of white iron.
*
y. Blast mixed with Superheated Steam.-Though this
remedy has proved effectual in some cases, it has, on the
other hand, the disadvantage of cooling the furnace.
According to Bunsen,† steam conducted over glowing coals.
forms a mixture of gases of constant combination, namely-
56.52 per cent
28.71
H
CO
CO₂ = 14'77
4 vols.
2 "
I
Langloist found such an admixture of gases to be com-
posed of—
H
CO.
CO₂.
CH.
54'52 per cent.
31.86
12'00
1.62
Tilghman || recommends the introduction of common salt
along with the blast in order to form volatile chloride of
sulphur, but this reagent is too volatile and too expensive.
d. Electrolysis has been proposed by Winkler.§ Upon
conducting the electric current into the liquid iron in the
hearth of a blast furnace, iron and manganese will form the
positive, and pure carbon the negative electrode, and sulphur,
phosphorus, and silicon will be separated. The practical
application of this method is involved in many difficulties,
and consequently it does not give the expected results.
b'. A certain amount of sulphur in the fuel (usually iron.
pyrites in mineral coal) must be removed by washing the
coal and by carefully coking it, whilst employing several
reagents to volatilise the sulphur. The different methods.
will be explained in the chapter on fuel.
If it is necessary to employ coal containing sulphur, a
greater addition of lime must be given to the smelting pro-
cess and a higher temperature must be applied. Charcoal
*
Polyt. Centr., 1857, p. 513. B. u. h. Ztg., 1860, p. 130.
+ SCHEERER, Met. 1, 524. Bgwkfd., vi., 142; vii., 239.
+
Polyt. Centr., 1855, p. 1185; 1858, p. 119.
|| Bgwkfd., xx., 45. LEOB. Jahrb., 1861, x., p. 328.
§ DINGL., Bd. 161, p. 303. B. u. h. Ztg., 1862, p. 240.
304
IRON.
pig-iron is always purer than coke or coal pig-iron when
produced from pure iron ores; white pig-iron produced with
raw pyritic coal is usually the richest in sulphur.
b. Ores containing sulphur in the form of sulphates (heavy
spar, gypsum) may now and then be dressed chiefly by hand.
sorting; still the effect is doubtful, particularly if the asso-
ciates are intimately mixed with the ore. It is best to smelt
these ores, with fluxes of lime or manganese, at a higher
temperature.
Basic sulphate of alumina, which is sometimes formed in
iron ores (some sorts of magnetic iron ore) and in the coal
ash, is only completely decomposed by a high temperature,
and the greater part of the sulphur liberated from the sul-
phuric acid seems to enter the iron. Basic sulphate of iron
is more readily decomposed at a high temperature.
The properties of pig-iron containing sulphur are the
following:-
A small amount of sulphur (0°4 per cent) renders grey iron
thinly liquid, light grey or mottled, and more compact, sul-
phurous iron ores are sometimes therefore intentionally em-
ployed when aiming at the production of solid castings; on the
other hand, the presence of phosphorus is injurious. The
mottled state of this cast-iron could also be induced by an
addition of manganiferous flux or by employing a lower tempe-
rature at the smelting; but in the first case a hard, brittle iron
would result, and in the latter case a thickly liquid and porous
iron. An increase of the sulphur in pig-iron prejudices the
carbonisation and gives rise to the formation of a thickly liquid
white iron which easily solidifies and has cavities on its surface
and in its body. These cavities or blisters are probably
formed by the reaction of the moisture upon the sulphide of
iron, and these and its property of expanding when chilled
render it unfit for foundry purposes. When tapped off this
white iron flows in a sluggish stream, showing a yellowish
colour; it throws out brilliant sparks, and at the moment of
solidifying it frequently separates into small globules of the
size of a pea, at the same time evolving a gas. These
globules rotate on the iron when still liquid, and are very
SULPHUR IN PIG-IRON.
305
hard after cooling, white, and rich in sulphur. Light grey
grained iron sometimes shows the same behaviour.
The amount of sulphur in grey foundry pig seldom exceeds
0*5 per cent, and is usually somewhat higher when the iron
is reduced with coke than it is in charcoal iron. Schafhäutl
found in English and French varieties of good foundry pig from
0'177 to 1'105 per cent of sulphur. According to Eggertz,*
the Swedish cannon iron contains from 0'07 to o‘I per cent
of sulphur, and sometimes more; the Swedish forge pig
contains from o'02 to o'04 per cent. Pig-iron with an
amount of sulphur above o'4 per cent causes the production
of red-short wrought-iron. All better sorts of forge iron
from Styria mostly contain only o'02 per cent, and o°04 per
cent at the highest.
As the smallest amounts of sulphur render wrought-iron
and steel red-short, both good wrought-iron and steel must
be produced from pig-iron as free from sulphur as possible.
White iron rich in sulphur cannot be converted into a fair
wrought-iron, as its conversion takes a shorter time than is
required for the removal of the sulphur; on the other hand,
grey iron with a smaller amount of sulphur may be applied,
although it never yields so good a wrought-iron as pig-iron
free from sulphur. Pig-iron rich in silicon at its conversion
into wrought-iron allows the removal of more sulphur
than iron poor in silicon. According to Karsten, even
1-10,000th part of sulphur renders wrought-iron somewhat
red-short, but several thousandths are required to impair
its applicability. Wrought-iron containing o‘02 per cent
of sulphur is no longer fit for punching. The superior
sorts of steel may contain as much as o'oi per cent of
sulphur.
Whilst a certain amount of phosphorus renders wrought-
iron cold-short, injuring the quality of wrought-iron in a
similar way as sulphur, it has been suggested that sulphur
and phosphorus present in wrought-iron at the same time
might neutralise each other. Janoyert believes that the
VOL. II.
* B. u. h. Ztg., 1862, p. 96.
+ Bgwkfd., xviii., 501.
X
B. u. h. Ztg., 1855, p. 194.
306
IRON.
phosphorus removes part of the sulphur, forming sulphide of
phosphorus, and Jullien* supposes that the phosphorus tends
to reconstitute the cohesion of the iron, which at a higher tem-
perature was impaired by the presence of sulphur; but such
iron is never as strong as iron free from sulphur and phos-
phorus, and the wrought-iron produced from it shows cold-
and red-shortness in a less degree, for the reason that upon
mixing pig-iron containing sulphur with pig-iron containing
phosphorus, the injurious substances are distributed through-
out a larger mass.
Eggertz has suggested a method of judging the applica-
bility of a pig-iron to the manufacture of wrought-iron and steel
with regard to its contents of sulphur. He judges from the
shade which sulphuretted hydrogen gas produced from the pig-
iron in question imparts to a silver plate. This method might
also be used to ascertain the amount of sulphur contained
in iron ores and fluxes; but a correct determination of sulphur
can only be obtained by quantitative chemical analysis.‡
3. Phosphorus.-Pig-iron usually contains more or less
phosphorus, as the iron ores, fluxes, or fuel are frequently
contaminated with phosphates, and as the phosphoric acid
upon the reduction and carbonisation of the iron is reduced
by carbon, as well as by metallic iron, at a higher temperature
than sulphuric acid, forming phosphide of iron. Owing to
its easy fusibility, iron containing a larger amount of phos-
phorus cakes together, and thus renders its carbonisation
difficult, and it enters the furnace hearth as white pig-iron
containing phosphorus and poor in carbon if the temperature
of the furnace is low. If the temperature is very high at the
smelting zone of the furnace, the phosphatic iron-like sili-
ceous iron, but not like sulphurous iron-still combines with
carbon, forming a phosphatic light grey iron; as one part of
the phosphorus, like silicon, to a certain extent replaces the
chemically combined carbon, so facilitating the separation
* B. u. h. Ztg., 1859, p. 49; 1862, p. 97.
TUNNER, das Eisenhüttenwesen Schwedens, 1858, p. 31. LEOB., Jahrb.,
1863, xii., 150.
LEOB., Jahrb., 1861, x., 496. B. u. h. Ztg., 1862, p. 89; 1863, p. 121.
DINGL., Bd. 166, p. 279.
PHOSPHORUS IN PIG-IRON.
307
of graphite.
An amount of phosphorus much exceeding
15 per cent almost prevents the iron from combining with
carbon; upon solidifying no graphite will be separated, and
the resulting white pig-iron is rich in phosphorus and silicon.
In some instances silicon and phosphorus do not prevent
the carbonisation of iron; a white iron produced at Meppen
from bog ores contains-
C
Si
6'00 per cent.
P
As .
S
2.79
1.82
0'15
0'50
Struve supposes the phosphorus to be contained in pig-
iron in different states of combination, as upon dissolving
the iron in acids sometimes phosphoric acid only is formed,
and sometimes phosphuretted hydrogen at the same time;
and also as the residue is of variable nature.
The amount of phosphorus in pig-iron depends on whether
it is contained in the ores, in the flux, or in the ash of the
fuel. According to Caron, the ash of the fuel contains from
o'008 to o'i per cent of phosphoric acid; ores and lime used
as fluxes sometimes contain petrifactions of phosphate of
lime. According to Karsten's investigations all the phos-
phoric acid combined with oxide of iron becomes reduced
and enters the iron, whilst phosphoric acid present in other
combinations partly enters the slag, and afterwards is con-
ducted more or less into the iron according to circumstances.
More phosphorus contained in the slag will enter the
iron the longer the liquid slag is kept in contact with the
pig-iron, the higher the temperature, and the more siliceous
the slag happens to be. According to Wrightsont and Abel,‡
hot blast conducts more phosphorus into the iron than cold
blast, and Nicholson and Price state that such an influence
does not exist when producing grey iron, as all the phosphorus
of the ore would then enter the pig-iron. Von der Mark§
ERDM., J. f. pr. Ch., Bd. 79, p. 321. Polyt. Centr., 1860, p. 890.
+ DINGL., Bd. 114, p. 319; Bd. 116, p. 207. B. u. h. Ztg., 1850.
+
WAGNER'S Jahresber, 1860, p. 37.
B. u. h. Ztg., 1863, p. 296.
|| DINGL., Bd. 138, p. 124. B. u. h. Ztg., 1856, p. 73.
§ HARTMANN, Fortschr., V., 126.
X 2
308
IRON.
found that some phosphorus also volatilises, and is contained
in the deposits of the furnace mouth. Pig-iron produced
from bog ores sometimes contains above 6 per cent of phos-
phorus.
If a mechanical dressing does not avail to separate the
phosphatic substances, the amount of phosphorus must be
lessened by forming basic slags rich in lime, by employing
fuel poor in phosphorus, and by producing grey iron at as
low a temperature as possible. Pig-iron produced by these
means may be fit for foundry purposes, and sometimes for
forge purposes, but it will be totally unfit if white iron results,
owing either to a too considerable amount of phosphorus in
the mixture, or to too low a temperature. At Kertch* in
the Crimea this white and very brittle iron resulted from
brown iron ore associated with petrifactions and vivianite
when employing lime as a flux and anthracite as fuel; it was
composed as follows:-
•
Fe.
P.
Si.
94'49
2.65
I'09
I'90
C.
Some sorts of grey pig-iron produced from bog ores showed
the following composition :-
Fe.
C, combined.
Graphite.
P.
Si .
S
Mn.
I.
II.
III.
91'54
89*13
94.72
0*56
2.82
3'02
2°51
3.26
0'41
0.82
2°13
4'64
I'44
trace
3'00
trace
No. I is grey iron from Nalibok, produced from bog ores con-
taining 62.91 per cent of protoxide of iron and 3°19 per cent
of phosphoric acid, with an addition of lime and charcoal.
Upon puddling this pig-iron with an addition of lime, and at
as low a temperature as possible at the commencement, but
at a higher temperature at the end of the operation, wrought-
iron of the following composition was produced :—
*
ERDM., J. f. pr. Ch., Bd. 169, p. 321. Polyt. Centr., 1860, p. 891.
PHOSPHORUS IN PIG-IRON.
309
Fe.
P.
Si .
C.
98.55
0.68
0'13
0.64
Upon piling and welding this wrought iron, the following
wrought-iron was obtained :-
Fe.
P.
C .
99*27
0'55
0.06
No. 2 is grey fine grained pig-iron produced from bog ores
in Finland, containing from 40 to 50 per cent of iron when
employing an addition of lime, charcoal, and wood as fuel,
and hot blast of 180° C., analysed by Arppe. No. 3 is grey
pig-iron from St. Annä in the district of Snojärvi.
Different sorts of grey iron produced from bog ores in
Meppen, containing from 2 to 3 per cent of phosphoric acid,
showed as much as 18117 per cent of phosphorus; we will
give an analysis of this iron later on.
English oolitic iron ores with 15 or 2 per cent of phos-
phorus yielded grey iron containing from 0.71 to 1 per cent.
The following are the properties of phosphatic pig-iron:—
Phosphorus renders grey pig-iron harder and less strong than
silicon, but only if the amount of phosphorus exceeds o˚5 per
cent; such an iron containing a larger amount of phosphorus
is more thinly liquid than iron free from phosphorus; it is
well adapted to castings, giving very sharp outlines not
requiring a further mechanical treatment, which would be
rendered difficult by the greater hardness of the iron. The
different sorts of grey iron contain from traces up to 1'5 per
cent of phosphorus and more; if containing 5.6 per cent
the iron is still fit for castings; it is light in colour and has a
fine grain and a compact texture.
Grey iron poor in phosphorus is more or less fit for conversion
into wrought-iron and steel, and very slight differences in the
amount of phosphorus seem essentially to influence the
quality of the iron, and particularly of the steel. A certain
amount of manganese in the pig-iron seems to assist the
removal of phosphorus, as, according to Struve*, at the
* B. u. h. Ztg., 1860, p. 448.
310
IRON.
puddling process phosphate of manganese is formed, which
enters the slag, and is very difficult to reduce. Caron* did
not find that manganese facilitated the removal of phos-
phorus. According to Karsten, o'3 per cent of phosphorus
does not impair the quality of wrought-iron, iron containing
o'5 per cent stands the blowing test (schlagprobe), and iron
containing from o'75 to o'80 per cent shows cold-shortness.
Wrought-iron containing 1 per cent of phosphorus can be
applied to but few purposes.
White pig-iron rich in phosphorus always yields inferior
wrought-iron for reasons before mentioned, and is never,
therefore, intentionally produced. Good white forge-iron
produced from purer ores usually contains but little phos-
phorus, and spiegeleisen scarcely contains even traces of it;
such traces are even found in iron produced from ores per-
fectly free from phosphorus ; they are introduced by the ash
of the fuel. Most varieties of steel of high repute contain
from o‘o1 to o‘02 per cent, and though showing some cold-
shortness, wrought-iron containing from 0'25 to 0'30 per
cent of phosphorus is still fit for various fine forge purposes,
such as the manufacture of nails, wire, &c.
Eggertz and Nicklès‡ have suggested an easy and accurate
method of determining phosphorus in iron by means of
molybdate of ammonia.
4. Nitrogen. This substance has lately been found in
iron by different investigators, and it also seems to combine
with other substances contained in iron, such as silicon,
titanium, carbon, &c., which appear to influence the carbides
of iron chiefly at the formation of steel; but there are
various theories touching this point, and fresh investigations
are required to settle it. Whilst Frémy and Boussingault
consider nitrogen as an essential component of pig-iron,
Caron and Gruner believe it to be an accidental component,
as, when undergoing analysis, iron in a pulverised state
combines easily with ammonia.
DINGL., Bd. 168, p. 380.
B. u. h. Ztg., 1860, p. 415.
Ibid., 1863, p. 124. DINGL., Bd. 166, p. 279.
|| Erdm. J. f. pr. Ch., Bd. 84, pp. 82-100.
DINGL., Bd. 158, p. 209.
WAGNER'S Jahresber. über d. Fortschr. d. chem. Technol., 1861, p, 60.
MANGANESE IN PIG-IRON.
311
Buff* found o*26 per cent of nitrogen in pig-iron produced
from bog ores. Schafhäutlt ascertained that nitrogen is an
essential component of all sorts of English white iron; he
also found in grey cast-iron from Creuzot, France, 07202 per
cent; in English cast-steel, o'1831 per cent; and in spiegel-
eisen 1°20 per cent, but he only discovered mere traces in
some varieties of German iron. Rammelsberg‡ found lately
or.ly o'002 per cent of nitrogen in spiegeleisen, which, though
again proving that nitrogen combines with iron, leaves it
doubtful whether its presence is essential to the physical
properties of the iron.||
5. Manganese.-As before mentioned, manganese con-
tained in the iron ores or fluxes is reduced with much more
difficulty than protoxide of iron; it produces slags easier to
fuse, and consequently white pig-iron which may become
spiegeleisen if the iron ores employed are pure; moreover,
it purifies the iron by forming chemical combinations with
silicon, sulphur, &c., which separate on the surface of the
liquid iron. Spiegeleisen then contains a proportionally
small part of the manganese (usually only 7 per cent) reduced
in the furnace, which, however, under certain circumstances
(page 274) may increase to 22 per cent and more. The
favourable influence of the manganese contained in white
iron upon the removal of silicon in the puddling process
has been stated on page 299, and we have spoken of its
influence upon the removal of sulphur. We also suggested
on page 309, the influence which manganese may perhaps
have upon the removal of phosphorus, and we now proceed
to give an extract from the interesting report on the subject in
Dr. Ure's "Dictionary of Arts, &c.," vol ii., p. 726-
The presence of manganese in pig-iron does not appear
to exert much influence, either for good or for bad, on the
quality of the metal, and even when it exists in quantity,
amounting to 4 or 5 per cent in the crude iron, it disappears
almost entirely during the conversion of the cast-iron into
* Ann. d. Chem. u. Pharm., Neue Reihe, vii., 375.
† ERDM., J. f. pr. Ch., Bd. 67, p. 257. DINGL., Bd. 153, p. 349.
+
DINGL., Bd. 168, p. 127.
|| BAKER and G. STUART dispute the statement of nitrogen being contained
in steel.
Chem. Soc. Journ., 1864, p. 395.
312
IRON.
malleable or wrought. The cinder from iron smelted from
manganiferous ores contains, generally speaking, more
sulphur than slags or cinders from iron ores containing no
manganese. We have had numerous opportunities of con-
firming this, and have, therefore, on this account alone,
attached more importance to the existence of manganese in
iron ores; but our attention has more recently been directed
to another point which we think especially worthy of the
notice of iron manufacturers, namely, to the almost perfect
removal of phosphorus from pig-iron containing a very large
proportion of that element, and, at the same time, a high per-
centage of manganese. As our experiments on this import-
ant point are still in progress we shall merely here quote a
few in illustration of the purifying action we have alluded to.
"] Iron made from a highly phosphorised ore containing no
manganese-
Pig.
Puddled bar
Rough down bar
Phosphorus
per cent.
3'030
0838
0'572
"The finished bar was cold-short in the highest degree; it
was, in fact, nearly worthless.
“Iron made from highly phosphorised iron containing a
large per centage of manganese-
Pig.
Puddled bar .
""
Finished bar.
Phosphorus.
Manganese.
2.60
7.20
0°301
0.20)
0°30
ΟΙΙ
"The iron was carefully watched during the puddling pro-
cess. It melted very thin, and took rather more work than
usual; as soon as the boiling commenced it was very violent,
the metal forcing itself out of the door hole until it was
checked. When it "came to nature," as the workmen term
it, it worked beautifully, and stood any amount of heat; in
fact, the heat could with difficulty be raised to the requisite
degree. The yield was 22 cwts. 2 qrs. 24 lbs. of pig, to pro-
duce 1 ton (20 cwts.) of puddled bars; this is about the
yield of good mine iron when properly puddled. The
finished bar exhibited none of the cold-short quality; it was
MANGANESE IN PIG-IRON.
313
exceedingly ductile; indeed, excellent horseshoes were made
from it. The puddling cinder had the following composi-
tion :
Silica.
8.240
Protoxide of iron.
70*480
Oxide of manganese .
Phosphoric acid
12.800
7.660
Sulphur .
O'535
99*715
"Other observations have shown that highly manganiferous
pig (without phosphorus) is puddled with difficulty and some-
times with considerable waste, so that the advantages of an
alloy of manganese would seem to be confined to those
varieties of crude iron into the composition of which phos-
phorus largely enters."
As grey iron results from ore mixtures which are more
difficult to fuse than similar mixtures for white iron, the
former mixture must not contain any perceptible amount of
manganese, or it must be melted with an admixture of lime
containing magnesia if the production of grey iron is intended.
A certain amount of manganese renders such grey iron harder
(probably owing to the formation of silicide of manganese),
mottled, and less fit for some sorts of casting than pig-iron
containing sulphur. When converting this pig-iron into
malleable iron, its amount of manganese will exert the same
favourable influence which it has on white iron. Pig-iron
produced from Franklinite* (FeO + MnO)+(Fe2O3 + Mn₂O3),
(page 270), containing from 3 to 11.5 per cent of manganese,
has nearly the properties of spiegeleisen. Such pig-iron is
exceedingly hard, has large crystalline plates in the middle,
and is grained at the edges; its fracture is silver white and
of metallic lustre, and it contains on an average one-eighth of
the manganese which was contained in the ores, the other
seven-eighths having been scorified. The wrought-iron pro-
duced from these pigs is exceedingly strong.
According to Karsten, 1'85 per cent is the highest amount
* B. u. h. Ztg., 1860, pp. 37, 465. TUNNER, Ber. über. d. London, Auss-
tellung, 1862, p. 42.
314
IRON.
of manganese which wrought iron may contain without
deterioration in quality.
6. Copper. An admixture of copper pyrites with iron ore
introduces copper into the pig-iron. The amount of copper
is usually under o'2 per cent, which does not prejudice the
quality of foundry pig (1 per cent of copper renders pig-iron
harder and more difficult to dissolve in acids), but the pig-
iron does not lose its copper upon conversion into malleable
iron, as it probably remains in combination with sulphur and
is concentrated. According to Eggertz,* some tenths per
cent render iron red-short, and steel is more sensitive to
the effect of copper, therefore the Swedish iron ores being
free from copper pyrites are particularly fit for the production
of pig-iron for steel manufacture.t Wrought-iron containing
0'5 per cent of copper shows traces of red-shortness, whilst
steel produced from such iron was useless. According to
Stengel an amount of o˚44 per cent of copper renders wrought-
iron red-short. Jullient states that o˚4 per cent of sulphur
and 0°4 per cent of copper cause the formation of cracks in
steel and bar-iron, and also injure the property of welding of
pig-iron when puddled. Schafhäutl! is of opinion that the
noxious influence at the puddling process which has been
attributed to copper does not exist; moreover, that it is chiefly
the state of aggregation and also the metalloids mixed with
the carbide of iron which exert the greatest influence.
7. Arsenic.—The investigations of Walchner,§ Stein,¶
Schafhäutl,** and others have shown that this element is
frequently present in iron ores, and sometimes in the fuel,
and, like phosphorus, it renders pig-iron hard, brittle, and
of great fusibility, and wrought-iron cold-short and difficult
to weld.tt Schafhäutl states that a certain small amount of
arsenic-perhaps present in the form of arseniuretted carbon-
* B. u. h. Ztg., 1862, p. 218.
† TUNNER, das Eisenhüttenwesen Schwedens, 1858, p. 12.
JULLIEN'S Eisenhüttenkunde, 1861.
B. u. h. Ztg., 1861, p. 39.
DinGL., Bd. 103, p. 227.
¶ ERDM., J. f. pr. Ch., li., 302; liii., 37.
Ibid., xxi., 247. B. u. h. Ztg., 1847, p. 364.
tt B. u. h. Ztg., 1862, p. 408.
ZINC IN PIG-IRON.
315
renders steel fine-grained and of greater strength; a greater
amount of arsenic causes red-shortness.
As arsenic occurs in the iron ore sometimes as arsenical
pyrites, and now and then in combination with iron pyrites,
it is advisable for the removal of the arsenic to submit the
ores to a good, partly reducing, roasting; an addition of lime
is sometimes advantageous. The decomposition is more
difficult if the iron ores (of younger formations) contain
arseniates, chiefly arseniate of iron.
Berthier, Wöhler, Schafhäutl, Mrazek, and others found*
in pig-iron from Alais 4'08 per cent of arsenic, in wrought-
iron from Dannemora o'017 per cent, and in cold-short iron
from Hungary, which was difficult to weld, 0°375 per cent of
arsenic and o‘29 per cent of phosphorus.
Antimony and bismuth behave like arsenic; an amount of
03 per cent of antimony renders wrought-iron cold-short.
Antimony, arsenic, and bismuth volatilise at the welding
heat of wrought-iron.
8. Tin.-Pig-iron containing tin has a fine-grained steel-
like texture, a clear sound, and great hardness without being
brittle in excess, and it is easy to fuse. A cold-short wrought-
iron containing about o'19 per cent of tin is produced from
pig-iron containing 1 per cent of tin; upon heating it is easy
to forge, and emits a white vapour. Iron containing a larger
amount of tin is difficult to weld during the puddling pro-
cess; it causes a great loss of iron, and yields wrought-iron
so cold-short that it breaks like glass merely on falling; this
wrought-iron also welds badly; its fracture is fine-grained,
white, and dull.
9. Zinc.-Calaminet or zinc blende in the iron ores intro-
duces but little zinc into the iron. Most of the zinc vola-
tilises and forms zinciferous deposits in the furnace, and is
also found in the flame escaping from the furnace mouth and
the tymp, characteristically colouring it. A considerable
volatilisation of zinc may carry off so much heat as to cause
PRECHTL, Techn. Encykl., xv., 367, 377; Bd. 15, p. 376. Ann. d. Chem. u.
Pharm., Bd. 31, p. 95.
† B. u. h. Ztg., 1861, p. 355.
316
IRON.
the formation of white iron; zinc also requires a considerable
quantity of fuel for its reduction and volatilisation. The
experiments for the production of both pig-iron and zinc
from zinciferous iron ores have not been successful (vol. i.,
page 459). Pig-iron produced from Franklinite (page 313)
is well adapted for the manufacture of wrought-iron, owing
to the presence of manganese.
Whilst pig-iron produced from ores associated with cala-
mine does not usually contain zinc, a whitish grey iron of little
strength is formed when removing the zinciferous deposits
which then fall partly into the hearth. When tapped off,
this iron emits smoke and a bluish green flame, and yields a
fairly malleable iron, though with a greater loss of iron, ast
it requires a more frequent heating.
10. Lead.-Galena associated with iron ores, which is
frequently the case, has no influence on the pig-iron; part
of it volatilises and part is found in a metallic state beneath
the sole stone of the blast furnaces, its contents of silver being
concentrated in it.
II. Vanadium, Molybdenum, Chromium, Tungsten,
and Titanium are found only in small quantities in pig-iron,
and seem to influence its quality but slightly. When
smelting titaniferous iron ores (titaniferous iron sand of New
Zealand),* or upon adding titanitet to the iron ore mixture,
grey titaniferous pig-iron is produced (not white iron, as at
this production most of the titanium enters the slag), which
is said to yield an excellent wrought-iron and steel; but it is
still doubtful whether any portion of the titanium is retained
in the wrought-iron or steel made from such pig-iron, so that
the improvement attributed to the use of titaniferous ore is
probably due to some indirect action rather than to the
actual presence of titanium in the finished product. Mr.
Robert Mushet has taken out no fewer than thirteen patents
for alleged improvements in the manufacture of iron and
steel, and titanium plays the chief part in all.
Berggeist, 1861, No. 54. Polyt. Centr., 1854, No. 3.
pp. 118, 412.
B. u. h. Ztg., 1862,
↑ Berggeist, 1862, No. 75. B. u. h. Ztg., 1860, p. 143; 1863, pp. 16, 51.
TITANIUM IN PIG-IRON.
317
A larger amount of titanium renders the ore mixture more
difficult to fuse, owing to the impeded carbonisation of iron,
as the titanium combined with iron unites with carbon and
nitrogen and separates in the form of copper-red cubical
crystals, which consist of cyano-nitride of titanium. An
addition of fluxes containing potash renders a titaniferous
mixture easier to fuse; a certain amount of lime and silica
serves the same purpose, forming a combination of silicate
and titanate of lime which is easier to fuse.
About I per cent of titanium may be present in pig-iron;
it increases the strength of the metal, at the same time giving
it a peculiar mottled character, the fractured surface showing
a series of dull dark grey patches apparently set in white
network.
Mayrhofer supposes that the good effect of the above.
named and also of some other substances is less the result
of the combination of those substances with the iron, than of
their capacity for combining with noxious associates of the
iron, such as sulphur, phosphorus, silicon, &c., and sepa-
rating them in the same way as by manganese. Schafhäutl+
and others suggest that steel takes into combination small
quantities of chromium, nickel, silver, platinum, iridium,
rhodium, titanium, aluminium, &c., thus becoming harder
and more tenacious; tungsten has chiefly the property of
rendering cast-steel very hard and tenacious. According to
Bernoulli, when an intimate mixture of finely divided grey
cast-iron and tungstic acid is heated to a very high tempe-
rature, the graphitic carbon is burned by the oxygen of the
tungstic acid, and steel is formed, which alloys with the
reduced tungsten. No diminution in the amount of carbon
was, however, perceived when the experiment was repeated
with spiegeleisen or ordinary white iron, carbon in the com-
bined form being apparently unable to effect the reduction of
tungsten. Siewert examined six samples of so-called tungsten
steel; four of them contained from 1 to 3 per cent of tungsten,
whilst none was found in the other two.
+
Oesterr. Ztschr., 1861, p. 37+.
PRECHTL, Techn.. Encykl. xv., 367.
BAUERMAN's Treatise on Metallurgy, p. 47.
318
IRON.
According to Guen,* cast-iron containing tungsten is
stronger than iron without tungsten.
Chevreult suggests that the above-named substances may
totally replace the carbon.
Steel containing 12 per cent of chromium gives a beau-
tiful damask surface when etched by sulphuric acid.
12. Nickel and Cobalt.-These metals seldom occur in
iron. Rubacht found in one variety of wrought-iron 153
per cent of nickel and o'63 per cent of cobalt, and Mrazek ||
0°021 and 0'017 per cent respectively. Streng found o˚003
per cent of nickel in pig-iron from Königshütte (Hartz).
Nickel alloys readily with iron without affecting its
malleability; this is exemplified in meteoric iron, which
contains a considerable quantity of the nickel, and has
occasionally been used by savages for the manufacture of
knives. A horse-shoe made from the great Australian
meteorite was exhibited at South Kensington in 1862, and is
now in the Museum of Practical Geology. The mass from
which it was taken, weighing 3 tons, has been presented to
the British Museum.
Cobalt is said to increase the whiteness and brilliancy of
iron, but has no marked effect upon its physical properties.
13. Aluminium, Calcium, and Magnesium are reduced
in small quantities by the influence of the temperature of
iron blast furnaces, if the slag present is basic; they
then enter the pig-iron. Larger amounts of these metals.
injure the quality of the pig-iron as well as that of the
wrought-iron produced from the pig-iron ;¶ on the other
hand, they are said to improve the quality of steel** (Indian
wootz steel). Whilst Faraday and Stoddart found from
0'013 to 0.69 of aluminum in this steel other investigators
did not detect traces of that metal in Indian wootz steel.
Oesterr., Ztschr., 1863, p. 286.
↑ B. u. h. Ztg., 1862, p. 108.
+ DINGL., Bd. 117, p. 395.
+
B. u. h. Ztg., 1862, p. 408.
BAUERMAN'S Treatise on the Metallurgy of Iron, 1868, p. 46.
¶KARST. Archiv., 1 R., ix., 417.
** Ibid., 1 R., viii., 193. DINGL., Bd. 134, p. 157.
ALKALI METALS IN PIG-IRON.
319
According to Gruner and Lau,* a larger addition of lime to
the ore mixture impedes the reduction of silicon, but it in-
creases, at the same time, the reduction of aluminium,
magnesium, and calcium, which injure the tenacity of pig-
iron to a greater degree than is generally supposed. Some
sorts of English pig-iron are found to contain from o'5 to I
per cent of aluminium, the Swedish varieties as much as
15 per cent of calcium and magnesium, and o'75 per cent
of aluminium. Pig-iron from Königshütte (Hartz) contains
0*258 per cent of calcium, and the spiegeleisen from Müsen
all the above-named metals.
Lohaget states that an addition of alumina at the manu-
facture of cast-steel has a great influence on the grain and
lustre of the steel, silicates of manganese and aluminium
being formed at the same time and separating on the surface
of the liquid steel.
+
According to Dumas, the apparently pure commercial
aluminium contains o˚47 to 0.7 per cent of silicon, and from
3*37 to 6.8 per cent of iron; this shows the affinity between
those substances.
Schafhäutl, upon dissolving grey iron, found, together
with the graphite, bright white scales containing aluminium,
whilst the iron did not contain more than 1'01 per cent of
aluminium at the highest. He considers silicide of iron and
aluminide of iron as characteristic components of grey iron,
and carbides of silicon, copper, and arsenic as components
of steel.
14. Alkali Metals, sometimes contained in the ash of the
fuel, or in the flux, and also in some iron ores (carbonaceous
ores), are reduced and alloy with the pig-iron. According
to Tissier,¶ they have a favourable influence during the con-
version of the pig-iron into wrought-iron, as they combine
directly with sulphur, phosphorus, and arsenic, thus separating
B. u. h. Ztg., 1862, p. 254.
† Ibid., 1861, p. 160.
+
WAGNER'S Jahrsber., 1859, p. 3.
|| ERDM., J. f. pr. Ch., Bd. 67, p. 257.
FRESENIUS, Ztschr. f. analyt. Chem., 2 Jahrg., p. 39.
¶ DINGL., Bd. 160, p. 122. Bgwkfd.. xx., 740.
320
IRON.
these substances, and as they oxidise silicon and carbon if
present in the oxidised state. The better quality of malleable
iron produced from charcoal, in comparison to that made
from coke pig-iron, is said to be partly caused by the presence
of an alkali metal.
At Witkowitz* trials were made to add some chloride of
sodium to the iron, but without success, owing to the vola-
tility of the salt. Tapping the pig-iron into moulds coated
with common salt, soda, or potash, was also unsuccessful.
Fresenius has ascertained the amount of alkalies contained
in spiegeleisen from Siegen (page 270).
Chemical Constitution of Pig-Iron.
Analyses of white and grey pig-iron have proved that the
amount of carbon may be the same in both kinds, the physical
properties of pig-iron do not therefore depend on the quantity
of carbon present but on the different allotropic state in which
carbon is contained in pig-iron. This allotropic state of
carbon is indicated by the behaviour of different sorts of grey
pig-iron to acids (page 292), and also by the reaction of
liquid tin and tungstic acid on pig-iron.
Eyferth observed that upon stirring the liquid grey pig-
iron with 25 per cent of tin, a great deal of graphite and
silica separated, forming a carboniferous alloy of tin and
iron, whilst white pig-iron, when treated in the same way,
gave up no graphite, though forming some little alloy.
Bernoulli states that white pig-iron does not reduce
tungstic acid, but when heated with grey iron a raw cast-
steel is formed, the graphite serving as an agent to reduce the
tungstic acid, and tungsten enters the steel.
* LEOB., Jahrb., 1861, x., 331.
KARST., Arch., 2 R., XXV., 235. ERDM., J. f. pr. Ch., Bd. 61, p. 30.
B. u. h. Ztg., 1854, p. 107, DINGL., Polyt., Journ., Bd. 145, p. 155; Bd. 141,
p. 432; Bd. 166, p. 279; Bd. 167, p. 291.
Ztschr., d. ver. Deutch. Ingen.,
114, p. 507; Bd. 124, p. 59.
Ztschr., 2 Jahrg., p. 39.
+
B. u. h. Ztg., 1861, p. 142.
DINGL., Bd. 159, p. 359.
1861, p. 224.
LEOB., Jahrb.,
Polyt. Centralb., 1858, p. 59.
Ann. d. Chem. u. Pharm., Bd.
1861, x., 496.
FRESENIUS,
CHEMICAL CONSTITUTION OF PIG-IRON.
321
The carbon in pig-iron may therefore be divided into
chemically combined carbon, and into mechanically com-
bined carbon or graphite. White pig-iron only contains the
former carbon, usually in quantities of from 3 to 5'93 per
cent (page 267), whilst grey pig-iron contains both modifi-
cations, altogether from 3'15 to 4'65 per cent, 2'57 to 3'75 of
which are graphite.
It was formerly supposed that the carbon in white iron was
combined with the whole mass of iron, but not in definite pro-
portion, and Karsten, Berthier, and others suggested that a
proportionally small part of the iron in grey pig was combined
as polycarbide,* which was then dissolved in the remainder
of the iron, supposed to be in a ductile state. Karstent has
lately modified this view, and Schafhäutl‡ has suggested that
polycarbide may be a graphite alloyed with other substances,
such as iron, sulphur, silicon, arsenic, &c.
Spiegeleisen may fairly be considered as tetracarbide,
Fe,C, which, according to Gurlt, changes upon heating into
octocarbide, FegC (?), and graphite, corresponding to the
composition of grey iron. Upon heating spiegeleisen to a
lower temperature less graphite is separated, and the
resulting mixtures of carbides of iron with a smaller
amount of graphite are represented by the mottled pig-iron.
Such mixtures, but without graphite, are also found in the
porous and flowery white pig (page 276) which is not satu-
rated with carbon owing to an insufficient temperature, and
owing to the iron passing too quickly through the smelting
and carbonising zone.
Besides the carbon, the other foreign substances in pig-
iron (sulphur, phosphorus, silicon, manganese, &c.,) were
formerly considered as accidental admixtures, but they are
now supposed to form distinct chemical combinations, re-
placing either the electro-positive iron (manganese, alkaline
or earthy metals); or the electro-negative carbon (silicon,
sulphur, phosphorus, &c.); more investigations, however,
* KARST., Arch., 1 R., viii., 3.
† Ibid., 2 R., xxi., 500.
+
VOL. II.
ERDM., J. f. pr. Ch., xix., 159, 404; xx., 465 ; xxi., 129.
Y
322
IRON.
are required to elucidate the properties of these combinations
and their influence upon the nature of the pig-iron.
Gurlt has calculated the following formulæ from a number
of the most trustworthy analyses of pig-iron :—
4(Fe, Mn)C, FegC, 4(FeMn)Si, 4(FeMn)P, 8(FeMn)S, &c.
But it appears also that these foreign admixtures combine
with each other and modify the properties of iron according
to their state of combination; therefore the differences.
between pig-iron and steel are not caused by the different
amount of carbon only; this also requires further investi-
gation.
*
Limit between Pig-iron, Wrought-iron, and Steel.-
With regard to their chemical composition no distinct limits.
exist between these varieties of iron, and the carbides only
differ in the mode of their production and in the peculiar
properties of each product.
Karsten terms carbide of iron, pig-iron which is not mal-
leable or ductile at the common temperature, and which
liberates graphite when slowly cooled; this takes place
when the iron is combined with at least from 2.25 to 2.3 per
cent of carbon, both chemically and as graphite; this amount
therefore constitutes the limit between pig-iron and steel.
But the separation of graphite may be prevented if sulphur,
phosphorus, and other substances which facilitate the for-
mation of white iron, are present, and then the separation
only takes place if the product is exposed, without being
melted, to a red heat for about 48 hours.
Steel is characterised by its fusibility, its property of
welding, and of hardening upon sudden cooling. It welds but
slightly if containing 1'75 per cent of carbon, is difficult
to forge if containing 1'9 per cent of carbon, and when ham-
mered breaks into pieces if containing 2 per cent, then
approaching in properties to pig-iron. Steel containing from
14 to 15 per cent of carbon seems to be strongest and
hardest, and though it increases somewhat in hardness if
the amount of carbon increases, on the other hand, its pro-
perty of welding and its tenacity then decrease. Steel retains
*DINGL., Polyt. Journ., Bd. 158, p. 209.
IRON ORES AND THEIR PROPERTIES.
323
•
its property of welding if its amount of carbon decreases,
but is then more difficult to fuse, and becomes less hard after
sudden cooling. It passes into malleable iron if containing
less than o'5 per cent of carbon. Wrought-iron is charac-
terised by a property of welding, by its infusibility and
its softness after sudden cooling. Iron containing from
0'5 to 0.65 per cent of carbon is very soft steel. Malleable
iron is softer the less carbon (about o'08 per cent) it contains.
As before mentioned, these limits may be modified by ad-
mixtures of foreign substances, frequently even by small
quantities of them.
DIVISION I.
PRODUCTION OF PIG-IRON FROM IRON ORES.
1. Smelting Materials.
The smelting materials comprise iron ores and different
ferruginous substances (refinery and puddling cinders, &c.)
which are either smelted by themselves or in admixture with
iron ores; also the fluxes and fuel.
A. Iron Ores and their Preparation.
Minerals are termed iron stones or iron ores when they
are ferruginous in so high a degree, and, at the same time,
so free from injurious admixtures, as to allow the profitable
production of a useful product. Whilst the noble metals
occur mostly near the equator, iron is found chiefly in the
North and in the temperate zones. Some iron ores improve
in quality as they occur nearer the surface of the earth.
Where ores and fuel do not occur together the question
will arise as to whether it is more advantageous to carry the
coals to the iron ore or the iron ore to the coals; this ques-
tion depends on local circumstances, chiefly on the price of
the freight, the value of labour, &c.
Y 2
324
IRON.
The ferruginous minerals may be chemically classified as
follows:-
I. Oxides of Iron.
a. Magnetic Iron Ores, (FeO, Fe2O3), containing 31 percent
of peroxide of iron, 69 per cent of protoxide, and 72'4 per cent
of metallic iron, are generally more difficult to smelt as they
are less inclined to reduction and carbonisation, owing to
their great compactness; furthermore, they cake easily upon
roasting, and also melt easily in the blast furnaces before
being reduced; this imperfect reduction causes the formation
of fusible ferruginous slags, and this formation is facilitated
by the protoxide of iron contained in the iron ore. The
smelting* of magnetic iron ores requires a careful and fre-
quently repeated roasting at a low temperature (Altenau in
the Hartz), a finer division, a considerable addition of lime,
and a prolongation of the upper part of the furnace hearth.
The higher furnace hearth keeps the iron in contact with
glowing coal for a longer time, thus carburising it more per-
fectly. The larger addition of lime prevents the formation
of siliceous slags, renders the mixture more difficult to fuse,
and consequently impedes the sinking of the charges, thus
causing a longer reaction of the reducing and carbonising
agents.
Owing to the difficult fusibility of the ore mixtures from
magnetic iron ores, a grey or mottled pig-iron is chiefly pro-
duced, which is well adapted for foundry purposes (Swedish
cannon iron) and also for conversion into malleable iron
(Dannemora iron) supposing that the ores are also pure.
Admixtures of foreign earthy and metallic substances in
magnetic ores either increase or lessen the difficulty of
smelting these ores. The most advantageous are those ores.
of uniformly grained texture, free from sulphur and phos-
phorus compounds (iron, magnetic, copper, and arsenical
pyrites, zinc blende, galena, apatite, &c.) when in admixture
with silicates, &c. (chlorite, hornblende, garnet, tourmaline,
quartz, manganiferous minerals, gneiss, calc spar, &c.), as
they contain the components required for a desirable slag,
* HARTMANN, Fortschr., iii., 147.
OXIDES OF IRON.
325
rare cases
besides some manganese; they melt at low temperature with
cold blast, combine with a less amount of earthy bases,
chiefly silicon, and yield an iron of a strongly mottled, fre-
quently radiated, and almost specular appearance, and in
cases a slightly mottled or grey iron, which is in
the highest degree adapted for conversion into malleable
iron (Swedish Dannemora iron*). Magnesian admixtures,
such as those occurring in iron mica slate, itabirite and
catawbarite in North Carolina,† render the ore difficult
to fuse.
Magnetic iron ore containing a prevailing amount of
quartz in coarse grains also renders the mixture difficult to
fuse, but the scorification of oxidised iron and its reaction
upon carbide of iron, which cannot be prevented at the
commencement of the operation, may be avoided later on by
smelting at a higher temperature. For instance, the Swedish
Ferola ores (magnetic iron ores with coarse grains of quartz,
besides some oligoclase, amphibole, and iron pyrites) yield at
Finspong a mottled iron‡ well fitted for the manufacture of
cannons, in which the silicium is limited by the presence of
manganese. A little sulphur retained in the ore, owing to
an imperfect roasting, renders the pig-iron more tenacious.
The smelting of magnetic iron ores is much more difficult,
as they contain finely disseminated quartz which cannot be
observed by the naked eye, and converts the ore into siliceous
magnetic iron ore (Altenau in the Hartz). This sort of
quartz has the greatest inclination to scorification at the
roasting as well as in the smelting; these ores are therefore
roasted twice at a low temperature and are added to other
ore mixtures in small proportions. The quality of the
resulting pig-iron is lower the larger the amount of these
sulphurous and phosphatic minerals contained in the ore.
These impurities are frequently so predominant as to render
the ore unfit for smelting (Schwarzenberg in Saxony;
Traversella in Piedmont, where an electro-magnetic dressing
of the ores has been introduced).
*PRECHTL, techn. Encykl.
Eisenhüttenwesen, 1858, p. 12.
† B. u. h. Ztg., 1857, p. 242;
‡ Ibid., 1857, pp. 262, 269.
xv., 378, 418. TUNNER, das Schwedische
B. u. h. Ztg., 1857, p. 361.
1860, p. 9.
326
IRON.
The magnetic ores are not usually manganiferous, but
when they are they yield, supposing them to be pure, an excel-
lent pig-iron, admirably fitted for conversion into malleable
iron.
Magnetic iron ore occurs particularly in igneous and
metamorphic rocks, either in distinct crystals or, as in many
basalts, disseminated through the mass, when it frequently
imparts magnetic properties to the rocks, especially to green-
stone, serpentine, or basalt. It also forms beds in gneiss,
in chlorite, mica, hornblende, and clay-slates, in marble,
greenstone, and other rocks, but seldom appears in veins.
The largest known formations occur in the northern parts of
the globe, in Scandinavia, Lapland, Siberia, and North
America. Less extensive deposits occur in the Hartz, in
Saxony, Bohemia, Silesia, and Styria; and in Southern
Europe, in Elba, and Spain. Magnetite is the most im-
portant iron ore of Norway, Sweden, and Russia. The
Dannemora mines in Sweden, excavated in an open quarry
150 feet broad and 500 feet deep, furnish the fine Ocregrund
iron, largely imported into England for the manufacture of
steel. Some highly magnetic varieties, especially from
Siberia and the Hartz, form natural magnets, possessing
distinct polarity. Others become polar only after contact
with magnets of sufficient power. Magnetic ores sometimes
occur in admixture with specular iron ore. They are also
found in some parts of England.
Two specimens of magnetic ore from Cornwall were of
the following composition (Dr. Noad)* :-
Water
2'50
3°20
Protoxide of iron
20'00
13'00
Peroxide of iron.
44'40
66'50
Oxide of manganese
0*16
0*56
Alumina.
5.20
3'60
Lime
0.60
0'56
Magnesia
I'00
I'52
Sulphuric acid
0'04
0'04
Phosphoric acid
0'50
0'57
Insoluble residue
24°20
9'40
* Dr. URE's Dictionary of Arts, &c., ii., p. 682.
FRANKLINITE.
327
Chrome Iron Ore* is sometimes met with in a similar state
of minute octohedral crystals, and may readily be mistaken
for magnetic iron sand, but being non-magnetic, may be
instantly distinguished from it.
The following iron ores are similar to the magnetic iron
ore, and are now and then used for the production of iron:
a. Titaniferous Iron Ore, a chemical combination of
mFe₂O3+nTi₂03, in different proportions, and associated with
small quantities of other substances, such as silica and oxides
of manganese, calcium and magnesium. At Avellino (Naples)
this sandy iron ore is worked in open hearths, and titaniferous
iron sand occurs in large masses on the coast of New Ply-
mouth (province Taranaki), in New Zealand. According to
Freytag, it contains-
FeO.
Fe₂O₁
Ti₂03
27°53 per cent.
66.12
6.17
""
and, according to other investigators, 88 45 per cent of
protoxide of iron and 11:43 per cent of protoxide of titanium,
besides silica and a trace of manganese, and yields 61 per
cent of iron of the best quality.
The titaniferous iron sand from Egersund, in Southern
Norway, contains as much as per cent of oxide of tin.
B. Franklinite, 3(FeO, ZnO, MnO) + (Fe₂O3, Mn2O3),
usually in admixture with red zinc ore and oxide of iron.
When composed of-
+
7-12ths 3FeO) (Fe₂O,
5-12ths 3ZnO)†13 Mn₂¤¸
it contains-
Fe
Mn
Zn
44'73 per cent.
9°33
21.07
The average of several analyses by Rammelsberg gives the
following composition :-
Iron.
Manganese.
Zinc.
Oxygen
45'16
9°38
20°30
25'16
100'00
PERCY, Metallurgy, p. 198.
328
IRON.
This mineral is only found in two or three places in New
Jersey* (North America) where it occurs in metamorphic
silurian limestone, as a bed from 20 to 30 feet thick, overlaid
by 6 or 8 feet of red zinc ore.
Both minerals are first treated for zinc, and the residues
are then smelted for white pig-iron or spiegeleisen.
b. Hæmatite, Fe₂O,, containing 69:34 per cent of iron.
It occurs in the crystalline form, constituting specular iron
ore, or micaceous iron ore, according to the dimensions and
character of the crystals; and it is massive or earthy, gener-
ally in reniform or botryoidal nodules of a more or less fibrous
and radiating structure.
a. Specular Iron Ore is compact, of metallic lustre, steel
grey, and gives a red streak. It is usually purer than mag-
netic iron ore, and yields a grey, soft, tenacious pig-iron,
owing to its great compactness, which makes it more difficult
to reduce, and to its want of manganese. This pig-iron
when produced with charcoal is especially adapted for the
manufacture of steel and wrought-iron (Isle of Elba). If
aiming at the production of white pig manganiferous fluxes
must be added (Fellonica in the Isle of Elba).
The specular ore occurs chiefly in the older crystalline
rocks in large beds or veins. The mines of the Isle of Elba,
celebrated from antiquity, still furnish the finest crystals,
which occur in the massive variety, along with pyrites and
quartz. This ore is also associated with some psilomelan;
it is rich and easy to reduce and to smelt; it is also found
elsewhere, chiefly in Sweden, Lapland, and the Ural.
It also occurs in volcanic rocks, as in Auvergne, on Vesu-
vius, Etna, and the Lipari Islands, especially Stromboli,
where some fine crystals, 3 inches broad and 4 inches long,
have been procured.
The micaceous variety occurs at Tincroft in Cornwall, at
Tavistock in Devonshire, and in Wales, Cumberland, and
Perthshire.
B. Red Hæmatite is less compact than the specular ore.
It varies in colour from steel-grey to brown-red, and always
B. u. h. Ztg., 1860, p. 463.
RED HEMATITE.
329
produces a characteristic red streak when drawn across a
piece of unglazed pottery. It occurs in fibrous, columnar,
botryoidal, granular, pisolitic, and compact forms. Special
names are given to the different varieties; for instance—
Kidney Ore, or Rother Glaskopf, includes the hard botry-
oidal forms, such as those of Cumberland, which are devoid
of metallic lustre ;
Red Ochre, or Iron Minium, are compact earthy varieties,
often containing clay, which are ground and used as colours.
Puddler's Ore is a peculiar, unctuous, compact form from
Cumberland, which is largely used for lining the hearths of
puddling furnaces.
The red hæmatite usually yields a soft tenacious grey iron,
fit for foundry and forge purposes owing to its compactness,
and to its seldom containing manganese, whilst the protoxide
of iron of the magnetic iron ore facilitates the formation of a
slag which combines with part of the protoxide, and has a
decarbonising reaction upon the pig-iron. In a few cases
manganiferous red hæmatite also occurs; as, for instance, at
Ulverstone, where red hæmatite containing 50 per cent of
oxide of manganese is smelted. The earthy associates are
usually quartz, calc spar, chlorite, siliceous iron (eisenkiesel),
chalcedony, hornblende, felspar, heavy spar, &c.; the metallic
associates are titaniferous minerals, iron pyrites, &c.
Red hæmatite as a rule contains more noxious foreign
associates than specular iron ore, and less than magnetic
iron ores; it is more or less difficult to fuse according to its
earthy associates.
Whilst the presence of quartz renders it more difficult to
fuse, silicates, fluor spar, calc spar, marl, and brown spar
increase its fusibility. Lime renders the ores less compact
upon roasting, and consequently facilitates their reduction in
the furnace, as is the case with magnetic iron ores.
Red hæmatite is the chief ore for the production of iron.t
It occurs in many parts of the Continent, and in great abun-
dance in the United States, in immense masses at Lake
Allgem., B. u. h. Ztg., 1863., p. 194.
This observation does not apply to England; for in this country by far
the largest quantity of iron is reduced from clay ironstone.
330
IRON.
7
Superior, and in the States of Missouri, Maine, New York,
Pennsylvania, Virginia, Arkansas, &c.
In this country* these ores are found in the greatest
abundance in the mountain limestone formations. The most
abundant deposits are those of Lancashire and Cumberland.
. The hæmatite of Whitehaven occurs in the carboniferous
limestones near the outcrop or surface edge of the clay-slate
rocks upon which the formation rests. Most of the excava-
tions from which it is extracted are subterraneous, and the
masses of iron ore in which the workings are carried on are
often so extensive that it is difficult in such situations to
obtain a clear idea of the nature of this important deposit.
(Warington Smyth). 214,433 tons of the hæmatite of the
Whitehaven district are smelted on the spot, at the Cleator
Moor and other iron works, and 579,642 tons are sent into
the iron working districts. In the year 1865, 607,439 tons
were raised in the Ulverstone district, and no less than
937,386 tons of hæmatite were exported during 1865 (Hunt) for
the supply of Staffordshire, South Wales, and other districts,
from these two localities. Considering its quality, it fetches
a low price, viz., from 11s. 6d. to 13s. 6d. per ton.
The following analyses of some carefully selected samples
of the hæmatite of the carboniferous limestone are by Messrs.
Dick and Spiller ("Memoirs of the Geological Survey of Great
Britain;" The Iron Ores of Great Britain, Part I.) :—
Lindale Moor,
Ulverstone.† Ulverstone.
86*50
Cleator
Cleator
Moor.
Moor.
Peroxide of iron .
95*16
90*36
94°23
Protoxide of manganese.
024
Ο ΙΟ
ΟΣΙ
0*23
Alumina.
0°37
0'51
Lime.
0'07
0'71
2.77
0'05
Magnesia
0.06
I
1*46
trace
Phosphoric acid
trace
trace
trace
trace
Sulphuric acid.
trace trace
ΟΙΙ
0°09
Bisulphide of iron
Water, hygroscopic
""
combined.
Insoluble residue.
trace 0*06
0°03
•
0°39
0*17
5.68
8.54
6.55
5'18
Carbonic acid.
2.96
* Dr. URE's Dictionary of Arts, &c., vol. ii., p. 682.
This column represents an unctuous variety of hæmatite (Gill brow ore)
used for lining puddling furnaces.
BROWN IRON ore.
331
The carboniferous limestones of Derbyshire and Somerset-
shire also contain veins and deposits of hæmatite, though
not equal in quality to those of Lancashire. The same ore
is also met with in the Devonian series of Devonshire, West
Somersetshire, and Cornwall.
In the Whitehaven district* two distinct varieties of hæma-
tite are recognised, the hard and the soft. The former
generally contains free silica in great excess, while the latter
contains silica, which, although it may to a great extent be
free, is associated with earthy bases diffused throughout it.
The siliceous red ores from Ulverstone contain on an
average from 41 to 57 per cent of iron, those from Somerset-
shire 59 per cent, from Cornwall and Devonshire from 39 to
60 per cent, and from North Wales from 43 to 49 per cent.
The calcareous red ore from the Forest of Dean, Lancashire,
and Cumberland contains from 40 to 47 per cent of iron; the
iron ores of Ulverstone sometimes contain a little arsenic.
2. Hydrates of Oxide of Iron.
The amount of water contained in these ores varies, and
influences the physical properties, chiefly the colour of the
ores; they are accordingly classified as follows:-
a. Brown Iron Ores; these have usually a brown streak,
seldom brownish yellow.
The common brown hæmatite, hydrated peroxide of iron,
2Fe₂03,3HO, when pure, contains 59.89 per cent of iron and
1444 per cent of water. It occurs in different varieties,
either finely fibrous (hæmatite), of slag-like appearance, or
earthy and ochreous.
Brown iron ore is a product of the decomposition of iron
pyrites, and occurs either separate or together with still
younger iron ores; it is frequently associated with mangani-
ferous ores, from which it derives its darkish brown and
black colour. Amongst the metallic associates of the brown.
iron ore are frequently iron and copper pyrites, galena, cala-
mine, malachite, &c.; amongst its earthy associates are calc
spar, heavy spar, brown spar, fluor spar, quartz, and very fre-
quently clay, constituting the aluminous brown iron ore, which
PERCY, Metallurgy, p. 200.
332
IRON.
sometimes occurs in a stalactitic form, or spheroidal, and
very abundantly in small globules and grains (oolitic brown
iron ores). The Greywacke iron ore also belongs to the
brown iron ores.
Upon roasting, the brown iron ores lose their water and
thus become porous; they are easily reduced and carbonised,
and form mixtures more or less difficult to fuse according to
whether manganese is present or not, and in consequence,
they either yield a grey pig-iron, or white pig well adapted.
for the production of steel. A little alumina renders them
more difficult to fuse, but well fitted for the production of
good iron if properly mixed. The siliceous brown iron ores
are the most difficult to fuse and to reduce (English hard
ores in contrast to the soft ores). Lime in the ore besides
the clay (Stahlberg in Siegen, Schmalkalden, Styria, Carinthia)
is very desirable, as such ores then contain all the components.
necessary for the formation of a good slag.
Purer manganiferous brown iron ores in the Pyrenees
(Northern Spain) are treated in open hearths for the direct
production of malleable iron and steel. Very rich pure
brown iron ores containing 65 per cent of iron in Biscaya*
are reduced, by Gurlt's method, in a cupola furnace by means
of carbonic oxide, to spongy iron, which is then heated in
Catalonian hearths and forged.
In many countries this is one of the most abundant and
valuable iron ores. It occurs in its different varieties in
many parts of Germany, and in the oolitic form it supplies.
by far the greater number of the French iron works, and
almost exclusively the Belgian iron works; it also occurs in
Spain, Russia, &c. Brown iron ore is the chief ore of the
United States of America; it is found in this country in
great abundance in the Forest of Dean, where it was worked
most extensively at a very early date, and though as a class
these ores are not rich, yet from the great masses in which
they are found, the cost of raising them is very low, about
2s. or 3s. per ton (Blackwell). These ores are associated
with lead and zinc ores, and are, therefore, liable to produce
за
*
Berggeist, 1863, No. 64.
YELLOW IRON ORE.
333
a cold-short wrought-iron; for this reason they are mostly
used for the production of foundry pig. The brown iron
ores from Wales, Cumberland, &c. are of better quality.
The varying quality of the Forest of Dean ores is shown
in the following analyses (Dr. Noad) :—
Water.
Carbonate of Lime.
Carbonate of Magnesia
Oxide of Manganese
Peroxide of Iron
Alumina
Sulphuric Acid
Phosphoric Acid
Insoluble Residue
•
I.
3*16
27'00 25*50 39'60
II.
III.
IV.
5'80
2'90
2.II
14*10
15:00 18:30
2500
17 10
40*80 38 10 29'00 59'70
6'00 3*60
trace trace trace
I'89
trace
trace trace trace 0°20
8.84 8.70 3°50 5'10
100*80 100'00 100'00 100*20
Some of the brown hæmatites contain a large percentage
of manganese. Their general composition is illustrated by
the following analyses (Dr. Noad) :—
Water
Oxide of Manganese
Lime.
Magnesia
Alumina.
Peroxide of Iron .
Sulphuric Acid
Phosphoric Acid
Insoluble Residue.
•
I.
II.
12.85
12.80
3*08
9'60
I'72
ΙΙΟ
I'20
0'92
68.57
68.45
ΟΙΙ
ΟΙΙ
ΙΟΙ
I'02
12'00
9'50
100'54
IOI 20
b. Yellow Iron Ore.-These are mostly of more recent
origin, and contain more water than brown iron ores, having
a brownish yellow streak. They occur either together with
brown iron ore, or as isolated masses in younger formations,
as the result of the decomposition of pyrites, carbonates, or
other compounds of iron. They are also deposited by water
containing iron vitriol (Rammelsberg), and they are even
formed under our eyes in the following manner :-Organic
acids produced from decaying vegetables, together with water
containing carbonic acid, extract iron from ferruginous rocks,
forming soluble iron salts. An admission of air precipitates
334
IRON.
these salts as insoluble basic salts and hydrated oxides
(limonite), whilst combining with certain components of the
vegetables-with phosphoric acid, for instance. Pure yellow
iron ores are seldom found in larger masses, but chiefly in
the following admixtures.
a. Aluminous Yellow Iron Ore, hydrated oxide of iron
in admixture with clay and sand, sometimes associated with
siliceous iron, hydrated oxide of manganese, and calc spar;
it also occurs grained in loose pieces, or cemented by clay,
massive or earthy. The aitites, or eagle stones, are also a
variety of this ore. On breaking the balls so named, they
are observed to be composed of concentric coats, the outside
ones very hard, but the interior becoming gradually softer
towards the centre, which is usually earthy and of a bright
yellow colour; sometimes, however, the centre is quite empty,
or contains only a few drops of water. Aitites occur in
abundance, often in continuous beds, in secondary moun-
tains, and in certain argillaceous strata.
The associates of this ore are metallic sulphides (iron
pyrites, galena, zinc blende), manganiferous minerals, calc
spar, brown spar, iron spar, quartz, heavy spar, gypsum,
cœlestine, &c. When derived from pyrites, it frequently
contains basic sulphate and arseniate of iron, sometimes also
phosphate of iron. The quality of the iron produced from
this ore depends on the amount of foreign associates.
These yellow ores are found on the Rhine, in Bohemia,
Silesia, Saxony, where they occur in the coal measures, and
on the Weser, and in the Hartz, Baden, Bavaria, Würtem-
burg, Hesse, France, Switzerland, and England, in the
oolitic group, or brown Jurassic formation.
B. Bog Iron Ore, Limonite, Lake Ore, &c., is formed in
the same way as the yellow iron ore, and is usually hydrated
oxide of iron with phosphate of iron and hydrated oxide of
manganese. Sometimes it is admixed with sand, clay,
organic substances, and siliceous iron, and now and then
with lime, magnesia, oxide of chromium, and sulphate and
arseniate of iron; in some cases it contains pure hydrated
phosphate of iron. According to Ehrenberg, the formation.
of bog ores is in part due to infusoria (diatomacea), which
BOG IRON ORE.
325
have the power of separating iron from water, and depositing
-it as hydrated peroxide in their siliceous coverings.
The ores either occur on the spot of their original forma-
tion, or they are carried into lakes by flowing water (Sweden,
Norway, Finland). In these countries large quantities of this
sort of bog ore, known as lake ore (sjomalmer), are obtained
by dredging the bottoms of the numerous lakes. It occurs in
granular concretionary forms, varying in size from grains of
coarse gunpowder up to 6 inches in diameter. The task of
collecting these ores is confined to the winter months; they
are raised by a perforated iron shovel fixed at the end of a
long pole, which is lowered through a hole about 3 feet in
diameter cut in the ice.
These ores are universally found in the neighbourhood of
reed banks and on the slopes of the shallows in the larger
and deeper lakes, in layers from 10 to 200 yards in length,
from 5 to 15 yards in breadth, and from 8 to 30 inches in
thickness; they never occur in strong currents of water.
Within a short time lake ore is reproduced, and there are
instances of lakes where the ore, after having been com-
pletely exhausted, collected again in the course of 26 years
to such an extent as to form beds several inches in thickness.
Bog ores are not found in this country in sufficient quan-
tity to be worth working; they are abundantly developed on
the continent, chiefly on the great plain of North Germany.
Owing to their large amount of manganese and phosphorus.
these ores are easily fused, and are inclined to yield white
pig-iron; wrought-iron produced from this pig is cold-short.
These ores containing a larger amount of phosphorus are
therefore usually smelted at a higher temperature for the
production of grey thinly liquid foundry pig-iron.
Ores with an excess of manganese and but little phos-
phorus may be used for the production of white radiated or
specular forge pig-iron, as, for instance, some Swedish lake
ores, which contain as much as 20 per cent of manganese.
One of the chief drawbacks in smelting bog ores is often the
abundant intermixture of sand, which can only be separated
by sifting the dry ores. At Alexishütte, near Lingen,
Welkner has constructed a sifting apparatus connected with
336
IRON.
a gas drying furnace. On sifting the ores at Tangerhütte
they are reduced to 70 per cent of their original volume.
The bog ore of Alexishütte, near Lingen, is composed as
follows:
Peroxide of iron
Oxide of manganese
Sand
Phosphoric acid.
Sulphuric acid
•
Water and organic matter
62.59
8°52
II.37
I'50
traces
16.02
100'00
3. Carbonate of Iron, either in a pure state as spathic
ores, or in an admixture with clay (argillaceous carbonate of
iron, clay ironstone); clay ironstone containing carbonaceous
matter is termed blackband.
a. Spathic Carbonate of Iron, FeO,CO₂, containing 48°2
per cent of iron, more or less carbonate of manganese (up to
20 per cent), and always combined with carbonate of lime
and magnesia, so as to form (FeO, MnO,CaO,MgO) CO₂.
When this ore has a fibrous structure it is called sphe-
rosiderite; in some cases the ore is found in a pure state
(Müsen,' Vordernberg), but it usually contains admixtures.
of calc spar, fluor spar, heavy spar, quartz, red and brown
iron ores, manganese ores, and of different metallic sulphides.
(iron, copper, arsenical pyrites, galena, silver, cobalt, and
nickel ores, &c.); sometimes the ore contains all the earths
required for the formation of slag. In all cases, the higher
part of the lode is changed into brown iron ore to a greater
or smaller depth, by the action of atmospheric air and
water.
Sparry iron ore appears to range through nearly the same
series of formations as the anhydrous hæmatites; it occurs.
in beds and masses often of immense extent, especially in
Styria and Carinthia. In the Erzberg, or ore mountain, near
Eisenerz in Styria, it is worked in an open quarry. The
Erzberg rises to a height of about 2500 feet, and apparently
consists of a solid mass of carbonate of iron, but it is in
reality only covered by a coping or arch of the mineral,
which varies in total thickness from 200 to 600 feet, including
SPATHIC CARBONATE OF IRON.
337
a few interstratified schistose partings. The deposit lies
upon and apparently passes on either side into limestone,
and is covered by a breccia of limestone fragments and clay
slate. The associated minerals are iron and copper pyrites,
quartz, carbonate of lime, and occasionally cinnabar.
The deposits of spathic ore in Carinthia are smaller in
extent, but of similar character.
In the Devonian rocks large quantities of spathic ores are
found in the district of Siegen, the Stahlberg or steel mountain,
near Müsen, being the most important deposit, where a
nearly vertical wedge-shaped lode in clay slate has been
worked since 1313.
In Anhalt and the Hartz, sparry iron ore forms large veins.
in Greywacke or Devonian limestone.
Other extensive deposits of this ore are found in the
Pyrenees and the Basque provinces of Spain, as near Bilbao;
and at Pacho near Bogota in New Granada.
In England ft occurs at Weardale in Durham, in lodes in
the carboniferous limestone associated with lead and zinc
ores, at Perran in Cornwall, Exmoor in North Devonshire,
and Brendon Hill in Somersetshire; between the two last
mentioned localities the ore forms a chain of lodes in the
middle Devonian rocks, which is said to be about 5 miles.
long, and of a maximum thickness of 27 feet.
Overmann states that in North America this ore is generally
impure, and not worked to any extent.
The variations in the composition of this important
mineral are shown in the following analyses-
I.
II.
III.
IV.
Protoxide of iron
53°50 43°59 52°56 43.84
Peroxide of iron
O'SI
Protoxide of manganese
•
6.50 17.87
4.82 12.64
Magnesia.
Lime
Carbonic acid
Metallic iron.
manganese
0'70 0*08 2.41 3°63
0°24 I'25 0'28
39°20 38°22 38.65 38.86
99'90 100'00 99*69 100'06
34.65
9'73
VOL. II.
Z
338
IRON.
No. 1, from the Pyrenees; No. 2, from Siegen; No. 3, from
Somersetshire, analysed by Noad; No. 4, from Brendon
Hill, Somersetshire, streaked with red hæmatite, analysed by
Spiller.
In the Permian rocks of Thuringia, a large irregular mass
of spathic ores, has been worked in the Mommel and Stahl-
berg mines, near Schmalkalden, for more than 700 years.
Pure or calcareous sparry iron ore, such as that mined in
Müsen, Thuringia, Styria, Carinthia, and lately in England
also,* is used chiefly for the production of spiegeleisen and
flowery pig, as it nearly always contains an amount of man-
ganese sufficient to render the mixture of easy fusibility, and
as the ores are easily reduced when roasted. In Westphaliat
these ores are smelted in admixture with blackband ironstone
for the production of grey pig-iron.
A small amount of sulphur in the ore is neutralised by the
manganese, and such ores yield an excellent white forge pig.
Ores with a larger amount of sulphur are smelted with hot
blast at a higher temperature and with an addition of lime
for the production of grey iron. A larger amount of mag-
nesia, sometimes amounting to one-third of the ore, renders
it difficult to fuse, and facilitates the formation of grey iron.
This ore seldom contains more than 5 per cent of carbonate
of lime.
Sparry iron ore in an undecomposed state is less easily
reduced, and scorifies some protoxide of iron, owing to its
easy fusibility, caused by the amount of manganese present;
also it only loses its carbonic acid at so high a temperature
that after its escape the ore begins to melt before being com-
pletely reduced. When weathered, or better roasted, the ore
reduces and carbonises easily, owing to its porosity, and the
more highly oxidised iron is less inclined to scorify. Pig-
iron produced from raw sparry ores is richer in manganese
than pig from decayed or roasted ores.
b. Argillaceous Carbonate of Iron, Clay Ironstone.-
This ore is a compact carbonate of iron in admixture with
* TUNNER'S Bericht über d. Londoner Industrie-Austellung, 1862, p. 13.
Berggeist, 1863, p. 195. B. u. h. Ztg., 1862, pp. 62, 354.
↑ Berggeist, 1857, No. 41; 1858, No. 5.
1858, No. 28.
B. u. h. Ztg., 1857, No. 36;
CLAY IRONSTONE.
339
clay, quartz sand, marl, &c., and sometimes associated with
calc spar, heavy spar, iron pyrites, galena, zinc blende, phos-
phate of iron and lime, &c. When occurring in the lower
schists of the coal formation it usually contains some bitumen,
and also a certain amount of manganese, up to 7 per cent
or more. When decayed these ores change into argilla-
ceous brown and yellow iron ore, and the action of the heat
transforms them into aluminous red iron ore.
These ores occur in the coal formations of Westphalia,
Saarbrücken, Silesia, &c., and very abundantly in the coal
fields of Great Britain, furnishing nearly two-thirds of the
total annual production of the United Kingdom.
The earthy or lithoid carbonates occur in some regions in
the upper limestone shales, and they extend upwards through
the coal measures proper towards their higher limits; they like-
wise occur in extensive beds in the Jurassic formation, particu-
larly in North Yorkshire; near the upper limit of the lias or
base of the oolite proper; and again, higher up, as nodules, and
perhaps in beds, in the middle oolite or Oxford clays. They
are also found extensively as courses of nodules in the Wealden
series, and as beds in the green sand. When these grey car-
bonates contain lime in abundance, and when clay is not
largely present, they are sometimes changed by atmospheric
influences into hydrated hæmatites; in Northamptonshire, for
example, and extensively in France. The only great coal field
in Great Britain in which these ores do not occur in sufficient
abundance to form the basis of a large production of iron are
those of Northumberland and Durham, and of Lancashire.
In England, these ores contain on an average 30 or 33 per
cent of iron, from 15 to 25 per cent of alumina, small amounts
of sulphur (in blende, galena, iron and copper pyrites), and
always from 0.5 to rather over 1 per cent of phosphoric acid.
The iron produced from the argillaceous ironstone is of an
extremely good quality, provided the coals used for smelting
are good. The ore is always used in a calcined state, by
which it loses in weight about one-third or one-fourth, the loss
consisting of carbonic acid and water. The production of
iron in South Wales and South Staffordshire rests almost
entirely on the great beds of this mineral.
Z 2
340
IRON.
In France, the clay carbonates of the coal measures are
only worth working in three localities:—in the coal fields of
the Gard, of the Aveyron, and, to a very limited extent, in
the coal field of the Loire, near St. Etienne (K. Blackwell).
In America, they are largely distributed over Pennsylvania,
Maryland, Virginia, Ohio, Illinois, North Carolina, and
Kentucky, but here they are not in general use, owing to the
expense of raising and working them.
The varying proportions of iron, silica, and alumina
which they contain, is shown in the subjoined analyses of
the ore from different localities:
Scotch Varieties, analysed by Dr. Colquhoun.
I.
II.
Carbonic acid.
32'53
30.76
Protoxide of iron
35°22
38.80
Protoxide of manganese
0°07
Lime
8.62
5°30
Magnesia
5°19
6'70
Silica
9'56
10.87
Alumina
5'34
6.20
Peroxide of iron
1.16
0'33
Carbon .
2.13
1.87
Sulphur
0.62
0.16
100°37
ΙΟΙΟΟ
Welsh Varieties, analysed by Dr. Noad.
Red Vein.
Red Vein Pin.
Silica
8.31
15°40
Alumina (insoluble).
3'13
5'00
Carbonate of iron
73'79
57'99
Oxide of manganese
0*92
o'64
Carbonate of lime
2.95
3'45
Carbonate of magnesia
3.80
8.58
Alumina (soluble)
2.52
3'52
Phosphoric acid .
0'53
0*75
Sulphuric acid
trace
trace
Bisulphide of iron
0'17
0*24
Potash.
0'48
0'45
Organic matter and water
2.36
2.34
98.96
98.36
Metallic iron.
35.62
28'00
BLACKBAND IRONSTONE.
341
c. Blackband Ironstone.-This ore is a clay ironstone
containing carbonaceous matter, and occurs in layers in the
upper strata of the coal formation. Its composition is nearly
that of the clay ironstone, but it always contains less sandy
clay and the carbonaceous matter already mentioned. When
heated this ore loses half its weight, and sometimes contains
enough intermixed lime for the formation of slag. The
smaller amount of sandy clay causes the blackband ironstone
to be more easily fusible, and its being less compact after
As
roasting renders it easier to reduce than clay iron ore.
blackband usually contains a larger amount of metallic sul-
phides, especially iron pyrites, and not less phosphorus than
clay ironstone, an inferior pig-iron results, usually in the form
of grey foundry pig, produced at a high temperature with hot
blast and a large addition of lime. The treatment of black-
band containing many impurities is very difficult, owing to
the great fusibility of the ore. This great fusibility and the
amount of bitumen in the ore occasion a considerable saving
of fuel, whilst the richness of the ore allows a large produc-
tion. The blackband may be calcined by burning it in heaps
without any additional fuel, the residues yielding from 50 to
70 per cent of iron.
This ore was discovered in Lanarkshire by Mushet in the
year 1801, but only came into use at about 1830. In the
western coal-field of Scotland* seven principal blackband
measures are known, of the following average thicknesses
respectively,—12, 16, 6, 8, 6, 15, and 8 inches. The yield of
blackband is at the rate of 2000 tons of calcined ore, equal
to 1000 tons of pig-iron per acre for each foot of thickness.
Blackband also occurs in North Staffordshire and South
Wales.
In South Wales a carbonaceous spathic ore, termed coal
brass, occasionally accompanies the coal. It differs from
blackband in containing considerable quantities of carbonates
of lime and magnesia.
The Scotch blackband iron ores contain from 20 to 25 per
cent of coal, from 10 to 15 per cent of alumina, and from
*
BAUERMAN'S Treatise on Metallurgy of Iron, 1868, p. 76.
342
IRON.
34 to 41 per cent of iron. Upon calcining the ores and thus
evolving from 40 to 50 per cent of volatile substances, this
amount is increased to from 55 to 60 per cent.
In 1855 blackband was discovered in Westphalia in for-
mations poor in coal, but in much less quantity than in
Scotland and England. In a few instances only do black-
band and coal occur together, whilst in this country this is
always the case. When calcined, the Westphalian black-
band loses 30 per cent of its weight; it contains 40 per cent
of iron when raw, and 60 per cent after roasting, besides as
much as 0.4 per cent of sulphur, and o'2 per cent of phos-
phorus.
Blackband also occurs at Gablau* in Lower Silesia, and
thin layers of it have been discovered at Ostraut and at Steier-
dorf in Banat.
Analyses of Blackband.
Dr. Noad found blackband from the neighbourhood of
Pontypool, South Wales, to contain-
Carbonaceous matter
Carbonate of iron.
magnesia
lime.
I.
II.
15.00
13°42
61.00
64°44
10 90
13°54
13°20
8.60
ΙΟΟΙΟ
100'00
Metallic iron per cent.
• 29.60
31°10
""
""
Analyses of Scotch Blackband.
Scotch blackband ironstone, analysed by Dr. Colquhoun:-
Ferric oxide
•
Ferrous oxide
Lime
Magnesia
Silica
Carbonic acid.
Water, combined
Carbonaceous matter
I.
II.
0*23
2.72
53.82
40'77
1'51
o'go
0.28
0'72
2.00
ΙΟ ΙΟ
34°39
26.41
I'00
7'70
17.38
The following analyses show the composition of blackband
from Yorkshire and South Wales:-
Allgem., B. u. h. Ztg., 1859, p. 97.
† LEOB., Jahrb., 1861, x., 284.
CLEVELAND IRON ORES.
343
I.
II.
III.
Ferric oxide
I'45
Ferrous oxide
36.14
37°07
42.64
Manganous oxide
1*38
0*23
0'26
Alumina
6'74
Lime.
2.70
6.61
5°24
Magnesia
2.17
7°40
5°26
Potash
0*65 Clay
270
Silica
17°37
Carbonic acid
26'57
37°14
36.89
Phosphoric acid
0'34
0'23
0'17
Sulphuric acid
trace
Iron pyrites
ΟΙΟ
trace
0°22
Water
(hygroscopic.
0.61
combined
1.16
Carbonaceous matter
2'40
9.80
8.87
99'78
100'18
99*55
No. I is blackband from Low Moor in Yorkshire, analysed
by J. Spiller. Nos. 2 and 3 are blackband (coal brass) from
Aberdare in South Wales, analysed by Price and Nicholson.
Cleveland Iron Ores.-These important iron ores were
only discovered in 1851 at the Cleveland Hills in Yorkshire;
they occur in the middle lias; in its best development the
bed of ore is about 20 feet thick; from 12 to 17 feet are
workable, the bed being made up of various interstratified
bands of ore, shale, and iron pyrites. The two seams of the
stone bed are principally distinguished as the pecten and
avicula seams, from the respective prevalence in them of fossil
shells belonging to these genera.
The average composition of these ores may be seen from
the following analyses by Mr. Crowdar :—
Silica
Peroxide of iron
Eastern Nab Hutton Low
(Main Seam).
Cross.
II 95
15.65
6.73
1.80
Protoxide of iron .
39'05
35'75
Alumina
13.83
4'95
Lime
2°52
7.39
Magnesia
2.72
2.98
Sulphur.
trace
trace
Phosphoric acid
I'02
5'05
Carbonic acid.
16.38
23.47
Water
5.80
4.89
4
344
IRON.
4. Silicates of Iron.
a. Siliceous Iron Ores are composed of silica with either
peroxide or protoxide of iron, in some cases also with both
oxides; also they are usually combined with other silicates,
hydrates, &c. The following are some of these ores:-
Chamoisite, 2(3FeO,SiO3)+6(FeO,Al2O3) + 12HO, con-
taining 49 per cent of iron.
Iron garnet, 3FeO,SiO3+Al₂O,,SiO3, containing from 18 to
31 per cent of iron.
Nontronite, Fe2O3,2SiO3+6HO, containing from 21 to
26 per cent of iron.
Yellow earth, Al2O3,SiO3+2(Fe2O3,SiO3)+6HO, containing
26.5 per cent of iron.
Upon smelting these ores, the ready fusibility of the proto-
silicate of iron, and the difficulty of its reduction, give rise
to the formation of a ferruginous slag and the production of
white pig-iron, i.e., to an irregular process, as the ores have
an inclination to smelt before being reduced and carbonated.
The reduction of the ores is also rendered still more difficult
by their compactness. When roasting the ores in order to
render them less compact and to transform the protoxide,
which readily scorifies, into peroxide, a caking of the ores is
likely to take place. It is advisable to throw the hot roasted
mass into cold water; the larger pieces will break up so
as to be easily crushed, when they will be more readily
reduced.
Owing to these properties, the silicates of iron are seldom
smelted by themselves, as for instance, chamoisite at Arden,
kieseleisen-glimmerschiefer* (a sort of mica slate containing
silicate of iron) in North Carolina. Ferruginous silicates
are therefore more used as fluxes, part of their iron being
extracted in this way.
In their behaviour these iron ores approach that of
ores containing protoxide of iron and an intimate ad-
* B. u. h. Ztg., 1857, P. 243.
SLAGS OR CINDERS.
345
mixture of quartz, like the siliceous magnetic iron ores
(page 325).
Clay iron ores contain silica, not combined with protoxide
of iron, but with other bases; they are, therefore, more
readily reduced, although they are less fusible. Iron ores
containing in admixture silicates suitable for the formation
of slag (Swedish magnetic iron ores, Styrian spathic iron
ores, some carbonaceous iron ores) are usually of appropriate
reducibility and fusibility.
b. Slags or Cinders produced in the operations of
puddling and re-heating, containing 40 or 75 per cent of iron.
These substances consist essentially of protosilicate of
iron, intermixed with more or less peroxide of iron; they
behave like siliceous iron ores with regard to reduction and
fusion; but they contain part of the phosphorus and
sulphur of the pig-iron from which they were produced, and
require somewhat more fuel than the ores, as their power
of conducting heat is greater. With regard to purity, fusi-
bility, and power of conducting heat, those slags may be
classified as follows:-re-heating slags, puddling slags; with
regard to their reduction, they stand in inverse proportion;
these slags are purer if resulting from pig-iron which
has been previously refined. Roasting renders these slags
somewhat less compact, the re-heating slags the least so;
the puddling slags are somewhat more reacted upon and the
influence is greatest upon slags resulting at the conversion.
of pig-iron into wrought-iron in open hearths. The
chief advantage of the roasting is, that an easily fusible
silicate, 6FeO,SiO,, which is difficult to reduce, and contains
the greater part of the phosphoric acid (Percy * found 13
per cent), will liquate, whilst a purer compound, 3FeO,SiO3,
will remain; this purer compound is more difficult to fuse,
but easier to reduce in the blast furnace. The coal con-
sumed at this liquation processt amounts to about half the
weight of the slags under treatment. Besides the expense
of this process it causes a great loss of iron.
* TUNNER'S Bericht über d. Londoner Industrie-Austellung, v., 1862, p. 36.
† Revue Universelle, 1860, 4 année, 4 livr., p. 57.
346
IRON.
This unfavourable behaviour of the slags has led to their
being mostly added to the ore mixture in such a degree as
the quality of the produced iron will allow. For example, in
this country, white forge-iron of different quality is pro-
duced by a variable addition of puddling and refinery slags
to the ore mixture; and a mixture of refinery slags and
clay slate is only used for the production of slag iron in-
tended for ballast blocks for vessels; the amount of phos-
phorus in pig-iron has not been observed to increase, pari
passu, with the increased addition of slag. In Scotland, 29
per cent of slags is added to the ore mixture, and at Königs-
hütte, in Silesia, as much as 42 per cent has sometimes been.
added.
In former times frequent attempts were made to smelt
slags by themselves, and also to reduce them in open
hearths and Stücköfen, or high bloomery furnaces; and
Berthier, Reichenbach, Calvert, Hinde, and others have re-
peatedly proposed that roasted and finely-divided slags
should be smelted in intimate mixture with lime and coal.
This method has, of late, been practically and successfully
applied by Lang and Frey,* and has been introduced in
many smelting works, professing to cause the production of
a purer iron, chiefly poorer in silicon, than that resulting
from smelting slags with an addition of iron ores. The
produced pig-iron, of course, is of different quality, according
to the purity of the slags. Even spiegeleisent has been
produced by this method. The following is stated to be
its composition:-
Iron .
Carbon
Sulphur.
Phosphorus
Silicon
94'03
5°14
trace
0°32
0*40
But the smelting of slags is always more expensive than
that of iron ores.
Tunner
1861, p. 327.
recommends smelting slags together with from
* Oesterr. Ztschr., 1860, p. 321; 1861, pp. 36, 43, 52, 60.
TUNNER'S Leoben. Jahrb., 1861., X., 289, 415;
† Oesterr. Ztschr., 1860, p. 326; 1861, p. 43.
‡ Ibid., 1860, No. 24. B. u. h. Ztg., 1861, p. 223.
B. u. h. Ztg.,
1862, xi., 299.
PERCENTAGE OF IRON IN ORES.
347
6 to 8 per cent of granulated pig-iron, and from 10 to 15
per cent of lime, when the pig-iron will introduce carbon
into the slags, and thus exert a strong reducing action.
5. Scrap Iron.
This is also used, in some cases, as smelting material,
and given in addition to the ore mixture.
a. Scraps of Cast-iron in fragments and other forms are
sometimes added at the commencement of an operation of
a blast furnace to rapidly increase the quantity of iron in
the hearth; and if the re-melting in cupola furnaces is too
expensive, or the material is not present in sufficient quan-
tity for carrying on the smelting in the cupola furnace.
These scraps may be added in large pieces at greater
intervals, without impeding the process, if the quantity of
fuel is charged accordingly; but one piece of scrap iron
must not be heavier than one charge of ore mixture. Ferri-
ferous bears may also be made useful in this manner. At
Witkowitz, pieces of cast-iron weighing 40 cwts., and ferri-
ferous bears, weighing 34 cwts. are re-melted in blast fur-
naces. Cast-iron borings and turnings are best caked by
a previous roasting, which may be facilitated by moistening
them, either with salt water or with hot water and muriatic
acid, when pressing the material into moulds.*
b. Scraps of Wrought-iron, ends of rails, &c., are less
often smelted in blast furnaces as an addition, than added
at smeltings in cupola furnaces, and at the conversion of
pig-iron into wrought-iron. At Ebbw Valet such scraps,
chiefly rails-ends, are treated with coke and an addition of
lime in a cupola furnace 15 feet high.
The Adaptability of Iron Ores to the Smelting Process
depends-
1. On their Percentage of Iron, which varies between
30 and 40 per cent; sometimes they contain even more than
70 per cent, and in some cases so little, owing to a great
admixture of foreign substances, that the ore is not worth
*
Oesterr. Ztschr., 1861, Nos. 25, 26, 38. Polyt. Centr., 1857, p. 828. LEOB.,
Jahrb., 1862, xi., 47.
+ TUNNER'S Bericht über d. Londoner Industrie-Austellung, v., 1862, p. 28.
348
IRON.
smelting by itself, but may perhaps be used as a flux, being
often preferable to fluxes which contain no iron.
Magnetic, red, specular, and carbonaceous iron ores are
classed amongst the richer ores, whilst different brown and
argillaceous ironstones form the poorer class. The amount
of iron required in an ore to make it worth smelting de-
pends on local circumstances, prices of materials, labour,
&c.; at the Dannemora Iron Works, ore mixtures containing
less than from 37 to 40 per cent of iron are considered too
poor, and the usual mixture contains 50 per cent and more ;
in most other places a profitable mixture contains only
about 30 per cent and even less. Iron ores containing less
than 15 or 20 per cent cannot be profitably smelted by them-
selves; the foreign admixtures often cause a richer ore to
be less valuable than a poorer ore.
The production of pig-
iron* in a certain time is limited by the amount of iron
contained in the ore; whilst 25 cwts. of pig-iron and
slags are produced weekly for every cubic yard of the con-
tents of a blast furnace, if smelting clay ironstones con-
taining 38 per cent of iron, the production amounts to
34 cwts. when treating carbonaceous iron ores containing
60 per cent.
2. On the State of Aggregation of the Ore, and the
State of Oxidation of the Iron contained in it.-Protoxide
of iron is more inclined to scorify than peroxide, and the
inclination is increased the more compact the ore is. Iron
ores containing peroxide of iron are also more difficult to
reduce, and to carbonise, the more compact they are. In
such cases the ores must be exposed for a longer time to the
influence of the reducing and carbonising agents, and this
may be effected by rendering the ore mixture difficult to fuse;
otherwise, non-reduced protoxide of iron will partly decar-
bonise, in the hearth, the pig-iron formed. With regard to
this behaviour, brown, yellow, and sparry iron ores, which
disintegrate when strongly roasted, are preferable to the
compact magnetic, specular, and red iron ores, provided they
do not contain disseminated carbonate of lime.
*
Allgem., B. u. h. Ztg., 1863, p. 230.
Foreign
}
FOREIGN ASSOCIATES OF IRON ORES.
349
admixtures sometimes impede the reduction and carbonisa-
tion of the ores (clay ironstone). Precautions are required
if ores containing protoxide of iron are replaced in an ore
mixture by ores containing peroxide.
If protoxide of iron is scorified, owing to an imperfect
reduction, the silicate of iron formed can only be reduced at
a great consumption of fuel. The formation of silicate of
iron is facilitated if the ore contains protoxide of iron or
finely disseminated quartz, if the ore is rich and very compact
or if the temperature is either too high or too low; too high a
temperature causes a smelting of the ores before they are
reduced; and a low temperature is not sufficient for
reduction.
3. The Foreign Associates, with regard to quality and
quantity, greatly influence the smelting process as well as
the product.
In very rare cases only do iron ores occur free from
admixtures of earthy and metallic substances (magnetic,
spathic, and specular ores); besides water and carbonic acid
the ores usually contain silica, alumina, and lime, either one
or more of these substances, if not all; sometimes they also
contain sulphur, phosphorus, barium, manganese, copper,
zinc, lead, and other metals. Some of the substances have a
favourable influence and others are hurtful. Though to a
certain extent it is possible to neutralise the influence of the
injurious substances at the smelting process, it can seldom
be done sufficiently, and usually causes economical disad-
vantages; these substances must therefore be removed as
much as possible by suitable chemical and mechanical
preparations, previously to the smelting operations (sulphur,
page 301; phosphorus, page 306; arsenic, page 314; copper,
page 314; zinc, page 315), provided essential components
of the mixture are not separated (silicon, page 296; earthy
metals, page 318), or they are contained in the fuel (silicon
in the ash of coke, sulphur in coal and coke, phosphorus in
turf). Some iron ores also contain a little tin.
A few substances only improve the quality of iron, chiefly
manganese (page 311) in iron ore and fluxes, and titanium
(page 316); in some particular cases the presence of
350
IRON.
phosphorus (page 309) and sulphur (page 304) even may
be desirable.
The ores are classified, in respect to smelting, into refrac-
tory and easily fusible ores, and one medium sort. And, as
the ores are seldom pure, and usually contain some compo-
nents required for the formation of slag, Sefström* classifies
them into ores with, and ores without, such components.
To smelt the purer ore without these components addi-
tions must be given suitable for the formation of slag
(mostly slag of the same process, if poor ores are not to be
had), as otherwise the pig-iron formed will be again oxidised
in the hearth by the reaction of the blast, and also as the
hearth walling will be rapidly destroyed. The thickness
of slag cover upon the pig-iron in the hearth is regulated
chiefly according to the pressure of the blast, which depends
on the compactness of the fuel, and also according to the
intended quality of the pig-iron, as grey iron requires more
preservation from the blast than white pig.
The richness of the ore mixture varies accordingly.
The ores containing components for the formation of slag
may be subdivided into—
a. Ores containing all the components required for the
formation of a suitable slag.
These are chiefly ores containing clay and lime at the
same time, or certain minerals which are easily fusible
(garnet, amphibole, augite, chlorite, felspar, &c.)
Garnet, 3(CaO,MnO)SiO3+(Fe₂O,,Al2O3)SiO3, smelts well,
yielding an excellent slag, and contains from 10 to 12 per
cent of protoxide of iron. Augite,
3(CaO,MgO,FeO),2(SiO3,Al2O3),
containing up to 12 per cent of protoxide of iron, behaves
similarly. Hornblende,
CaO(SiO3,Al2O3)+3(MgO,FeO),2(SiO3,Al2O3),
is easily fusible if containing up to 15 per cent of FeO, and
still more so if in admixture with felspar; it changes into
greenstone with an amount of about 20 per cent of iron.
*
LEOB., Jahrb., 1843, p. 96; 1853, iii., 261.
CALCAREOUS ORES.
351
Chlorite, 3(MgO,FeO)SiO, +Al2O3,SiO3+2MgO,HO, is a
good easily fusible component owing to its iron (up to 15 per
cent FeO). Felspar, (Al,O,,SiO,+KO,SiO3), is somewhat
more refractory if smelted by itself, but is rendered of easier
fusibility if other more basic minerals are present.
The ores sometimes allow the production of white pig-iron
and sometimes of grey pig, according to whether they con-
tain more or less manganese, and whether they become more
or less porous upon roasting.
b. Ores containing all the components, but not in the
appropriate quantitative proportion.
These ores are those containing highly siliceous compo-
ponents (talc, much felspar), or too basic components (ser-
pentine, &c.); 2(3MgO,SiO3)+3HO+3(MgO,HO), or compo-
nents too rich in magnesia (dolomite), which render the ores
difficult to fuse. These ores then require additions of lime
and quartz, or clayey marl, or clay slate and lime, &c.
c. Ores not containing all the components required for the
formation of slag, but some of them, viz. :—
Siliceous Ores, which are fusible to a different degree,
and also vary in their behaviour according to whether they
contain quartz in coarse grains (page 325), or finely dissemi-
nated (page 325), and whether the silica is combined with
per- or protoxide of iron (page 344), or partly combined with
other bases. Ores containing peroxide of iron and coarse.
grained quartz or clay are the most difficult to smelt; the
ores containing protoxide of iron are less difficult to smelt.
as they are more inclined to form silicate of iron, chiefly if
the ores contain the silica in intimate admixture, and pro-
vided the ore does not already contain the iron as silicate
(page 345). A certain amount of manganese increases the
fusibility. These ores require different fluxes according to
the state of the silica present, and the resulting pig-iron will
be richer in silicon the more intimately the silica is mixed
with the ore.
Calcareous Ores.-Carbonate of lime alone renders the
ores difficult to fuse, and more so if the lime contains mag-
nesia (dolomite, brown spar), but less so if it contains alumina
(marl) or manganese (some kinds of brown spar). Owing to
352
IRON.
the presence of magnesia, the brown spar is usually sorted
out when dressing the ores. Fluor spar facilitates the
smelting, and its fluorine, besides the lime, assists the
removal of silica, therefore one part of fluor spar may
replace several parts of carbonate of lime. Gypsum and
heavy spar, owing to their amount of sulphur, are not usually
present in sufficient quantity to influence the smelting pro-
cess. A small amount of these substances may be rendered
innocuous by the presence of manganese; a larger amount
gives rise to the formation of much sulphides of calcium and
barium, thus rendering the slag difficult to fuse.
The Dressing of the Iron Ores.
The dressing or preparation of the ores aims at the sepa-
ration of injurious components, and of such non-metallic
associates as are not required for the formation of slag. The
more carefully they are prepared, the better will be the
quality of the produced iron, and the less fuel will be used
(12 to 16 per cent) at the smelting process.
The proportionally low prices of pig-iron limit the pre-
paration of the ores to the most simple operations, and
make it almost impossible to be done in this country, where
the price of workable ores is low, and the supply sufficient;
the high price of labour would also form an impediment.
The dressing operations usually consist of hand sorting
or washing, according to the nature of the ores.
rare cases the ores are crushed and treated on racks. For
dividing the ores, crushing mills are preferable to pounding
mills, as they produce less ore dust, but Blake's stone-breaker
gives still better results.
In some
In order to save carriage it is advisable to prepare the
ores on the spot where they are raised.
When sorting by hand, the larger pieces of ore are kept
separate from the small ore, and broken with a ragging-
hammer, whilst the non-metallic associates and injurious
substances, such as iron pyrites, heavy spar, &c., are picked
out. The mass is previously washed on an inclined plane
in order more readily to recognise the pieces of ore; but ores
THE DRESSING OF IRON ORES.
353
containing hæmatite in froth-like aggregations are sorted as
soon as they come out of the pit, as a shower of rain will
coat the ore pieces with the iron froth, thus rendering them
indistinguishable.
In some ores, magnetic iron ore, for example, when first
raised, the ore cannot be discerned from the gangue. This
is the case when the ores have been exposed for some time
to the atmospheric air. When first raised, argillaceous ores
of the coal measures, likewise occurring in nodules, are often
difficult to separate from adherent fragments of shale. If,
however, they are exposed to the air for some time, a super-
ficial oxidation takes place, and the shale disintegrates, and
can readily be removed. Instead of this tedious weathering
process Harding employs steam to separate the clay from
the English spherosiderite, thus completing the preparation
in a few hours.*
As before mentioned, iron ores are seldom crushed and
washed either in settling pools or on racks. This is done at
the Schreckendorfer smelting works, purifying an impure
roasted magnetic iron ore so much that it may be given to
the ore mixture in such quantity as to form two-thirds of the
mixture. The cost of separating 1 ton of ore amounts to
about 7d., the loss of ore to from 0.8 to 12 per cent, and
about 100 cwts. of ore are dressed in 24 hours.
At the smelting works in Hassel (Norway) an iron mica
disseminated in gneiss, and containing 25 per cent of iron,
is enriched by this dressing so as to contain 55 per cent. At
Taberge in Smaland, magnetic iron ore disseminated in basalt
is likewise enriched from about 28 per cent to 43 per cent.
Sella has tried to prepare magnetic iron ores containing
iron and copper pyrites from Traversella in Italy by means
of a magnet, and Chenot has endeavoured to liberate iron
ores from gangue after having made them magnetic by means
of an electro-magnetic wheel.
Iron ores intermixed with clay, sand, loam, &c., and the
pisolitic ores of the oolitic formations, are frequently
much enriched by a washing operation. Sandy bog-ores
VOL. II.
*
Berggeist, 1858, p. 165.
2 A
354
IRON.
may be separated from the adherent sand by dry sifting
(page 335).
The following is an account of the treatment of an argilla-
ceous yellow ochrey ore at the Carolina mine, near Stolberg,
in Rhenish Prussia:
After a preliminary breaking and sorting by hand, the ore
is placed in a cylindrical trough, in which revolves a hori-
zontal shaft armed with projecting paddles, whilst a stream
of water flows through it and removes the fine muddy par-
ticles. The remaining ore is then thrown on to an iron
riddle, and the small pieces falling through are washed in an
inclined gutter in a stream of water. The slime, together
with that from the trough, is carried into settling pools, de-
positing a ferruginous mud, which is moulded into bricks and
burnt, yielding a product containing from 40 to 43 per cent.
of iron. The production from two washing apparatus, em-
ploying seven men, is 500 cwts. in 12 hours, at a cost of 17S.,
or 7åd. per ton for washing.
Extensive preparations are made in Belgium for washing
loamy brown iron ores; this is effected on the pit by
stirring the ores in wooden troughs through which a stream
of water is kept flowing. This washing is repeated in the
smelting works, where the ores are treated in inclined cylin-
drical drums of cast-iron, 2 metres long and o°8 metre in
diameter, armed internally with projecting spikes, which are
made to rotate while the ore is fed in at the upper end, a
stream of water passing through at the same time. The
particles of ore and clay, loosened by the tearing action of the
spikes, are discharged at the lower end into a trough, where
they are separated by a current as in the preceding instance.
The cost for I ton of enriched ore, containing from 35 to
40 per cent of iron, amounts to about 7s.
In Würtemberg and Baden pisolitic brown iron ores, con-
tained in a ferruginous and calcareous marly matrix, are
cleaned by jigging.
As we have before mentioned, iron ores are frequently
exposed previous to smelting to the action of the air for
a longer time, in order to disintegrate them, to separate
clayey admixtures in a mechanical way, to oxidise certain
CALCINING IRON ORES.
355
deleterious components, thus converting them into a soluble
state, and to oxidise more highly the protoxide of iron; all
these reactions will facilitate the smelting process. Man-
ganiferous sparry iron ores, when first raised and added to
the ore mixture, yield a pig-iron richer in manganese than
that produced by similar ores which have been weathered.
Very compact iron ores, particularly if they contain sul-
phides, are submitted to a weathering after being roasted;
we shall explain this process later on.
Roasting or Calcining Iron Ores.
The experience of many ages teaches that iron ores when
first raised, upon smelting, give less favourable results with
regard to reducibility, consumption of fuel, largeness and
expense of the production, than ores which have been either
weathered (page 334) or roasted. Dusty and friable ores or
small ores, particularly if free from pyrites, are not roasted;
when roasting pure ores the fuel employed must be of a kind
unlikely to contaminate the ores.
Spathic ores in a raw state are employed without being
weathered, if the process is for the purpose of introducing a
greater amount of manganese into the pig-iron, and an addi-
tion of raw ores is given to roasted ores to facilitate the pro-
duction of mottled iron.
In this country all kinds of iron ores (clay ironstone,
Cleveland ores, spathic and brown iron ores) are subjected
to the process of roasting before smelting, with the exception
of red iron ore.
The process of roasting renders the iron ores easier to
reduce, and they require a less refractory slag for smelting;
it also saves fuel and lime flux, whilst the production is in-
creased and the quality of the pig-iron improved.
The process of roasting is intended to effect-
1. A disintegration of compact iron ores in order to divide
and reduce them more easily, as by the effect of the fire the
thick ore pieces split and crack, and are thus more easily
divided and more perfectly permeated by carbonic oxide;
magnetic iron ore splits more than red and specular iron ore.
2 A 2
356
IRON.
For this purpose quartzose ores are best quenched in water
when still hot. These ores have a finely grained fracture;
magnesian ores are more coarse grained and feel soft to the
touch; calcareous ores show a white dust upon their frac-
tures; clayey ores show some lustre ; and manganiferous ores
sometimes have an iridescent fracture.
2. The removal of volatile substances, such as water, from
brown and yellow iron ores, carbonic acid from spathic iron
ores and from ores having a calcareous gangue, and carbon
from blackband. The ores thus become not only very porous,
and consequently well adapted for the reduction and car-
bonisation of the iron, but they may also be economically
smelted with advantage. Raw ores in the upper parts of the
blast furnace absorb part of the heat intended for the expul-
sion of the volatile substances, and part of the carbonic acid
is only evolved at temperatures sufficient for it to be de-
composed by the fuel, thus causing a waste of fuel.
3. The decomposition of sulphides and arsenides, which
are transformed, on the admission of atmospheric air, into
oxides and salts; at higher temperatures the salts are mostly
converted into oxides. For this purpose steam has been
successfully employed. The decomposition will be most
perfect if the ore becomes sufficiently disintegrated, if the
admitted air is rich in oxygen, and if the ore pieces are of
suitable size and have been exposed to a suitable tempera-
ture for a sufficiently long time. Reducing gases evolved in
the roasting process (from blackband or intermixed fuel) im-
pede the oxidation of the sulphides, causing the formation
of lower sulphides or basic sulphates. These lower sulphides
are preferable, as they may be removed by a weathering and
washing, whilst the basic sulphates are insoluble in water.
On the other hand, a reducing reaction may facilitate the
decomposition of arsenical pyrites. Carbonates present in
the iron ore are transformed into sulphates, and some of
them (sulphate of lime and magnesia) may be lixiviated out,
On
whilst heavy spar and phosphates are not reacted upon.
throwing water upon roasted and still glowing ore it not only
becomes disintegrated, but a formation of sulphuretted
hydrogen gas also takes place.
ROASTING IRON ORES.
357
4. The transformation of protoxide of iron into peroxide,
which scorifies less readily, by the oxygen of the atmosphere
or by the evolved carbonic acid. This transformation is
effected whilst the above mentioned processes are going on
(magnetic iron ores, spathic iron ores). It will be more per-
fect the more porous the ores are and the greater the admis-
sion of air. 100 parts of protoxide of iron yield 1114 of
peroxide; and 100 of magnetic oxide about 103 of peroxide.
The loss upon roasting the different varieties of ironstone
are as follows:
Magnetic iron ores lose from 3 to 6 per cent of hygroscopic
water, but their weight is again somewhat increased by the
higher oxidation of the protoxide; specular and red iron ore
lose from 3 to 5 per cent, brown iron ores from 10 to 15 per
cent, limonite from 8 to 20 per cent, and sparry iron ores up
to 40 per cent, but their weight is again somewhat increased
by taking oxygen into combination. Spherosiderites lose
from 18 to 30 per cent, and blackband frequently above
40 per cent.
Upon laying out the roasted ores in the air they again
absorb water in different quantities, according to their state
of aggregation; dusty and clayey ores absorb about 6 per
cent; they are therefore preferably kept under shelter. Cal-
careous ores when roasted contain caustic lime, which
disintegrates by absorption of water, transforming the ore
into a pulverised condition; therefore, they must also be
kept under shelter. 100 parts of carbonate of lime yield
56 parts of caustic lime and 74 parts of hydrated lime.
The proper regulation of the temperature and method of
roasting are of essential influence in the roasting process.
The temperature must not rise so high as to cake or melt
the ores, as they would become more instead of less com-
pact. Magnetic iron ores are particularly inclined to cake,
especially if they contain silica intimately admixed; iron-
stones, chiefly those containing protoxide, if they are asso-
ciated with easily fusible silicates or substances facilitating
the fusibility, such as manganiferous compounds, also have
an inclination to cake. Ores capable of bearing only a dark
red heat must not be roasted together with ores which may
358
IRON.
C
be heated up to light red heat without caking, and ores free
from sulphur must not be roasted with sulphurous ores.
Easily fusible ores containing sulphur and compact at the
same time are twice slightly roasted (magnetic iron ores in
Sweden and in the Hartz); they are sometimes weathered
between the two roasting operations (Sweden).
Besides the intensity of the temperature the length of the
operation is important, that the single ore pieces may be
thoroughly heated so that the oxygen may perfectly per-
meate them. The size of the pieces must be judiciously
chosen; pure ore may be roasted in pieces the size of one-
quarter of a cubic foot, but the pieces of ore containing inju-
rious substances must not be larger than from 30 or 40 cubic
inches.
Ores are generally roasted stronger, the more compact,
tenacious, rich in iron, and refractory they are; friable and
easily fusible ores require a slight roasting, whilst siliceous
ores are in some cases refractory and in others easily fusible;
calcareous ores require a higher temperature in order to
expel the carbonic acid, and consequently they consume
more fuel.
Ores containing sulphides are more strongly roasted if
intended for the production of forge iron, and less strongly
if intended for the production of foundry pig. Dusty ores,
preventing a free circulation of air in the roasting furnace,
are more slowly roasted than ores in pieces, allowing a free
passage of the fire, which must be frequently impeded by an
admixture of small ore.
Well roasted ore usually changes its original colour; it is
friable, without any raw kernel, and not caked.
The choice of the roasting method depends partly on the
prices of fuel and labour, but chiefly on the purpose of the
roasting, namely, whether the temperature only is to be used
as a reagent, or the oxygen of the atmospheric air at the same
time. Dependent on these conditions one or other of the
following methods of roasting are used.
1. Roasting in Open Heaps. This method is the
cheapest with regard to labour, and allows a favourable
oxidation owing to the free admission of air, but the fuel
ROASTING IN OPEN HEAPS.
359
has an imperfect effect, and a great deal of it is thus wasted.
More time is also required for roasting, and the resulting ore
is not uniformly roasted, as the interior of the heaps is usually
heated to excess whilst the outer parts are sometimes under-
heated. This method is still employed where fuel is very cheap,
and it is applied to blackband iron ores which contain sufficient
carbon to effect the roasting without an addition of fuel;
these ores are also sometimes roasted in heaps, which do not
require a high temperature.
Open heaps containing iron ores with intermixed fuel are
usually formed as shown by Fig. 65, in which d represents a
layer of large ore pieces, preventing the fuel, a, from becoming
CA
FIG. 65.

C
moist. The layer of fuel, a, consists of wood, small coal,
wood chips, fir cones, &c. Upon this a layer of iron ore, b,
2 or 3 feet high, is given; upon this a layer of fuel, c, from
3 to 6 inches thick, and so on, iron and fuel alternately.
The upper layers of ore are thicker than the lower ones, as
the heat rises to the top, causing a caking of the ore if the
heap has not a pyramidal form, which prevents caking by
cooling from the outer parts.
Heaps, of a longish pyramidal shape, 3 or 4 feet high, 6 or
7 feet broad, and of a corresponding length, are employed if
intending a speedy roasting of an easily roasted ore. The
most suitable dimensions for conical heaps are usually from
15 to 20 feet in diameter and from 6 to 7 feet in height.
Larger heaps may be employed when roasting ores in large
pieces than when roasting small ores, which easily impede
the draught; but the height of the heaps must not exceed
9 feet, otherwise it will be too difficult to pile the ores on the
top of the heap.
360
IRON.
At Königshütte (Hartz), calcareous iron ores are roasted
in heaps in the shape of a truncated pyramid, 60 feet long
on its basis and 9 feet high; they are roasted in from 8 to
14 days. The levelled sole for the heaps is formed of forge
slags, upon which is laid a bed of ironstone 6 inches thick,
next a layer of small coal also 6 inches thick, and then alter-
nate layers of ore and coal in such a manner that the layers
of ore increase in thickness up to 10 inches towards the top,
whilst the layers of coal decrease to 3 inches. One cubic foot
of small coal roasts 3 cubic feet of ore.
At Altenau (Hartz) the heaps are 30 feet long, 15 feet
broad and 8 feet high, containing alternate layers of iron ore
I foot thick and layers of coal 6 inches thick. The heaps
are covered with a layer of small coal 3 inches thick, to pre-
vent too much draught from the outside; calcareous red iron
ores are roasted in this manner in 8 or 10 days.
At Witkowitz,* magnetic iron ores are roasted for 3 or 4
months in heaps about 120 feet long, 30 feet broad, and 6
feet high; 100 lbs. of ore, consuming oʻor cubic foot of wood
in pieces, 0.72 per cent of coal of middling size, and 3*71 per
cent of small coke; one-quarter more of small coal is used
when roasting argillaceous spherosiderites.
Blackband ores are usually roasted without any addition
of fuel, as they contain organic combustible components in
such quantity that when lighted they burn by themselves.
Some sorts of blackband require a small addition of fuel.
The roasting of these ores is facilitated by sifting out the
small ore particles.
The blackband ores in Hasslinghausen (Westphalia), con-
taining from 15 to 30 per cent of carbon, are roasted, without
an addition of fuel, in heaps 3 feet high and of any length
and breadth.
At Heinrichshütte (Westphalia) the blackband ores are
roasted in heaps from 20 to 30 feet broad on the basis, from
10 to 15 feet high, and varying in breadth, for from one to
three months, according to the nature of the ores and the
size of the heaps.
LEOB., Jahrb., 1860, x., 303.
ROASTING IN OPEN HEAPS.
361
The ore pieces for roasting are chosen of a larger size the
more carbon the ores contain. The roasting of these ores is
not without drawbacks, as most of them scorify easily and
always contain a certain amount of sulphur. When roasting
ores rich in carbon, low heaps (2 or 2 feet high) are em-
ployed in order to prevent a caking of the ores, whilst for
roasting ores poor in carbon, heaps 8 or 9 feet high are made,
containing about 10,000 cubic feet of ore. As a rule, the
ores poor in carbon are also poor in sulphur.
Well roasted ores retain their slaty structure and are more
or less red-coloured according to their richness; caked ores
are bluish black and of slag-like appearance, and ores which
have been insufficiently roasted are friable, and disintegrate.
As the presence of reducing gases prevents a complete
oxidation of the sulphur, Grundmann* recommends that the
heaps should be covered with a coating of small ore, and
that this coating, rich in sublimed sulphur, should be removed
after roasting; the heap must then be weathered, i.e., exposed
to air for a long time, and occasionally watered. Grundmann
also recommends piling the blocks of ore with their planes
of stratification vertical to facilitate the escape of sulphur,
as the pyrites contained in the ore is usually found inter-
spersed between the divisional planes.
In order to ignite the heaps they are either surrounded
with a channel I foot broad and deep, filled with wood
which is then lighted and covered with iron ore; or holes
3 feet deep, 3 feet broad, and 3 feet distant from each other,
are made in the heap itself and filled with wood, or some-
times with glowing iron ore.
As the temperature and the draught in open heaps is diffi-
cult to regulate, so that scorification of the ores sometimes.
takes place, and at other times an imperfect roasting of the
ore, in Westphalia heaps of the following construction are
employed in order to obtain a more uniform roasting:—the
heaps are 120 feet long, 30 feet broad, and 4 feet high, en-
closed between walls built up of the larger ore pieces, small
openings of I square foot being left at intervals of 12 feet
GRUNDMANN, d. Entschwefl. der Eisenerze, &c., Hagen, 1855.
362
IRON.
along the sides. These draught holes communicate with
passages 3 feet deep in the interior of the heap, which are
filled with wood. The larger blocks are placed in the middle,
whilst small ore is heaped against the sides of those passages
in order to conduct the flame as much as possible into the
heart of the heap. After the heap is fully ignited, the wall
is pulled down and thrown upon those places where the fire
shows a tendency to come too quickly to the surface, in
order to damp it. A heap of the above dimensions contains
about 17,000 cubic feet of ore, and takes about a month to
burn out.
Blackband containing a large amount of pyrites and coaly
matter will easily raise the temperature so high that the ores
in the inside of the heaps cake, whilst the surface does not
become sufficiently heated. In Westphalia these ores are
therefore roasted in heaps only two feet high; after having
burned out this heap, and before cooling, a second layer 2 feet
in thickness is placed above the first, and in some cases a
third after the second has burned out. The second and
third layers of ore are ignited by the addition of some
pieces of fire-wood, upon which red-hot masses of ore are
shovelled. If the roasting does not proceed uniformly the
hottest places are checked by damping with small ore, whilst
those that are cold and black are started afresh by digging a
hole in the pile and filling it with red hot ore. The loss of
weight varies between 27 and 41 per cent, according to the
amount of carbonaceous matter in the ore.
In Scotland* and Staffordshiret blackband ores are roasted
in heaps from 3 to 9 feet high, resting upon a foundation of
coal, at an expense of 8d. per ton, namely, 4d. for labour
and 4d. for coal.
The roasting in heaps is also employed at Yniscedwin and
Ystalifera in South Wales, but in England and Wales most.
of the ores are roasted in kilns. At Yniscedwin 4 or 5 cwts.
of coal are used for roasting 1 ton of ore; roasting in kilns
costs 24d. per ton of ore, and is much cheaper than roasting
in heaps, which causes an expense of 6d. per ton.
Preuss. Ztschr., iii., Bd. 70.
↑ B. u. h. Ztg., 1862, p. 211. Allgem., B. u. h. Ztg., 1863, p. 21.
ROASTING IN KILns.
363
2. Roasting in Mounds (Stadeln).-Mounds, or walled-
in areas, although they allow the temperature and the air-
draught to be better regulated, and therefore effect a more
perfect roasting than the former method, are inferior to kilns
with regard to expense of fuel and labour, and they also yield
a less uniformly roasted product.
The height of the walls varies between 6 and 12 feet, ac-
cording to the nature of the ore, two ranges of draught
holes about 4 inches square being pierced through them about
3 feet vertical distance apart, the lower series being close to
the level of the ground. Larger mounds are provided with
air shafts inside the walls and communicating with flues on
the sole of the mound. The flues are sometimes replaced
by a foundation of wood, thus giving the air access to the
air shafts.
The mounds used at Ilsenburg (Hartz) are 19 feet 7 inches.
long, 12 feet broad, and 6 feet high.
At Heinrichshütte (Westphalia) blackband ores are roasted
in mounds from 12 to 15 feet wide and high, and 20 feet long,
which are furnished with air holes in the walls and in the
sole; if necessary a small addition of fuel is given.
This method of roasting is not used in this country.
3. Roasting in Furnaces or Kilns. This method of
roasting is generally preferred, as it allows the use of an
inferior fuel and a more perfect control of the temperature,
thus yielding a more uniform product; and though, on the
other hand, the expenses for labour are greater, the total cost
for roasting in kilns is lower than that of roasting in heaps.
The furnaces are mostly so constructed as to allow con-
tinuous operations; when roasted the ore is removed from
the lower part of the furnace, whilst fresh ore is charged at
the furnace mouth.
Rich and easily fusible ores are best roasted in furnaces of
cylindrical or prismatic shape, and furnaces intended for
roasting compact and refractory ores are constructed tapering
towards the top as well as towards the base. When roasting
friable ores the furnace mouth is about two-thirds of the
width of the largest diameter of the furnace shaft, thus
causing the ore to lie in a loose condition.
364
IRON.
Furnaces tapering a great deal towards the top cause an
increase in the temperature at the furnace mouth, occasioning
a waste of fuel, caking of the ore, and an irregular sinking
of the charges; a contraction of the furnace towards its.
base impedes a quick sinking of the charges. It is therefore
advisable when roasting friable ores to have the furnace
cylindrical in its upper part and tapering towards the base.
Furnaces of larger dimensions (above 6 feet diameter) are
frequently constructed in preference to circular furnaces; a
rectangular shape with truncated corners causes a more
uniform temperature and a more regular process, as in
circular furnaces the charges sink more quickly in the middle
than on the edge, where they are retained by the friction of
the walls. The walls also exert a cooling action, so that
easily fusible ores may cake in the centre of the furnace
whilst they are insufficiently heated at the margin.
The height of the furnaces depends chiefly on the size of
the ore pieces and the fusibility of the ore; furnaces 20 feet
high and more, are chosen for roasting ores in large pieces.
Too narrow a furnace easily chokes, whilst too wide a fur-
nace is not uniformly heated. It has proved advantageous
to furnish the top of the furnace with an iron cone to dis-
tribute the ores so that they may lie looser in the centre than
towards the furnace walls.
These roasting furnaces differ chiefly in the manner in
which they are heated, which again causes the reaction to
be more or less oxidising.
These furnaces may therefore be classified into-
A. Furnaces in which the fuel is charged in alternate
layers with the ore.
Ores and fuel are charged in alternate layers, and the tem-
perature is regulated chiefly by the proportional quantity of
fuel charged.
These furnaces make a perfect use of the fuel charged, but
they do not admit of a strongly oxidising reaction, as the
carbonic oxide produced from the fuel prevents it. At the
points where ore and fuel come into contact a caking of the
ore easily takes place, even with ores containing peroxide of
iron, which becomes reduced to magnetic oxide. The ash
FURNACES WITHOUT A GRATE.
365
of the fuel may also contaminate the ore with noxious sub-
stances.
This mode of firing is therefore chiefly adapted for com-
pact ores containing peroxide and free from sulphur, requiring
only a disintegration. Steam may be less advantageously
applied in these furnaces than in furnaces heated by flame,
as the glowing fuel will decompose it, but steam may be still
more effective than atmospheric air alone.
According to the more or less easy combustibility of the
fuel, the furnace is provided with a grate (of different con-
struction) or is without a grate; fuel in the form of gas is
also sometimes employed.
Furnaces without a Grate are used for easily combustible
fuel when roasting ore in large pieces; air is admitted by
the discharging openings.
A furnace of this kind used in Lerbach (Hartz) is repre-
sented by Fig. 66. a is the shaft of the furnace, walled in
with refractory bricks; b is the rough walling; c, the furnace
mouth; d, the discharging openings.
FIG. 66.

ཅན་་་ ་་་་་
d
α
20
When roasting siliceous red iron ores in this furnace
67 cubic feet of small charcoal and 133 bundles of brush-
wood are used for 21 cubic feet, and 5 cubic feet of charcoal
and 8 pieces of brush-wood for 21 cubic feet of calcareous
ore. Two workmen roast daily about 285 cubic feet, and
are paid 3d. per 21 cubic feet. When roasting brown and
clay ironstones the production is increased three-eighths, and
the expense of roasting 3d. per 21 cubic feet.
366
IRON.
The kilns at Dowlais* in South Wales are of a flattened
elliptical plan (rectangular with semicircular ends), being
9 feet wide at the top and contracting to 2 feet in width at
the bottom; they are 20 feet long and 18 feet high. The
floor is made of cast-iron plates 2 inches thick, and the inte-
rior is lined with fire-bricks, with an exterior casing of rough
masonry. Two arched passages slightly bowed outwards
are left in the lower part of the masonry, on one side ex-
tending back to the inner fire-brick lining; this is perforated,
within the space covered by the arches, by four rectangular
openings at the floor level for withdrawing the calcined ore,
as well as by a numerous series of smaller holes above, which
serve for the admission of air. The top edge of the kiln is
covered by a flanged cast-iron ring which protects the brick-
work from abrasion by the lumps of ore when filling.
The method of working is as follows:-Two or three small
coal fires having been lighted on the floor of the kiln, raw
ironstone is placed on the top and around them until the
whole is covered by a layer about 9 inches thick. When
this has attained a dull red heat, a second layer is added with
about 5 per cent by weight of small coal, and so on, fresh
layers of stone being added as soon as the preceding charge
has been heated to redness. When the kiln is completely
filled the lowest portion will be sufficiently cold and fit for
drawing. The capacity of a kiln of the above dimensions
is about 70 tons, and will calcine 146 tons weekly, so that
the average time of burning the charge is about three days
and a-half. The consumption of small coal is at the rate of
I cwt. per ton of ore, whereas in calcining in open heaps
2 cwts. of small and cwt. of large coal are required to do
the same amount of work.
The average loss of weight of Welsh argillaceous ores
when calcined is 27 per cent; of blackband from 40 to 60
per cent; of red hæmatite about 6 per cent; and of Cornish,
Devonian, and similar brown hæmatites from 12 to 14 per
cent.
Gjers's calcining furnace, largely employed in the Cleveland
* BAUERMAN's Treatise on Metallurgy, p. 111.
FURNACES WITH A PLANE GRATE.
367
The
district, is constructed like the newer blast furnaces.
body or shell of the furnace is of fire-brick, only 15 inches
thick, cased with wrought-iron plates. The diameter at the
top is 18 feet, at the boshes or wide part, which is the
junction of two cones, 20 feet, and at the bottom 14 feet; the
horizontal section is everywhere circular. The bottom of
the brick-work rests upon a flat cast-iron plate 4 inches
thick, which is supported by vertical cast-iron columns
27 inches high, leaving an open space all round between the
bottom of the kiln and the floor. The floor is covered in with
a cast-iron plate 20 feet in diameter and 2 inches thick,
cast in segments, carrying in the centre an upright cone
8 feet in height and diameter. The total height from the
foundation plate to the filling gallery at the top is 24 feet,
and the capacity 5500 cubic feet. The ore remains in the
furnace about 2 days, the fuel required being about 1 ton
of coal slack for every 20 tons of ore. The admission of air
from the exterior is regulated by a series of holes penetrating
the brick-work near the bottom, and a further supply is intro-
duced into the centre by means of a series of radiating flues
in the brick-work of the foundation and the hollow in the
overlapping part of the cone. The roasted ore is drawn
through the openings between the pillars, being directed
outwards by the slope of the interior cone. In larger kilns
of similar construction, 34 feet high, the consumption of fuel
is reduced to 1 ton per 25 tons of ore.
Furnaces with a Grate permit a more uniform admission
of air and combustion; they also allow the advantageous use
of a more compact fuel, such as small coke.
These furnaces may again be subdivided according to the
construction of the grate, namely—
a. Furnaces with a Plane Grate.-The grate, of 2 or
2 square feet, is usually constructed of wrought-iron bars
1½ inch broad and inch thick, in two halves, which are sup-
ported in the middle where the halves join by a movable iron.
cross beam, so that the grate bars may be partly or wholly
removed in order to discharge the furnace; but the furnace
is frequently provided with special openings for discharging,
368
IRON.
as the removal of the grate bars is troublesome and inter-
rupts the process.
When starting a furnace, wood chips, coal, &c., are placed
upon the grate and ignited, and the furnace is then filled
with alternate layers of ore and fuel up to half its height.
When the fire has risen so as to appear on the surface of the
topmost layer, the other half of the furnace is filled in the
same manner.
The furnace used at Upper Silesia is represented by Figs.
67 and 68.
FIG. 67.
FIG. 68.

ΑλΘ
e
யர்
10
20
FT
a is the inner walling of the furnace of refractory bricks;
b, the rough walling; c, a space between the two wallings
filled with substances which are non-conductors of heat;
d, the grate; e, the ash-pit; f, the discharging openings;
g, vaults leading to the grate and the discharging openings.
Before starting the furnace, bricks with intermediate spaces
are placed loosely upon the grate to prevent too much draught.
At Gleiwitz and Königshütte (Silesia) 100 lbs. of ore,
causing a loss of 28 per cent, consume 013 cubic foot of
small coal; one charge of ore consists of 75 cwts., and one
charge of fuel of 10 cubic feet.
At Vordernberg the furnace shown by Fig. 69 is used for
roasting the spathic and brown iron ore from the Erzberg.
a is the furnace shaft, 10 feet long and 4 feet broad on the
top; b is a brick wall covered with iron plates, dividing the
FURNACES WITH PLANE GRATES.
369
lower part of the furnace into two divisions, each provided
with a grate, c, 4 feet 3 inches long and 4 feet broad, which
rests upon three cross beams, d.
་་
•
FIG. 69.

19
a
€
€
C
đ
f
f
யர்
3
T
6.
9
12
FEET
Two or three grate bars are removed when the well-
roasted ore has cooled somewhat in the space, e, allowing
it to fall into the cooling place, f, which is provided with a
sole inclining 45°. Before removing the ore it is moistened
with water in order to cool it still more and to prevent its
dusting; g are levers for lifting the doors, h, with which the
furnace mouth is closed.
One charge of ore consists of 78 cwts. and a charge of fuel
of from 31 to 34 cubic feet of charcoal. From three to six
charges pass the furnace in fourteen hours according to the
nature of the ore, whilst during this time the grate is three
times opened. Besides these furnaces others are used in Styria*
TUNNER'S Leob. Jahrb., 1842, p. 136. Oesterr. Ztschr., 1855, p. 126.
VOL. II.
2 B
370
IRON.
of the shape of a four-sided truncated pyramid, and from 8 to
12 feet high; they are 5 feet long at the top, and from 3 to
4 feet on the base. They consume 3 or 4 lbs. of small coal
for roasting I cwt. of spathic iron ore, which loses in weight
19 or 20 per cent. The grate is placed 4 feet above the sole,
and the grate rods are provided with handles.
*
In Siegen spathic iron ores are roasted in furnaces 17 feet
high, 6 and 3 feet wide respectively at the top and at the
bottom, and provided with two grates, the second one lying
2 feet below the upper one. The rods of the latter grate lie
3 inches apart, and are covered one hand high with large
charcoal, upon which a layer of ironstone is placed. On this
FIG. 70.

a
B
A
7
4-
8
IF
12
layers of ore and cinders are alternately charged. A wood
fire is sustained on the lower grate for three days, thus
粼
​CARNALL, Preuss. Ztschr., iii., A. 186. Allgem., B. u. h. Ztg., 1860, p. 28.
FURNACES WITH PLANE GRATES.
371
igniting the fuel above the upper grate. The well-roasted ore
is discharged by removing the rods of the upper grate, and
FIG. 71.

* *
Ъ
fresh ore is charged anew. Thus the roasting process is
carried on for months without interruption, yielding daily
about 100 cubic feet of well-roasted ore, and saving 5'4d. per
7 1-9th cubic feet of ore when compared with the former
method of roasting in open heaps.
The roasting furnaces used in Siegen are constructed by
Jhnet so as to serve as steam roasting furnaces; their con-
struction is shown in Figs. 70 and 71.
Steam is admitted by the iron tube, b, which conducts it
into a hollow iron cone, a, 18 or 20 inches high, and 18 or
20 inches in diameter on its base. This cone is pierced with
numerous small holes which keep open owing to their
small size and the power of the escaping steam. It is
supported by three bars, c, which rest upon the castings, d,
and is placed at such a height from the furnace that the
steam will react most advantageously. ƒ are openings for
admitting air. These furnaces are more effective for decom-
posing sulphides, but flame cupola furnaces are preferable.
· At Seraing (Belgium) 200 cwts. of ores are roasted in
24 hours, at a consumption of 72 cubic feet of coal and the
same quantity of small coal.
*
Schles. Wochenschr., 1861, p. 276. HARTMANN, Fortschr., v., 90.
2 B 2
372
IRON.
At some French iron works from 300 to 400 cwts. of ore
are roasted in 24 hours, consuming from 4 to 5 lbs. of coal
per cwt.
b. Furnaces with a Grate of Conical Shape.-One of
these furnaces, used at Neudeck*, in Bohemia, is represented
by Fig. 72. a is the conical grate standing above an opening
FIG. 72.

Ъ
d
10
15
J FEET
of the furnace sole formed of iron plates; the rods forming
the grate are held together by a sort of hood, which is sup-
ported by a fixed iron rod. b is the furnace shaft ; c are dis-
charging openings; d is a space through which the air enters.
When putting this furnace into operation the grate is
covered to within I foot above its point with 30 cubic feet of
small coal; upon this 220 cubic feet of ore are charged,
next 24 cubic feet of small coal, then 200 cubic feet of ore
and 20 cubic feet of small coal, and, finally, 200 cubic feet of
After lighting it, the discharging openings are closed,
and the furnace is wholly emptied in 48 hours, when it is
again filled.
ore.
c. Furnaces with Step Grates.-These furnaces have
given favourable results at Mariazell; they are shown in
* RITTINGER's Erfahr., 1860, p. 38. B. u. h. Ztg., 1860, p. 103; 1861, p. 56.
Oesterr. Ztschr., 1859, No. 32.
FURNACE WITH STEP GRATE.
373
Fig. 73. a is the cast-iron step-grate; b, the discharging
openings; c, inclined plane on which the roasted ores are
removed; d, rough walling; e, interior walling; f, space
between the two wallings filled up with ash, &c.
FIG. 73.

..
d
e
a
α
e
d

The furnace is capable of containing Soo cwts. of spathic
iron ore; the charges consist of 100 cwts. of ore and 23 cubic
feet of small coal. From 80 to 100 cubic feet of ore are
roasted in 24 hours, consuming about 2 cubic feet of small
coal per cwt. of roasted ore.
In roasting spathic ores containing iron pyrites it is neces-
sary to provide as much as possible for the free access of air,
in order to oxidise the sulphur, which is partly volatilised as
sulphurous acid, the remainder forming sulphate of iron.
At Mariazell in Styria, Wagner has constructed for this
purpose an annular kiln, the exterior of which consists of a
series of flat cast-iron rings, 1 inch thick and 9 feet in
internal diameter, placed one above another, but kept apart
by piers formed of bricks. In the centre of the enclosed
space is placed a cylindrical brick chimney 2 feet in diameter,
with six holes 3 inches square for regulating the draught. A
374
IRON.
cast-iron water-pipe provided with numerous small jets sur-
rounds the kiln at the level of the lowest ring.
A less expensive construction, based upon the same prin-
ciple, and employed at Gollrath (Mariazell), is shown by
Fig. 74.
A, A, are two spaces, each 63 feet long, separated from
each other longitudinally by the partition wall, a, and
divided in their lower parts into ten divisions, each 4 feet
long, by the sloping walls, b. Each of these divisions is
FIG. 74.

1
TH
་་མ
Q
h
h
A
Ъ
Ъ
d
m
m
d
C
7
ཁང་བ་་ ་བ་
Z
с
....
N
wi
5
10
15
FI
provided on its sole with an inclined iron plate, c. The hot
iron ore gliding down on this plate is wetted with water by
means of the tube, d. e is a hollow space running along
the longitudinal sides of the furnace, communicating with
the atmosphere by the channels, f, and conducting the air
into the furnace by the openings, g. h are air-holes on the
:
FURNACES WITH AN INTERIOR FIRE-PLACE.
375
smaller sides of the furnace. i are openings in the partition
wall, communicating with the air by the channels, k and l.
m are iron plates, and ʼn are water reservoirs.
B. Furnaces Heated by Flame.-These furnaces have
the following advantages over the preceding furnaces:-A
stronger oxidation when roasting ores containing sulphur
and protoxide of iron; the better regulation of the tempera-
ture, whereby the ores are less easily scorified; the possibility
of using a sulphurous fuel without contaminating the roasted
On
product; and a more effective application of steam.
the other hand, these furnaces consume more fuel (from
one-half to twice the quantity), which must burn with a long
flame; they require a larger stock of castings and more
frequent repairs, and are less adapted for roasting pulverulent
ores, which impede the draught.
This last disadvantage may be lessened by enlarging the
lower parts of the furnaces. The size of the grate depends
on the quality of the fuel; too large a grate wastes fuel, and
too small a one does not produce sufficient flame.
These furnaces vary in their construction according to
whether the fire-place is placed on the sides of the furnace
or in its inside, and whether or not steam is employed.
a. Furnaces having the Fire-place on one side.-These
furnaces are only in use in very few places, as the flame has
a tendency to rise on the furnace walls, particularly if the
furnaces are wide and the ore is small.
b. Furnaces with an Interior Fire-place.-These fur-
naces, which permit a uniform division of the heat, were
first employed in Norway and Sweden. They are sometimes
constructed so as to allow the admission of steam. This
latter furnace was introduced into Finland and the Ural
Mountains by Nordenskjöld, in 1845, for roasting iron ores
containing sulphur. They are shown in Figs. 75 and 76.
a is the furnace sole; b, the grate; c, dome formed of iron
bars; d, ash pit; e, discharging openings; f, openings for
admitting air; g, fire-place; h, iron tube, pierced with many
holes for admitting steam into the furnace; i, dome pro-
tecting the steam tube.
376
IRON.
FIG. 75.

பர்
10
5
JFT
They were introduced later into Silesia* and Hördet
(Westphalia) for roasting blackband ores rich in sulphur;
one furnace roasts daily about 350 cubic feet of ore.
FIG. 76.

9
d
a
f
When roasting blackband either by itself (Hörde) or
together with magnetic iron ore (Vorwärtshütte in Silesia) a
grate firing does not take place, though Nordenskjöld's con-
struction of the furnace has been adopted; the carbon
contained in the ore serves as fuel. The grate of the fur-
nace is then chiefly used to admit air.
The furnace employed at Vorwärtshütte, near Hermsdorf
in Silesia, is 25 feet high, 6 feet wide on the furnace mouth,
Allgem., B. u. h. Ztg., 1861, p. 232. Schles. Wochenschr., 1859, No. 18.
HARTMANN, Fortschr., v., 106.
*
† Berggeist, 1861, p. 536.
ROASTING FURNACES HEATED WITH GAS.
377
increasing in width down to the lower cylindrical part of the
furnace, 9 feet in diameter, and 6 feet high. When starting
the furnace, a layer of blackband rich in carbon and 4 feet
thick is first charged, next wood chips, coal, blackband, coal
and wood chips, then magnetic iron ore, and afterwards
alternate layers of blackband and wood chips. After filling
the furnace it is ignited from below, and alternate layers of
blackband and magnetic iron ore are afterwards added when
the contents of the furnace have sunk to half of the height of
the furnace.
When the furnace is fully heated, superheated steam of
about 36 lbs. pressure is admitted every half-hour for some
minutes, and atmospheric air is admitted at the same time
in order to decompose the sulphuretted hydrogen into water
and sulphurous acid; the sulphuretted hydrogen would
otherwise again form sulphide of iron in the upper part of
the furnace. The raw ores contain from 1 to 3 per cent of
sulphur, which is almost completely removed by this roasting.
When roasting without the application of steam only half of
the sulphur can be removed, the bisulphide of iron becoming
transformed into sulphide only. 250 cubic feet of magnetic
iron ore and 76 cubic feet of blackband are roasted daily.
c. Roasting Furnaces Heated with Gas.-When applying
the gas from iron blast furnaces to roasting iron ores no
other fuel is required, and the roasting is more perfect and
uniform. The low but uniform temperature required for
roasting cannot be obtained either in furnaces with inter-
mixed fuel or in furnaces heated by a common flame firing,
as the draught and consequently the heat always follows the
larger interstices formed by the ore, causing a caking.
The roasting can only be rendered more perfect by applying
more fuel and by retarding the process. The combustible
gases, on the other hand, may be uniformly distributed in
the furnace, and the temperature may be regulated by the
admission of air.
These gas roasting furnaces are differently constructed.
Those at present used at Dannemora are represented by
Figs. 77 and 78.
The gases from the blast furnace are conducted by a tube
378
IRON.
of iron plate 120 feet long and 10 inches wide into the tube
surrounding the roasting furnace, by means of the adjoined
tube, a. The tube surrounding the roasting furnace is pro-
FIG. 77.

་
fd
i
e
с
C
B
e
d
e
d
d
f
Ch
་།་ར།
པ་ཅ་ཐོག་་ས་ལྷས་
C
Y
0
12 FT
vided with openings, b, for cleaning it, and with 10 conduit
tubes, c, from which the gases enter the cast-iron boxes, d,
and then the furnace by means of the walled channels, e;
ƒ are valves for regulating the current of gas; the doors, g,
give access to the valves, and are provided with openings for
the observation of the temperature of the furnace. The
openings, h, serve the same purpose, as well as for intro-
ducing iron rakes into the furnace, if the sinking of the iron.
ore should necessitate such assistance; i are discharging
openings; k are bearing irons.
As this furnace has proportionally large dimensions, and
the gases enter the furnace on its periphery, there is a danger
that a great deal of the gases will burn on the inner wallings
and not reach the heart of the furnace.
ROASTING FURNACES HEATED WITH GAS.
379
This disadvantage is avoided at Vordernberg, where ten
small roasting furnaces are laid one beside another, and also
in the furnace which is used at Hof and represented by Figs.
79, 80, and 81.
FIG. 78.

D
h
d
s
h hh
€
h
0
-E
I
a is a gas tube provided with 16 nozzles, h; the eight middle
ones are 1 inches wide, and the other eight from 1 to 1
inches. b is a lid for cleaning the tube, and serving at the
same time as a safety valve; c are cast-iron bearings upon
which the gas tube rests; the tube is also supported by the
iron plates, d, standing inclined at an angle of 45°, and which
are kept in proper position by the gutters, e, in which
they rest; f is a channel for admitting air entering the
furnace by the openings, g, of the iron plates, d; i, a cast-iron
380
IRON.
dome protecting the gas tube and the nozzles; k, are dis-
charging openings; I, bearers.
Before starting the furnace it is warmed for some days by
allowing the gases to pass through it whilst empty. The
openings, k, are then closed by piling ironstone up to 1, when
some wood chips, coal and ore, and some charges of small
FIG. 79.
Ci

G
h
H
h
d
¡D
10
e
k
15
1
20
JFT
coke and ore are added. The wood chips are now ignited, and
the gases are conducted into the furnace when the coal
and coke are sufficiently burning. The furnace is then
gradually filled with ore, and the admission of gas regulated
by a sliding door.
Weathering the Roasted Iron Ores.-It has before
been mentioned that raw ores are sometimes exposed to the
influence of the atmospheric air and water, but this process
of weathering is more effective when applied to roasted ores,
WEATHERING THE ROASTED IRON ORES.
381
and is therefore more frequently applied to them. For this
purpose it is advisable to quench in water the red-hot
ores when taken out of the furnace, and then to break
them.
FIG. 80.

၁၁၁၄ဝဝ၁၀၀၀၀၀၈၀၁၀
α
JO
g
F
f
d
This process aims chiefly at the decomposition of sulphide
of iron, which has been reduced by the roasting to a lower
sulphide and made less compact; the sulphide is converted
into soluble sulphate by the reaction of air and water, and
lixiviated. Arsenical pyrites is also more inclined to de-
composition after roasting; it is transformed into soluble
sulphate of iron and arsenious acid. Copper pyrites is like-
wise easily decomposed after roasting. If lime is present
at the same time with such sulphides, gypsum will be
formed at the roasting process, thus facilitating the removal
of sulphur.
Roasted calcareous ores quickly disintegrate, and are
therefore weathered for a short time only, but siliceous ores
382
IRON.
containing sulphides sometimes require some years for
weathering and lixiviation.
Breaking up the Iron Ores.-The size of the ore pieces
to be smelted depends chiefly on the reducibility of the ore.
and the height of the furnace. A high furnace, smelting
easily reducible ores, may therefore smelt the ore in large
pieces up to 8 cubic inches, whilst the ore must be reduced
to I cubic inch when smelting, in low furnaces, ores difficult to
reduce; magnetic iron ore is usually still more broken up.
Ι
FIG. 81.

市​公
​d
B
0.0.0.0.0.0.0.0.0.0.0.0
០០០.០០០.០
-D
A
Charcoal blast furnaces usually smelt ore of the size of a
nut; coke furnaces use a larger size. Hard coal lying loose
in the furnace allows the use of smaller ore pieces than
friable coal which impedes the rising of the hot gases. A
uniform size of the ore pieces facilitates the regularity of the
process. Easily fusible ores in too large pieces are liable
to melt before they are reduced; puddling and other iron
slags must therefore be used in a finely divided state.
The ore must not be so much crushed that it takes the
form of fine powder, as it then lies too compactly in the fur-
nace, causing the rising gases to escape along the furnace
BREAKING UP AND MIXING THE ORES.
383
walls and even through the tuyeres and below the tymp;
they then corrode the tuyeres prematurely, and escape from
the furnace mouth without having sufficiently reacted; the
ores pass into the furnace hearth unprepared for it, as the
different levels of the furnace have a lower and less uniform
temperature; they cause the escape of ore dust from the
furnace mouth, &c.
The result will then be :-A large consumption of fuel, an
irregular process, and a quick wasting of the inner furnace
lining.
If at the breaking up of the ore much ore results in the
form of powder it is advisable to thoroughly moisten the
ore mixture; puddling slags in form of powder are some-
times formed into bricks, being mixed with lime. The
disadvantages of finely divided ore may also be partly
avoided by a suitable construction of the blast furnace;
friable ores behave like powdered ore.
The size of the ore pieces has also an essential influence
on the amount of silicon, sulphur, and phosphorus in the
resulting pig-iron. Finely divided ore is usually only imper-
fectly reduced and carbonised, and it yields a pig-iron poor
in carbon and more inclined to combine with the above sub-
stances than iron saturated with carbon.
The different methods of breaking up the iron ores have
been mentioned on page 352.
Mixing the Iron Ores.-In order to produce from iron ores,
iron of a certain quality and at the lowest possible consumption
of fuel, the earthy components of the ore must form a fusible
slag at the smelting temperature of the pig-iron, and the
quantity of resulting iron must stand in a certain proportion.
to the produced slag, which has to protect the iron from the
further influence of the blast.
These conditions are seldom obtained from iron ores,
which, therefore, require a suitable admixture, namely, a
mixing of rich with poor iron ores, so that the mixture con-
tains a certain average amount of iron which experience has
proved to be most advantageous as regards the yield and the
quality of the iron; at the same time the admixture of ores
which contain different earths, and assist each other at the
384
IRON.
smelting process, is aimed at. When smelting ores too rich
in iron the quantity of slag resulting is not sufficient to pro-
tect the iron before the tuyere and in the furnace hearth from
the influence of the blast and the carbonic acid formed, the
iron will be decarbonised and scorified; and more so the
stronger the blast is. Therefore, a larger quantity of slag
is usually required when smelting with coke a mixture diffi-
cult to fuse, than when smelting an easily fusible mixture
with charcoal; the production of grey iron also requires
more slag than that of white iron. But as the smelting of
rich mixtures is more economical as regards fuel, the mixtures
smelted are always as rich as possible.*
When smelting very poor ores too much slag will be pro-
duced, causing a greater loss of iron, as the iron separates
with difficulty from the great mass of slag, the consumption
of fuel is increased, and an excess of siliceous slag may
react upon the iron, extracting some of its carbon and re-
placing it by silicon. It is therefore advisable to produce
basic slag if the pressure of the blast requires the presence
of a great quantity of slag.
The average amount of iron to be aimed at by mixing
different iron ores, depends chiefly on the amount of iron
contained in the ores. The amount of iron in mixtures of
ores difficult to fuse and reduce usually varies between 30 and
40 per cent, and it is increased to 50 per cent, and in rare
cases even to more, when the ores are easily fused and
reduced.
This average amount also depends on the quality of the
fuel and the fusibilty of the ores.
Some few ores may be smelted by themselves, as they
contain in sufficient quantity the components required for
the formation of slag.
In other cases, siliceous, argillaceous and calcareous ores
must be used in sufficient quantity to be suitably admixed
for the formation of slag. But in most cases the ore mixture
contains an excess either of silica or lime, usually silica, and
Bgwkfd., x., 187. B. u. h. Ztg., 1862, p. 390.
MIXING THE IRON ORES.
385
rich ores do not contain the earths required for the formation
of slag in nearly sufficient quantity.
The components which are then wanting must be added
to the ore mixture.
The normal process of the blast furnace, and its advantages
with regard to the consumption of fuel and yield of iron,
depend on the judicious choice of the fluxes. The iron
smelter ought, therefore, to have an exact knowledge of the
composition of his iron ores, fluxes, and fuel, and of the pro-
perties of the silicates which especially influence the process,
and of the laws which regulate chemical combinations.
When forming a mixture the following points are to be
taken into consideration, viz. :-the fusibility of the intended
slag, the quality of the fuel, the pressure and temperature
of the blast, the reducibility of the iron ores, and the quality
of the pig-iron to be produced.
1. Mixing the Iron Ores with regard to the
Formation of Slag.
The earths chosen must generally be of such a composition
that the melting point of the slag formed is lower than the
highest temperature of the furnace, and more or less.
approaches the melting point of pig-iron (1400° to 1600°).
In general, the formation of slag must take place at the
same time that the carbonised iron is fused; otherwise, pro-
toxide of iron enters the easily fusible slag, or imperfectly
carbonised iron is produced. In some cases it may be
necessary to form slags difficult to fuse, and, in others, slags,
easy to fuse. The lower the melting point of the slag and pig-
iron is below the highest temperature of the furnace, the
larger ore charges may be given, and the smaller will be the
relative consumption of fuel.
The formation of white iron will be facilitated if the tem-
perature above the tuyere is not much higher than that re-
quired for melting the slag and pig-iron, and if these products
pass quickly to the hottest point of the furnace; on the other
hand, grey iron will be formed if the fusion point and the
highest temperature of the furnace differ much, and if the
melted products slowly pass into the furnace hearth.
VOL. II.
2 C
386
IRON.
The temperature of the blast furnace is increased, the more
compact the fuel and the hotter the blast of the required
pressure. Scheerer* has calculated the highest temperature
(P) of a charcoal furnace at different temperatures of the
blast (t):
t.
0° C.
100°
P.
2656° C.
2758
150°
39
2809°
200°
2860
""
250°
2911°
300°
""
2962°
350°
3032°
وو
400°
وو
3064°
,,
Lindauer,† supposing that the carbon at its combustion
forms one-third carbonic acid and two-thirds carbonic oxide,
calculated the highest temperature of a blast furnace as
follows:-
t.
0° C.
50°
100°
P.
2346° C.
2395°
2445
""
2494,
2543
150
""
200°
""
2592
2642°
2740°
""
250
300°
350°,,
Tunner‡ estimates the temperature before the tuyere of
a charcoal blast furnace carried on with coal blast to be
higher than 1900° C. and less than 2500° C.
According to Aubel,|| platinum (2534° C.) may be fused at
the hottest point of an iron blast furnace. (Carbide and silicide
of platinum fuse at a lower temperature than pure platinum;
the formation of these must therefore be avoided when
ascertaining the temperature).
When forming mixtures the formation of double silicates
* SCHEERER'S Met., ii., 21. Bgwkfd., vii., 417.
† LINDAUER, Hüttenchemie, 1861, p. 299.
LEOB., Jahrb., 1860, ix., 295. B. u. h. Ztg., 1860, p. 208.
|| DINGL., Bd. 165, p. 278. B. u. h. Ztg., 1862, p. 392.
MIXING THE IRON ORES.
387
of lime and alumina is aimed at; they constitute the slag
either as mono- or bi-silicates. They are formed at a tem-
perature of 1800° or 1900° C., and experience has shown their
melting point to be below the highest temperature of the
blast furnace.
These silicates have the following composition:-
Bi-silicate.
Mono-silicate.
3CaO,2SiO3+Al2O3,2SiO3. 3CaO,SiO3+Al2O3,SiO3.
Silica
Lime
Alumina
57°23
26.53
16.24
40°09
37°17
22.74
According to Bodemann, the following combination—
56 silica, containing 24°9 oxygen
30 lime,
14 alumina, „
8.6
"
8.9
""
is of the greatest fusibility, and corresponds to the formula-
4(3CaO,2SiO3)+3(Al2O3,2SiO3).
Each modification in the proportion of the different earths
contained in Bodemann's silicate (whose fusion point corres-
ponds to the temperature required for the formation of a
mottled or light grey iron), renders it more difficult to fuse,
increases the consumption of fuel, and causes the grey iron
produced to be of darker shade. But certain circumstances,
such as the presence of sulphur in the ores and fuel, hot
blast, &c., may necessitate the formation of more basic slags,
richer in lime, and therefore more difficult to fuse. An in-
crease of silica or alumina also renders the slags less fusible,
but in most cases this effect is obtained by an increased addi-
tion of lime, as a greater amount of silica gives rise to the
scorification of iron and to the formation of pig-iron rich in
silicon and sulphur. As alumina reacts both as a base and
as an acid, a small surplus of it may be advantageous, but
if much exceeding 15 per cent, the slags are very difficult to
fuse and dark grey iron is produced. 110 parts of alumina
contain as much oxygen as 100 parts of silica, and the action
of the alumina will be doubtful if present in nearly the
same amount as the silica. In order to avoid the formation
of the more doubtful alumina slags an increased addition of
silica is frequently made, thus decreasing the proportional
2 C 2
388
IRON.
amount of alumina, and forming a bi-silicate, if both silica
and alumina are calculated as acids.
When smelting blackbands* rich in alumina, the resulting
slags are mostly rich in alumina, containing as much as
26 per cent; and their slight fusibility, which is the result of
the large amount of alumina, is sometimes (Hattingen in
Westphalia, Scotland) neutralised by an amount of man-
ganese which the ores contain at the same time. The amount
of alumina in slags of coke blast furnaces seldom exceeds
15 per cent.
In rare cases only is lime wholly wanting in blast furnace
slags, but when this is the case, the lime is replaced by other
earthy or metallic bases, chiefly by oxide of manganese.
If part of the lime in slag is replaced by magnesia the
fusibility of the slag is decreased.
When intending to produce white iron requiring a more
easily fusible slag than those shown by the composition
ascribed by Bodemann, some of the lime is replaced by oxide
of manganese, without much altering the proportional amount
of silica in the slag. Sometimes silica is decreased so as to
form slags standing between mono- and bi-silicates (Vordern-
berg, Gittelde in the Hartz); the slag obtained when smelting
spiegeleisen is a mono-silicate. In rare cases only the oxide
of manganese is intentionally replaced by protoxide of iron,
as this would cause a loss of iron, and the protoxide of iron
would decarbonise iron that had been already carbonised, and
would also strongly attack the furnace lining.
Slags obtained at the production of white forge iron
usually contain more or less iron; and the green blackish
slags resulting from an irregular process contain from 10 to
20 per cent of protoxide of iron; whilst slags of the regular
process usually contain from 1 to 4 per cent of protoxide.
This small loss of iron is not to be avoided, even by devoting
the utmost care to the composition of the mixture, and
it is increased by grains of iron sticking to the slags,
causing a mechanical loss.
The fusibility of the admixture is also increased by a larger
* LEOB., Jahrb., 1861, x., 391. B. u. h. Ztg., 1858, p. 231; 1862, p. 323.
SLAGS FROM CHARCOAL FURNACES.
389
number of bases; and the presence of alkali contained either
in the ash of the charcoal, in the slaty gangue of the mineral
coal, or in some blackbands, reacts most favourably upon
the smelting process.
Important investigations concerning the nature of blast
furnace slags have been made by Rammelsberg,* Haus-
mann,† and Scheerer.‡
These slags differ greatly, owing to the great variety of
iron ores and of the associated minerals. Some of the slags
correspond, in point of chemical composition and crystalli-
sation, with certain native minerals, such as pyroxene
(Wollastonite, Diopside), Gehlenite, Humboldtite, felspar,
peridote, emerald, chytostilbite, amphibole, &c.
The state of aggregation of the blast furnace slags depends
on their chemical composition, on the process, on the time
of smelting, on the period of the operation at which the
slag resulted, and on accidental circumstances, such as the
rapidity of the cooling of the slags on their coming into con-
tact with water and other substances, &c. The state of aggre-
gation forms the following chief varieties of slags :-
Crystalline slags; porcelain-like and stony slags; glass-
like slag, porphyry-like slag; sometimes they are full of
blisters and air-holes, and also asbestiform and capillary.
The specific gravity of the slags is greater the more
quickly they are cooled, but they are less hard.
The chemical composition of some amorphous blast
furnace slags is shown in the following analyses:-
SiO 3
Al₂03
CaO
MgO.
FeO.
MnO
S.
1. Slags from Charcoal Blast Furnaces.
I.
II.
III.
IV.
70°23
70°12
600
60°44
6'37
6.25
7.4
3°27
20'41
19'71
20.6
19'73
0'70
7.2
7'01
0'15
I'45
3°0
4.89
2.70
I'40
3.6
4'28
0°36
* POGGEND., Ann., 1., xxiv., 95. Bgwkfd., xii., 545.
↑ Studien d. Götting. Ver., vi., 323 (1854). Nachrichten v. d. k. Gesellsch.
der Wissensch. in Gottingen, 1856, No. 12. LEONH. Hüttenerzeugn., p. 55.
+
Ann. d. Chem. u. Pharm., Bd. 94, p. 79
KOCH., Krystall. Hüttenproducte, p. 40.
(1855).
GURLT's Pyrog. Min., p. So.
§ RAMMELSBERG, Metallurgie, p. 89. KARSTEN'S Eisenhüttenkunde, iii., p. 218.
390
IRON.
V.
VI.
VII.
VIII.
SiO3.
53'0
52.8
59°42
53'79
A1203
I'O
3'4
14'94
13.04
CaO
15'0
5'6
19'79
25.67
MgO
8.0
9°0
O'II
0'57
FeO
ΙΟ Ο
I'4
6.03
2.44
MnO
ΙΟ Ο
26°2
trace
2.20
IX.
X.
XI.
XII.
SiO3.
49'57
49'0
46.371
49'70
Al2O3
9'00
21.8
4'301
9'10
CaO.
24'0
38.640
16.20
MgO
15°15 trace
7'400
10*30
FeO.
0*14
2'I
0*950
3970
MnO
25.84
0.6
1.860
10.80
KO
0*7
0*089
NaO
0°7
0*138
CuO
trace
POS
S.
trace
2.60
5
Loss
SiO3
A120,
CaO
MgO
FeO
}
trace
0°030
0°40
0°30
XIII.
XIV.
XV.
XVI.
31'1
37.8
27.48
29°0
8.9
2'I
25'78
6.3
3
14'1
25'47
23°4
34°2
8.6
0'41
15'7
ΙΟ
21'5
o'91
23°7
MnO
TiO2
PO5
S.
4.4
22
2'59
I'4
6.70
9.66
9'0
No. 1. A bluish white, enamel-like slag, resulting at the
production of grey pig-iron from spherosiderites, analysed
by Klasek. Formula of the slag:
11(3RO,5SiO3)+5(Al2O3,2SiO3).
No. 2. Slag from Peitz, bluish white, vitreous, pellucid at
the edges, produced from bog iron ore. According to Kar-
sten:—11(3RO,,5SiO3)+5(Al2O3,2SiO3).
No. 3. A vitreous bluish grey slag with blue stripes, from
Framont, produced from red and brown iron ores; by
Berthier:-3(RO,SiO3) + Al2O3,3SiO3.
No. 4. A pellucid, bluish or greenish grey, enamel-like
slag, from Finspang; it resulted at the production of pig-
iron for cannons; proportion of oxygen, 31°27 : 11·82.
SLAGS FROM CHARCOAL FURNACES.
391
No. 5. A blistered, black slag, from St. Helena in Savoy,
produced from spathic iron ore at an irregular process;
analysed by Berthier.
No. 6. A well-fused slag, partly stony and partly vitreous;
it resulted from spathic iron ore at the production of spiegel-
eisen; by Berthier. Formula :-
7(3RO,2SiO3)+Al2O3,3SiO3.
No. 7. Black slag from Steinrenne (Hartz), produced at
an irregular process; analysed by Bodemann.
No. 8. A vitreous, green slag, from Rübeland (Hartz); it
was formed at the production of mottled iron; analysed by
Rammelsberg :—3(3RO,2SiO3+2(Al2O3,2SiO3).
No. 9. Slag from Hammhütte, produced at a regular pro-
cess; by Karsten. Formula:-3(3RO,2SiO3)+Al2O3,SiO3).
No. 10. A darkish green and violet blue vitreous slag,
from Bley (Depart. Haut-Saône); it is difficult to fuse, has
graphite separated on its surface, and it was formed at the
production of grey pig; by Drouat. Formula :
3(3RO,2SiO3)+A₁₂O3SiO3.
No. II. Slag obtained from Dannemora mixture in Sweden
at the production of Bessemer pig-iron; proportion of oxygen,
24'077: 16'718.
No. 12. Slag from Dalekarlien, produced from phosphatic
ores.*
No. 13. Yellow, blistered slag, difficult to fuse, from Ekers-
holm in Smaland, produced from Taberg magnetic iron ore;
analysed by Zachrisson. Formula :
3(9RO,2SiO3)+2(Al₂O3,SiO3).
No. 14. Slag obtained from spathic and brown iron ore,
from Siegen, at an irregular process, producing white pig
poor in carbon.
No. 15. Pumice-like slag, from Concordiahütte; by Hess.
No. 16. Black, blistered slag, from Lietzen, in Styria,
produced at an irregular process; analysed by Mayrhofer.
B. u. h. Ztg., 1857, pp. 262, 374. Oesterr. Ztschr., 1860, No. 47.
392
IRON.
2. Slags from Coke Blast Furnaces.
I.
II.
III.
IV.
Sio,.
50'00
46.60
46.46
39'95
Al₂O
3
23°00
15.80
18.80
17.41
CaO.
27'00
10.80
25'60
29'64
MgO
2:28
3'50
6'47
FeO.
7:56 ·
0*24
MnO
3°40
2.80
O'91
KO .
2'40
I'46
S.
0'70
CaS.
11'70
3*60
PO
trace
V.
VI.
VII.
VIII.
SiO3 ·
40*20 41.64
42'94
45.64
A1₂O
3
16.45
13°20
16.29
10.84
CaO.
30'00
35'91
31 10
35'ΟΙ
MgO
7.29
4'21
4°16
3°16
FeO.
0'57
ΟΙΙ
0°34
0'71
MnO
0*84
0'74
0'51
trace
KO .
I'30
I'70
1.87
0.82
S.
CaS.
2'71
2.19
2.16
3°30
PO5 ·
trace
trace
trace
trace
IX.
X.
XI.
XII.
Sio,
4III
4I'II
37.84
38.80
A120
3
9'46
13'45
13°20
15'20
CaO.
37'90
29.82
20.68
37'00
MgO
2.II
4'75
2.93
3°20
FeO.
0'49
6.44
20.83
4'40
MnO
1.61
o'66
0.80
KO .
0'71
1.84
I'08
S.
0.80
CaS.
0'41
I'34
0'87
PO5
trace
•
0'15
I'77
XIII.
XIV.
XV.
XVI.
Sio,
A1203
34°25
34.90
28.80
20°02
16.70
5'80 12.30
30°00
CaO.
38.81
50*63
56°30 39'06
MgO
I'53
0'93
0'55
4:38
FeO.
0.84
6'52
0'70
3.86
MnO
1'51
I'04
0'21
KO .
I'21
trace
S.
0'08
I'15
CaS
5'48
C. 0°50
I'43
PO₁
trace
P. O'10
SLAGS FROM COKE BLAST FURNACES.
393
No. 1. Medium composition of slags resulting in England
and Belgium at the production of foundry pig with hot blast;
analysed by Mayrhofer.
No. 2. Slag from Séraing, obtained at the production of
forge pig-iron, analysed by Schmidt :-
6RO,9(SiO,,Al2O3)=9(3RO,2SiO3)+4(2Al2O3,3SiO3).
No. 3. Normal slag from Gleiwitz.
Nos. 4-11. Slags from Staffordshire, analysed by Wright-
son, in order to investigate the influence of hot blast upon the
phosphorus contained in pig-iron.*
Nos. 4 and 5. Slags produced at the production of grey
iron with cold blast.
Nos. 6 and 7. Slags obtained at the production of grey
iron with hot blast.
No. 8. Slag obtained at the production of grey iron with
hot blast from oolitic ore.
No. 9. The same from puddling slags.
No. 10. Slag, thin, liquid, and in appearance like black
bottle glass, obtained at the production of white pig-iron.
No. II. The same from oolitic ore at a very irregular pro-
cess.
No. 12. Dirty brown slag, glistening on its fracture, from
Chanon St. Etienne, obtained at a regular process at the
production of spiegeleisen; by Berthier:-
3RO,2(SiO,,Al₂O3)=9(3RO,SiO3)+5A1₂O,,SiO3.
No. 13. Slag from Heinrichshütte, near Hattingen in
Westphalia, obtained at the production of grey iron from
blackband, the blast being heated to 300° C.; the slag
approaches a bi-silicate if calculating the alumina as an
acid; proportion of oxygen, 25°65 : 12°36.
No. 14. Slag from Witkowitz, black when in thick pieces
but pellucid and lightish green at the edges.
No. 15. Slag from Königshütte (Silesia), obtained at the
contemporaneous production of zinc and pig-iron by Schmidt's
method; by Meyer; it disintegrates, forming dust :—
3 [2CaO(SiO,,Al2O3)] +3CaO(SiO,Al₂O₂).
No. 16. Slag from Hoerde (Westphalia), by Mayrhofer.,
* B. u. h. Ztg., 1856, p. 73.
+ Ibid., 1858, p. 231.
394
IRON.
3. Slag from Coal and Anthracite Blast Furnaces.
Sio,.
3
Al2O3
CaO.
CaS.
MnS.
35'34
20°47
38.72
I'35
5°39
Slag from the Iron Works of Gartsherrie and Govan in
Scotland, analysed by Schwarz :-*
6RO,5(SiO3,Al2O3)=3RO,SiO3+ Al2O3,SiO3.
The fluxes to be used for mixing with iron ores depend on
the nature of the ores. It has been before mentioned that in
rare cases only are ores found containing all the components
in the proportion required for the formation of a suitable
slag. Most ores require for this purpose some addition of
earthy substances, and may, in this respect, be classified as
follows:
1. Rich Ores without Earthy Substances.—They are
mixed either with poor ores, or with fluxes of blast furnace
slags, neutral silicates, &c.
2. Siliceous Ores.-These are the most common; they are
mixed with different fluxes according to the state in which
the silica is associated with the ore, namely:
a. Silica is mechanically admixed with the ore; ores gene-
rally difficult to fuse, requiring aluminous and calcareous fluxes,
best in the form of aluminous lime, if suitable iron ores are
not obtainable. Fluor spar has some advantages over common
lime, as its fluorine volatilises part of the silicon; it fuses
easily with heavy spar, gypsum, and phosphate of lime, and
is therefore a good flux for ores containing those substances.
Ores at the same time containing quartz in a finely dis-
seminated state and protoxide of iron are difficult to smelt.
The siliceous ores without manganese produce grey or
mottled iron.
b. The silica is more or less saturated by other bases
besides iron; the silica is sometimes saturated with bases in
such a manner as to form a suitable slag without the addition
*
Polyt. Centr., 1856, p. 998.
MIXING THE IRON ORES.
395
of other fluxes; or when contained in richer ores an addition
of neutral fluxes is required only to produce the sufficient
quantity of slag.
In most cases, however, the silica is not sufficiently satu-
rated, and it then requires an addition of basic minerals,
calcareous ores or lime.
In rare cases ores are smelted containing silicates so basic
as to require an addition of silica or aluminous marl, &c.
c. The silica is combined with per- or prot-oxide of iron, as
in puddling, refinery slags, &c. These substance are difficult
to smelt, and the best method is Lang and Frey's, according
to which, 25 parts of well burned fresh lime are slaked, and
mixed whilst still warm with 65 parts of pulverised slag and
10 parts of pulverised coal. The mixture is moulded into.
forms, dried, and broken up to the size usual for smelting.
3. Calcareous Iron Ores being very refractory when
smelted by themselves, are mixed with argillaceous iron ores
or clayey substances. Pure quartz is seldom employed as
an addition, for it requires a longer time for the formation of
silicates and is liable to scorify protoxide of iron.
4. Iron Ores containing Magnesia are very refractory,
and require an addition of argillaceous substances and
lime.
If these ores contain at the same time a certain amount
of manganese the reaction of the magnesia will be partly
neutralised.
5. Manganiferous Iron Ores sometimes smelt by them-
selves, and are inclined to yield white pig-iron, but they
usually require an addition of lime in order to produce a
slag free from iron. An increased addition of lime or of
magnesiferous substances is required, if the production
of grey iron is intended from these easily fusible ores.
Easily fusible ores poor in manganese, such as some sorts of
blackband, also require a greater addition of calcareous
fluxes.
6. Titaniferous Iron Ores, usually difficult to fuse, require
fluxes of lime and quartz; alkaline fluxes are also very
effective.
396
IRON.
II. Mixing the Iron Ores with regard to the Quality
of the Fuel Employed.
The quality of the fuel essentially influences the slag with
regard to its amount of silica. Wherever it is possible, the
formation of slag is attempted in the proportion of the earths
as proposed by Bodemann (page 387), as this proportion
necessitates the smallest consumption of fuel when producing
mottled iron.
1. Charcoal. This fuel, being free from injurious sub-
stances and not producing so high a temperature as coke,
(therefore reducing less silicon), allows the formation of a slag
consisting of a bi-silicate of lime and alumina, and suitable
for the production of mottled or lightish grey iron. According
to Lindauer, a good silicate of the composition—
3CaO,2SiO3+Al2O3,SiO3
was formed at Horovic; it contained :-
SiO 3
CaO
A1203.
50'00
31.00
18.95
and likewise a silicate of the formula 3CaO,SiO3+Al¸O3,2SiO3.
A scorification of iron takes place if the slag is composed
as follows:-
3CaO,2SiO3+2(Al2O3,2SiO3), containing—
Sio.
CaO
Al2O3 ·
60
17
23
Janoyer states that in many cases a slag of sufficient
fusibility is formed if the ore mixture contains equal parts of
clay (containing on an average 75 per cent of silica and
25 per cent of alumina), and carbonate of lime (containing
50 per cent of lime); the slag formed will contain—
SiO 3
Al2O3
CaO
50'00
16.65
33°35
Tri-silicates being more difficult to fuse, and at the same time
scorifying iron, are intentionally formed in rare cases only.
More basic slags, approaching mono-silicates, are formed
by a larger addition of lime, if either the iron ores or the
fluxes contain sulphur.
MIXING THE IRON ORES.
397
If intending to produce white pig-iron, part of the lime
is replaced by oxide of manganese, the resulting slag then
being nearly-
3RO,2SiO3+Al2O3,2SiO3, or 3RO,2SiO3+R₂O3,SiO3.
The ore mixture is not altered by replacing part of the char-
coal by wood. The slags of charcoal blast furnaces are
usually of viscid fluidity, vitreous, becoming crystalline upon
slowly cooling; if containing a larger amount of manganese,
they are thinly liquid, solidify rapidly, and are usually of
a brown or yellow colour; they are light green when con-
taining protoxide of iron at the same time, and of a violet
or amethyst tint if in a high state of silication.
2. Coke.—As a general rule the slags of coke blast furnaces
must be more basic and richer in lime than those of charcoal
furnaces, and approach the mono-silicate of lime and alumina,
containing—
Silica.
Lime.
Alumina.
40
37
23
Owing to the larger amount of lime these slags are more
difficult to fuse, but are less viscid; they solidify more rapidly
than charcoal slags, and are strong and even earthy, but
though difficult to fuse they melt to a thin liquid at the high
temperature obtained by the coke.
The greater amount of lime in an ore mixture for coke
blast furnaces is necessitated by the following circumstances:-
a. Upon combustion, coke yields a higher temperature
than charcoal, thus facilitating the reduction of silicon, pro-
ducing a less strong pig-iron, the more so the hotter the blast
employed; an excess of lime impedes the reduction of
silicon.
A larger amount of alumina also counteracts the reduction
of silicon, but cannot completely prevent it.
Experience has proved that the production of strong grey
iron by means of a hot blast of about 300° C., and of coke
containing 10 per cent of ash, requires the formation of
slags in which the silica and alumina together amount to
very nearly the same quantity as the lime and other bases
added together.
398
IRON.
At Meppen* (Hanover), where chiefly bog iron ores, con-
taining 20 per cent of silica and from 2 to 3 per cent of
phosphoric acid, and clay ironstone were smelted with coke
containing 10 per cent of ash, the resulting pig-iron was
poorest in silicon if the ore mixture was composed according
to rule before stated, and this is proved by the following
analyses :-
Analyses of Slags.
Silica.
Alumina.
Lime.
Magnesia
Sulphur.
I.
II.
III.
47*56
38°10
38 10 36.61
10°48
12.09
Ι2ΟΙ
40*73 47.85 49°35
0'63
0'97
Analyses of the Pig-iron obtained.
I.
II.
III.
Dark Light
Grey.
Mottled. White.
Grey. Grey.
Grey.
Graphite 2'7750
1'7875
Carbon,
combined)
+
0*5045 6'0000
Sulphur. o*1238 0*1875
Silicon 3.8480 +
•
Phosphorus 1.8117
0'5012
2*7936
2°22 2°39
I'5
Arsenic 0*0650
++
I'8200
0'72 0'55
•
0'1500
Temperature of the
blast, from.
Consumption of coke
for I part of pig-
iron.
125° to 150° C. 225-250° C. 300-360° C.
1.600
I'720
1.800
According to Mène,‡ the most regular process for the
production of a good white iron by means of coke and hot
blast takes place if a slag is formed which is composed of
4CaO,SiO3+Al₂O,,SiO3, containing
Sio,
Al2O3 ·
CaO
38
20
42
* Ztschr. der Deutschen Ingen., v., 7. HARTMANN, Fortschr., v., 115.
† Not determined.
+
Polyt. Centr., 1862, p. 824.
MIXING THE IRON ORES.
399
Janoyer* produced brittle pig-iron containing 3 per cent
of silicon, when employing a very hot blast and forming a
slag=7(3CaO,Al2O3),10SiO3, containing 48 per cent of silica,
16 per cent of alumina, and 36 per cent of lime. When
forming a slag of the formula 8(3CaO,Al2O3),10SiO3, the
produced iron was stronger, and contained only 1.8 per cent
of silicon. In order to produce a very strong and fusible iron,
Janoyer formed a slag of the formula 6(3CaO,Al2O3),7SiO3,
containing-
SiO 3
Al₂O
CaO
3
41°5
ΙΟ Ο
48.5
whilst consuming from 135 to 137 lbs. of coke per 100 lbs.
of grey iron. This consumption of coke increased to 143 lbs.
when forming a slag of the formula 20(3CaO,Al₂O3),19SiO3.
Gaulthier and Mayrhofer‡ have obtained similar results.
If alumina is present in such quantity (in some blackbands)
as to react as an acid, the mixture is to be formed in such
a manner that the oxygen of the bases amounts to half the
quantity contained in the silica and alumina together, viz. :—
20 containing 10'39 oxygen 26′26
Silica.
Alumina .
Lime.
34
46
15.87
I2*22
b. The coke ash, mostly from 5 to 10 per cent, approaches
a bisilicate in its composition. The following analysis is of
the ash of a coke from Westphalia :—
SiO3
A1203
Fe₂03
Mn₂O
3
Cao.
MgO .
FeS
Alkali and loss
54'78
30'92
7'79
0*64
1.26
0°51
0°19
3'91
The silica of the coke ash will be easily reduced at a high
temperature, if not counteracted by'an excess of lime in the
* B. u. h. Ztg., 1856, p. 306; 1862, p. 234.
† Ibid., 1858, p. 307.
+
LEOB. Jahrb., 1861, x., 371, 391.
400
IRON.
mixture. But lime cannot fix the silica of the ash so com-
pletely as the silica contained in the ore mixture, for the
reason that the ash is not liberated from the coke before it
passes the tuyere, where the other components of the slag
have already combined and fused, and the time is too short.
to combine all the ash in the slag.
Mayrhofer states that the injurious action of a larger
amount of ash may be neutralised by employing lime-coke
(vide chapter on Fuel), and several different silicates, instead
of silicates of alumina and lime only.
Whilst coke ash necessitates a larger consumption of fuel
and lime, charcoal ash reacts favourably.
c. An addition of lime is necessitated by the amount of
sulphur which coke always contains, and more lime must be
added, the temperature of the blast being raised correspond-
ingly, the freer the resulting pig-iron is intended to be from
sulphur. When smelting pure ores with pure coke, requiring
no large addition of lime, the resulting slags will be either
wholly vitreous, or they will show a stony kernel when in
thicker lumps; when hot this slag may be drawn into thread.
Slag resulting at the production of forge pig has a smaller
kernel than that from the production of foundry iron. Slags
with an increasing amount of lime gradually devitrify, are
less capable of being drawn into thread, are thinly liquid,
solidify rapidly, and are nearly white and thoroughly stony
if containing the largest amount of lime. On exposing
them for a longer time to the atmospheric air, they disin-
tegrate into dust, as the contained sulphide of calcium and
the basic silicate of lime become decomposed.
In order to produce pig-iron as free as possible from sul-
phur, basic slags are formed at Hörde,|| Hattingen,§ Porta
Westphalica,¶ Scotland,** &c.
**
LEOB., Jahrb., 1861, x., pp. 358, 361.
† Ibid., 1861, X., 324.
+
B. u. h. Ztg., 1861, p. 357.
TUNNER'S Bericht. über d. Londoner Industrie-Ausstellung, v., 1862, p. 30.
§ B. u. h. Ztg., 1858, p. 231.
¶ B. u. h. Ztg., 1859, p. 156.
** B. u. h. Ztg., 1862, p. 323.
HARTMANN, Fortschr., ii., 225.
Berggeist, 1859, p. 91.
MIXING THE IRON ORES.
401
When intending to produce white forge iron, part of the
lime is replaced by manganiferous fluxes.
d. The higher temperature and the slower sinking of the
charges in coke furnaces offer more opportunities than
charcoal furnaces for the reduction of silica by carbon or
iron, and therefore require a more basic mixture.
The addition of lime in English iron works* varies
between 0'4 and 1'1 part for each part of pig-iron produced.
3. Coke and Charcoal.t-When employing these kinds of
fuel in admixture, the properties of both must be taken into
consideration at the formation of the ore mixture.
4. Mineral Coal usually contains a larger amount of sul-
phur, necessitating a larger addition of lime if the production.
of good pig-iron is intended.
III. Mixing the Iron Ores with regard to Temperature
and Pressure of the Blast.
The higher the temperature of the blast, either for the
purpose of economising fuel or of increasing the production,
the more basic and richer in lime must be the slag, in order
to reduce as little silica as possible. Therefore, the slag
(Königshütte in Upper Silesia) produced in a coke furnace.
whilst employing blast of from 60° to 70° C. is a combination
of mono- and bi-silicates, containing :-
SiO
Al2O3
CaO
MgO
MnO
•
47'34
II 44
34'45
I'94
3'05
whilst slags formed when employing a blast of 300° C. are
either mono-silicates or mixtures of mono- and sub-silicates.
Iron ores difficult to reduce (some sorts of magnetic iron.
ores) or easily fusible ores (some sorts of spathic and blackband
ores) have a tendency to yield a white, imperfectly carbonised
iron, sometimes rich in sulphur and silicon if produced by
* B. u. h. Ztg., 1862, p. 253. Allgem., B. u. h. Ztg., 1863, p. 170.
† LEOB., Jahrb., 1861, x., 376, 389. Oesterr. Ztschr., 1861, No. 2.
B. u. h. Ztg., 1858, p. 227.
VOL. II.
2 D
402
IRON.
means of cold blast, even with a great consumption of fuel;
they must be mixed with lime in order to render the ore
more refractory, so that it may be exposed for a longer time
to the reducing action of carbonic oxide gas, and be smelted
with hot blast, thus producing the required temperature
without an excessive consumption of fuel.
If intending to retain the same quality of pig-iron, the
peroxides in an ore mixture must not be replaced by prot-
oxides, but by oxides only; the mixture of protoxides and
peroxides necessitates a larger addition of lime to render the
mixture more refractory.
The pressure of the blast depends chiefly on the compact-
ness of the fuel, and the fusibility of the ore mixture, and
is of great importance when calculating the quantity of
slag required. A certain amount of slag is required to pre-
serve the reduced iron from the oxidising reaction of the blast
and carbonic acid before the tuyere and in the furnace hearth,
and that quantity of slag must be increased if the pressure
of the blast is increased. Therefore the quantity of slags
must be greater when employing coke than when using
charcoal, and smaller when producing white pig from easily
fusible ore than when producing grey iron.
As a general rule, the quantity of slag to that of pig-iron
stands in the proportion as 100 to 100 when smelting an ore
mixture of moderate fusibility with coke and a pressure of
20 centimetres of mercury; this proportion may increase
to 200 to 100 if the mixture is more difficult to fuse and a
higher pressure of the blast employed. When smelting the
rich blackband in Scotland, the quantity of slag sometimes
decreases to 93; and, on the other hand, at Galicia we find the
abnormal proportion of 100 of pig-iron to 450 of slag.
The proportion of slag to pig-iron in charcoal furnaces is
usually below I: I, and may be as low as 40: 100 when smelting
an easily fusible ore mixture with a low pressure of blast.
The loss of iron increases with the quantity of the slag, and
as siliceous and aluminous ores require more fluxes than
calcareous ores the latter give a better yield of iron. The
least loss of iron takes place, for example, at the smelting
MIXING THE IRON ORES.
403
of the rich easily fusible Scotch blackband ores, which con-
tain nearly 60 per cent of iron.
IV. Mixing the Iron Cres with regard to the Quality of
the Pig-Iron.
The production of some sorts of pig-iron requires a certain
natural state of the ores, which cannot be obtained artificially,
or only with great difficulty. For example, spiegeleisen can
only be produced from easily fusible and reducible ores almost
free from sulphur and phosphorus, and rich in manganese,
and usually containing already a sufficient quantity of
slag-forming components, or requiring a small addition of
lime only; the production of these sorts of iron is therefore
confined to certain localities. Again, other sorts of iron ore.
allow the production of the most varying qualities of pig, if
the ore mixture, fuel, construction of the furnace, process,
and above all the smelting temperature, are suitably modi-
fied. Sometimes the same furnace produces alternately grey
and white pig, and sometimes always one certain variety. This
distinction of different sorts of iron for foundry, wrought-iron,
and the manufacture of steel, is not so strictly maintained in
Germany as in Belgium and Great Britain; in this country
the distinction is made not only in single furnaces but even
in whole districts.* For example, in Scotland foundry iron is
produced from blackband with raw coal; Yorkshire produces
pig-iron for the manufacture of best fibrous iron and boiler
plates from clay-iron ore with coke; Staffordshire with the
same iron ores and coke or coal produces pig-iron for con-
version into common wrought-iron; Wales produces iron
for rails from clay, red and brown iron ores with coke and
coal; Cumberland, steel-iron from red iron ore with coke.
The following are the chief rules for mixing the ores with
regard to the quality of the pig-iron :—
The production of spiegeleisen and flowery iron requires a
pure manganiferous iron ore, necessitating little or no fluxes.
If upon giving large charges of this ore, and thus lowering
Amt. Bericht über d. Londoner Industrie-Ausstellung, v., 1862. Berlin,
1863, Hft. i., p. 27.
2 D 2
404
IRON.
the temperature, the porous, flowery white pig-iron will
result, upon charging less ore, spiegeleisen containing more
or less of manganese is produced.
The amount of manganese in the pig-iron increases with
the quantity of raw spathic ores employed, and also if the
mixture is more basic than acid, if hot blast is employed, if
the gases in the furnace hearth are of higher pressure, and
under other circumstances which have not been sufficiently
investigated.
An easily fusible slag, standing between a mono- and bi-
silicate, in which the lime exceeds the alumina, counteracts
the reduction of silica. As coke produces higher temperatures,
flowery pig can only be produced with it with interruptions,
whilst spiegeleisen may be reduced continually with coke,
even when employing hot blast (Siegen employs charcoal,
coke, and hot blast; Vordernberg, charcoal and hot blast; the
Weardale Iron Company, at Ferryhill in Durham, coke and
cold blast).
If the production, at a regular process, of white pig-iron
is intended from iron ore containing sulphur and phosphorus,
a larger addition of lime is required, and also an addition of
manganiferous flux if the ores do not contain manganese;
sometimes the easy fusibility of the mixture is obtained by
scorification of iron; as, for instance, in this country, where
white forge iron is produced by means of an increased
addition of puddling slags.
If the ores contain a larger amount of sulphur, a mottled
or grey forge iron must be produced by increasing the addition.
of lime and using hot blast. If intending to produce a soft,
tough, grey pig-iron fit both for foundry and forge purposes,
ores containing phosphorus, sulphur, and manganese, and
slags must not be employed, and circumstances sometimes
also make it necessary to avoid the easily fusible fluor spar.
Iron ores containing a great deal of phosphorus are unfit
for the production of either white forge iron or grey forge
iron, but they are well adapted for foundry iron owing to its
thin fusibility. Mottled pig-iron is also produced to serve
as a suitable mixture with other sorts of forge iron in the
puddling furnace, or to be employed for certain kinds of
FLUXES.
405
castings. The different methods of converting pig-iron into.
wrought-iron performed in open hearths require different
forge iron, some methods requiring white and others mottled.
iron.
Grey foundry iron requiring to be of great strength is pro-
duced from basic ore mixtures, so that it may be richer in carbon
and poorer in silicon and sulphur. The mixture for common
foundry iron is more acid, and pig-iron intended for re-melting
is produced from a still more acid mixture. Sulphurous ores
are sometimes intentionally added to the ore mixture if a
small amount of sulphur in the pig-iron is desirable—cannon
iron for example. Grey forge iron, with an inclination to
become white on chilling, must be produced from a mixture
poor in silica and rich in lime, and not too refractory.
Blackish grey pig-iron is only produced when the forma-
tion of a mixture very difficult to fuse is required, owing to an
excess of alumina or magnesia, or if both ore and fuel con-
tain a larger amount of sulphur.
In this country, where most of the iron ores are easy to
fuse and reduce, and contain phosphorus with the exception
of the red and spathic iron ores, a very high temperature is
kept in the blast furnace in order to produce a grey pig-iron
poor in phosphorus. In Scotland, where sulphurous coal and
very basic mixtures are employed, the temperature of the
furnaces is kept so high as to produce darkish grey iron, the
slight tenacity of which is ascribed by Gruner and Lan to the
presence of aluminium.
The white grained forge iron used for the manufacture of
common rails is produced at a less high temperature and
with a large addition of slags.
According to Mayrhofer, the amount of mixture required
for the production of 100 lbs. of pig-iron is equal to
lbs., p expressing the percentage of iron in the mix-
I0,000
P.
ture.
B. Fluxes.
Fluxes are added to the iron ore mixtures for the following
purposes :
-
a. To supply the components required for the formation
406
IRON.
of slag if they are absent in the iron ores. These fluxes
may be
1. Basic Fluxes, namely-
Calcareous substances, of which the following are applied
according to requirement :-Pure limestone is added in ad-
mixture to aluminous ores; aluminous limestone (marl) to
siliceous ores; fluor spar to ores of difficult fusibility, chiefly
if they are siliceous; magnesian limestone (brown spar, dolo-
mite,† ankeritet) is added in admixture to easily fusible ores.
Caustic lime is sometimes used instead of limestone, and
produces a certain economy of fuel, as the local cooling, on
account of the absorption of heat in the blast furnace, caused
by expulsion and reduction of the carbonic acid, is dispensed
with. Caustic lime also increases the production, as, owing
to its smaller volume, more charges pass through the furnace
in a given time. The advantages will be greater the poorer
the ores are and the more flux they require. Comparative
experiments on this point have been made at Ougrée in
Belgium, and Königshütte in Silesia. In the former case,
26 per cent of lime replaced 40 of limestone, and the produc-
tion of metal was increased by 2'3 per cent, with a saving of
1.6 per cent of coke. In Silesia the saving was 2.85 per
cent, and the increase of production 3.1 per cent.
Phosphate|| and sulphate of lime, and sulphate of alumina,§
sometimes present in limestone, are injurious. Limestone
also contains now and then an undesirable amount of silica, T
which makes it necessary to increase the addition of lime.
The purest limestone usually contains 1 or 2 per cent of
silica, and the poorer sorts from 5 to 10, and even up to
25 per cent.
Basic silicates, such as greenstone, ** serpentine, &c., are
in some cases employed.
+
* LEOB., Jahrh., 1861, x., 318.
† B. u. h. Ztg., 1861, p. 286.
Jahrb. d. K. K. geolog. Reichsanst., 1853, p. 827; 1856, p. 806; 1859,
pp. 26, 30.
|| B. u. h. Ztg., 1861, p. 356.
§ LEOB., Jahrb., 1861, x., 325, 370.
Allgem., B. u. h. Ztg., 1863, p. 171.
** LEOB., Jahrb., 1861, x., 318.
FLUXES.
407
Manganiferous substances (Franklinite, brownstone, &c.,)
are used for the production of white pig-iron.
2. Acid Fluxes, mostly in the form of acid silicates
(clay slate, basalt, porphyry,* &c.,) are in some cases required
for mixing with strongly basic ores.
3. Neutral Fluxes, used for ores rich in iron; mixtures
of native silicates (garnet,† augite, porphyry‡) with limestone
and sometimes iron blast furnace slags are made use of.
Some of the fluxes, as we have before mentioned, act upon
the noxious substances in iron (sulphur, phosphorus, silicon).
These absorbing fluxes are chiefly lime, manganiferous sub-
stances, and perhaps alkalies, || besides fluor spar (page 394).
Some kind of fluxes are given to the ore mixture in order
to introduce certain substances into the pig-iron, thus in-
creasing its tenacity, and to purify it, so that it may yield
superior qualities of wrought-iron and steel. These fluxes
are chiefly manganiferous and titaniferous minerals and
tungsten.
La Gren§ states that tungsteniferous pig-iron is stronger
than iron without tungsten, and assumes the nature of steel,
as tungsten becomes reduced by the carbon of the iron.
Riley states that many sorts of English pig-iron contain
titanium, which, he says, acts like manganese (page 316).
Some fluxes have a solvent action, and are sometimes
used for the removal of obstructions or scaffolds in the fur-
nace, which may be caused by either too acid or too basic
mixtures, by too wet fuel and friable ores, by zinc contained
in the ore, &c.
Such solvent fluxes are—
Fluor Spar; this is used if the other fluxes do not answer,
as it forms a thinly liquid slag, combining with silica and
bases, and also as fluorine has great affinity to silicon; fluor
spar reacts strongly upon scaffolds formed either by too basic
or too acid mixtures.
* LEOB. Jahrb., 1861, X., 317.
+ B. u. h. Ztg., 1857, p. 103.
LEOB., Jahrb., 1861, X., 323.
|| Ibid., 1861, x., 329. Berggeist, 1856, p. 207. Oesterr. Ztschr., 1862, p. 379.
§ B. u. h. Ztg., 1861, pp. 167, 356, 358. Schles. Wochenschr., 1860, No. 9.
408
IRON.
At Königshütte in Upper Silesia, some not very large
scaffolds were removed by 10 cwts. of fluor spar, I cwt. of
which was added to each ore charge. Larger scaffolds were
removed by introducing 30 lbs. of fluor spar through the
tuyeres every four or six hours. But, according to V. Mayr-
hofer,* it is less advisable to charge fluor spar through the
tuyeres, as pig-iron richer in silicon will be formed; for the
reason that fluor spar then attacks silica at the place where
the liberated silicon comes into contact with the fused iron,
whilst otherwise the decomposition would take place in the
upper parts of the furnace where the iron is not yet in a
liquid state.
Iron blast furnace slagst of the regular process, of the
size of a walnut, are sometimes employed as solvent fluxes
in such a manner that the ore mixture is more or less de-
creased in weight, and the wanting ore replaced by such slag
in two or four charges.
Slags from puddling and re-heating furnaces are mostly
effective if the scaffolds originate from siliceous ores, but
they must be carefully employed, as they easily attack the
furnace lining.
The size of the pieces of flux must be nearly the size of
the ore pieces, but it is advisable in coke furnaces to use
limestone in pieces not exceeding 1 lb. in weight, and a still
greater division is advantageous if intending a more intimate
contact with fuel and ore in order to remove noxious sub-
stances from them by means of the lime. If the pieces of
lime (from 5 to 20 lbs. in weight) are too large they cause a
greater consumption of fuel, and form a slag difficult to fuse,
enclosing caustic lime, and containing a large amount of
iron if not perfectly fused.
C. Fuel.
The properties of the different kinds of fuel will be treated
in a subsequent chapter, and the following account con-
cerns only the applicability of fuel to the production of pig-
iron.
*
LEOB., Jahrb., 1861, x., 332.
† Ibid. B. u. h. Ztg., 1861, p. 358.
FUEL.
409
For reducing pig-iron the fuel produces the temperature
required for smelting the iron and the slag by burning
before the tuyere, thus becoming transformed into carbonic
acid, which, after being partly reduced to carbonic oxide by
glowing coal, reduces the oxidised iron and carbonises the
reduced iron.
Some consumption of coal may also take place by the
reduction of carbonic acid liberated from ores or fluxes, by
the reduction of oxide of manganese to protoxide, of sulphuric
acid, phosphoric acid, &c.
The amount of fuel required for the production of a certain
quantity of pig-iron depends on the following circumstances:-
I. On the nature of the ore mixture (whether it is rich,
easily fusible, &c).
2. On the quality of fuel (raw or carbonised, rich or poor
in ash, &c.)
3. On the nature of the produced iron (white or grey).
4. On the quantity and nature of the fluxes.
5. On the construction of the blast furnace.
6. On the quantity, pressure, moisture, and chiefly tempe-
rature of the blast; the consumption of fuel in English iron
works is stated to be 13 per cent larger in summer than
in winter, owing to the greater moisture in the air.
7. On the state of aggregation of the ore and fuel, as well
as on the care devoted to the charging operation.
8. On the more or less perfect use of the furnace gases.
In this country the consumption of fuel has been considerably
lessened since hot blast, raw coal, enlarged furnace mouths,
and the furnace gases have been made use of.
Kirchwegert asserts that the chemical and physical pro-
perties of the ores have no influence whatever upon the
consumption of fuel, which, moreover, stands in direct pro-
portion to the quantity of mixture (ore and fluxes) to be
smelted; and that the perfection of a blast furnace smelting
process may always be judged by the consumption of fuel.
Kirchweger arrived at this conclusion from a number of
* B. u. h. Ztg., 1852, p. 205.
† Berggeist, 1859, No. 103; 1860, Nos. 21, 32, 39, 58. B. u. h. Ztg., 1860,
pp. 284, 321, 503; 1861, pp. 78, 80, 271.
410
IRON.
calculations he made of the consumption of fuel in different
blast furnaces, some of which are the following:-
Aplerbeck (Westphalia)
Consumption of Coke. Nature of the Iron.
per lb. of
Mixture.
grey iron.
0°423-0*387 grey iron.
0*369-0*332 mottled and white
•
0°539
Georg Marienhütte (Hanover)
Hof (Bavaria)
Ougrée (Belgium)
Porta Westphalica .
Hörde (Westphalia)
0'332
0'394-0*360
• 0'411-0 424
Phönix (Rhine) .
. 0°427
iron.
white iron.
according to
the percentage
of the ore added.
according to
the percentage
of the ore added.
Neustadt (Hanover)
Consumption of Charcoal. Nature of the Iron.
0*356
Altenbecken (Paterborn). • 0'341
Dassel (Brunswick)
.0'369
grey.
grey, temperature
of the blast 252°-
315° C., the mix-
ture yielded 29°44
per cent of iron.
grey.
Lüneburg (Hanover)
• 0*385
Consumption
of Charcoal.
Nature of the
Iron.
Beckerode (Hanover). 0358
Rothehütte
0*383
Königshütte
0 385
""
flowery white pig
grey
Percentage
of the Ore
Mixture.
35/1/1
32/
Steinrennerhütte Hartz o°384
Lerbacherhütte
Altenauerhütte
0*381
0*362
mottled
27
301
32
Without admitting that the chemical composition of the
ores and fluxes, their degree of porosity and fusibility, &c.,
had no influence on the consumption of fuel, we consider
Kirchweger's calculations of value, as they call the attention
to the great differences in the relative consumption of fuel,
which to a certain degree might be avoided in so far as they
FUEL.
411
are founded on the following circumstances:-Moisture, im-
perfect mining or dressing, insufficient degree of division and
unsuitable largeness of the charges of fuel, ore, and fluxes,
moisture insufficiency or irregularity of the blast, moist
foundation and wrong construction of the blast furnaces,
irregular charging, ill-conducting off of the furnaces gases,
&c. The amount of ash contained in the fuel considerably
influences its consumption.
1. Carbonised Fuel.-Charcoal and coke are mostly used.
a. Charcoal yields a superior pig-iron, owing to the more
suitable composition of its ash, and to its property of pro-
ducing a less high temperature than coke; therefore when
smelting with charcoal a smaller production of superior pig-
iron is attempted; whilst a larger production of inferior pig-
iron results from smelting with coke, facilitated by a suitable
construction of the blast furnace, and the application of a
large amount of hot blast of high pressure.
An increased production is obtained by using coke in a
charcoal furnace without enlarging its dimensions. For
example, a charcoal furnace of the capacity of 50 charges,
each 33 cubic feet 1560 cubic feet of charcoal, is able to take
150 charges of II cubic feet 1650 cubic feet of coke. As
the effect of I cubic foot of coke is equal to 2.75 cubic feet
of charcoal, 1650 cubic feet of coke will correspond to about
4538 cubic feet of charcoal, from which it follows that the
production will be two and a half times as large when using
coke, supposing that the ore charges made in 24 hours are
smelted in the same space of time as is used when employing
charcoal.
Only two charcoal blast furnaces are used in this country,*
namely, in Lancashire (Newland and Backbarrow); and the
same number in Belgium. In Germany and France also a
great deal more coke iron than charcoal iron is at present
produced.
Owing to its being less compact, charcoal combines more.
readily than coke with the oxygen of the blast before the
TUNNER, Bericht. über d. Londoner Industrie-Ausstellung, v., 1862, p. 28.
B. u. h. Ztg., 1862, p. 169; 1863, p. 156.
412
IRON.
tuyere, thus preventing a decarbonisation of the fused pig-
iron which passes before the tuyere, and quickly reduces the
carbonic acid formed to carbonic oxide, in sufficient quantity
for the reduction of iron ore in the upper parts of the furnace.
This reduction of carbonic acid to carbonic oxide causes a
lowering of the temperature, and less pig-iron is oxidised by
carbonic acid. The lower temperature also partially pre-
vents the reduction of silica by iron, coal, or carbonic oxide.
In consequence of these circumstances the resulting pig-iron
is rich in carbon and poor in silicon, chiefly when employing
cold blast and soft charcoal; hard charcoal, being less com-
bustible, approaches coke in its behaviour.
Statements concerning the consumption of charcoal for
producing a certain quantity of pig-iron vary. Karsten cal-
culates the consumption to be as follows:-From 16 to
3 parts of soft charcoal and o‘66 to 12 parts of hard charcoal
to I part of pig-iron when smelting rich easily fusible ores
in lumps; from 06 to 12 parts of coal when producing
spiegeleisen; from o'91 to 2*21 parts, on an average 1'5 parts,
of charcoal when producing grey iron. These statements
differ from the experiments made in the iron works of the
Upper Hartz, where at least the same quantitity, and gene-
rally somewhat more, of hard charcoal is used, than soft
charcoal, for the production of mottled and grey iron (109 to
1*2 parts), supposing the other circumstances to be the same.
Lindauer states the consumption of charcoal, when using
cold blast, to be 16, and 13 when using hot blast. For the
production of porous white iron (luckige floss) in Styria the
consumption of charcoal amounts to from o'6 to o'65, and
for spiegeleisen in Siegen, from o'75 to o'9; for grey iron
usually from 1 to 1'5, seldom to 2.
b. Coke.—Its applicability to the production of pig-iron
depends-
1. On the quantity and quality of the ash it contains;
coke containing above 10 or 12 per cent is used with less.
advantage. The washing of coal previous to coking is there-
fore an important step in iron smelting, as the purer coke
*B. u. h. Ztg., 1855, p. 261.
FUEL.
413
thus obtained facilitates the smelting process, lessens the
consumption of fuel, and produces a superior iron. On em-
ploying steam or lime in the operation of coking a further
removal of sulphur takes place. Sulphur is also removed
when mixing the ore with caustic lime instead of limestone.
French coke contains from 10 to 12 per cent of ash; the coke
of this country much less.
2. On the state of aggregation, whether caking, sinter, or
sand-coke.
The state of aggregation has an essential influence on the
compactness, effect, and combustibility of the coke, and
accordingly it behaves differently* in blast furnaces, chiefly
in requiring a different height of the furnace.t
Owing to its greater compactness, coke combines with
more difficulty with oxygen than charcoal, and requires a
hotter blast of greater pressure in order to transform the
greater part to carbonic acid. The greater compactness of
coke also causes the carbonic acid to be reduced by it to
carbonic oxide with more difficulty, therefore a larger space
above the tuyere is filled with carbonic acid at a higher tem-
perature than in charcoal furnaces; consequently the pig-iron
is more readily decarbonised by the carbonic acid, and silica
is reduced; this is also facilitated by the slower combustion
of the coke causing a slower passing of the charges. In
order to preserve the pig-iron in the furnace hearth as much
as possible from the oxidising reaction of the blast and car-
bonic acid, the carbonic acid must be brought into contact
with a larger surface of coke (this is also necessary for the
reduction of the iron ores), thus reducing the acid to carbonic
oxide; this end can only be obtained by an increased quantity
of coke in proportion to the ore charges.
This circumstance explains the fact that the production of
a certain quantity of pig-iron requires more coke, sometimes
from one and a-half to twice as much coke as charcoal.‡
* B. u. h. Ztg., 1859, p. 171; 1863, p. 282.
† Ibid., 1850, p. 641; 1861, p. 169; 1862, p. 210. Allgem., B. u. h. Ztg.,
1863, pp. 158, 190. Schles. Wochenschr., 1860, No. 9.
‡ Bgwkfd., viii., 456. KARST., Archiv., 1 R., xi., 119. B. u. h. Ztg., 1860,
p. 208. LEOB., Jahrb., 1860, ix., 315; 1861, x., 352.
414
IRON.
The more compact the fuel is and the more ash it contains
the hotter the blast of higher pressure is required, and the
more carbonic acid will be formed, which will again be more
slowly transformed into carbonic oxide.
At Heinrichshütte in Westphalia, the consumption of coke
per I part of mottled and white pig-iron amounts to from I'I to
1*3 parts only when smelting rich spathic ores, which are easily
reduced and carbonised, the temperature of the blast being
from 280° to 300° C. Grey forge iron required 14, and grey
foundry iron from 15 to 175 parts of coke. 17 parts of coke.
are used in Belgium when smelting white pig from brown
iron ores with a hot blast of from 70° to 100° C., and 2'0 parts.
when smelting grey iron. The iron works on the Lower
Rhine consume 17 parts of coke when smelting red iron
ores; the iron works at Hasslinghausen from 1'9 to 2'3 parts,
smelting blackband with hot blast. The English iron works
consume on an average from 1'75 to 2'1 parts of coke to
I part of iron produced.
In many cases the consumption of coke is equal to the
quantity of pig-iron and slag added together.
The following are the chief differences in smelting with
coke and with charcoal :-
Owing to the greater height of the coke furnaces, ores and
fluxes are charged in larger pieces, and frequently in a raw
state, and the ore mixture is composed so as to form mono-
silicate slags, whilst the mixture for charcoal furnaces forms
bi-silicates. The pressure of the blast for coke furnaces
(4 to 8 inches) being higher than that for charcoal furnaces
(1 to 3 inches) requires a larger quantity of slags (page 402),
and also the temperature of the blast for coke furnaces is
higher than that of charcoal furnaces, and the consumption
of fuel is larger in coke furnaces.
Owing to the more difficult combustion, the charges in coke
furnaces pass more slowly. When producing grey pig-iron
with charcoal one ore charge requires 16 hours for passing
through the furnaces, and 40 hours when employing coke ;
the ore charges require about three quarters of that time.
FUEL.
415
when producing white pig-iron.* The production of coke
furnaces would therefore be smaller if they were not enlarged
in their horizontal section, whilst employing at the same
time a hot blast, thus producing a uniform temperature
throughout the section of the furnace. The enlarged cross
dimensions of coke furnaces which have lately been chiefly
employed, whilst they mostly retain the former height of the
furnaces, and without accelerating the passing of the charges,
cause an increase of the production, and a saving of fuel,
may be considered as an important improvement in iron
smelting. This enlargement of the furnace has, of course,
necessitated an increase of the blast, which is obtained either
by a higher pressure by employing tuyeres of larger size, or
by employing a greater number of tuyeres. The quantity of
blast required depends on the quantity of fuel to be burned in
a certain time to obtain the intended production. Under
the same circumstances, the production of a blast furnace
increases almost in direct proportion with its cross dimensions,
and therefore nearly the same ratio will result upon dividing
the number expressing the weekly production of pig-iron, by
the cross section of the furnace hearth expressed in cubic
feet.
The coke furnaces on the Continent produce on an
average from 140 to 200 tons per week, whilst in this
country some of these furnaces produce as much as 600 tons
and upwards.‡ The weekly production of charcoal furnaces.
averages between 20 and 45 tons; yet some charcoal fur-
naces produce even less, whilst others again a great deal
more; for instance, some Swedish furnaces produce from
50 to 75 tons, Russian furnaces 100 tons, some Styrian
furnaces from 150 to 175 tons, and the Rachette furnace 210
tons.
We shall explain later on the variations in the construc-
tion of the blast furnaces, which make them suitable for the
different kinds of fuel.
* B. u. h. Ztg., 1855, PP. 243, 261.
+ Ibid., 1857, p. 181; 1860, p. 332; 1862, pp. 208, 339.
+
Allgem., B. u. h. Ztg., 1863, pp. 23, 95. B. u. h. Ztg., 1861, pp. 335,
438; 1862, pp. 208, 413; 1863, pp. 156, 279.
416
IRON.
Owing to the larger consumption of fuel, coke furnaces
give off from the top a greater quantity of waste gases than
charcoal furnaces; these gases contain more carbonic oxide,
and are also hotter (228-330° C.) than the gases from
charcoal furnaces (100-120° C.), as the higher temperature
of the coke furnace causes some carbonic acid to be trans-
formed into carbonic oxide in the higher parts of the furnace.
The amount of hydrogen in the waste gases of coke furnaces
is less than that in charcoal furnaces, as coke contains fewer
hydrogen compounds (1—2 per cent) than charcoal (10—15
per cent.) Though consuming more fuel when producing
grey iron than when producing white iron, thus causing a
stronger heating of the former gases, they show no great dif-
ference in temperature; this is owing to the circumstance that
upon the production of white iron, some iron is generally burnt
before the tuyere, which increases the temperature of the
rising furnace gases.
The waste gases are sometimes composed as follows, and
may have the subjoined heating powers :-
Gases from
Coke. Mineral Coal.
Charcoal.
I.
II.
Nitrogen .
63'4
59'7
64°4
56.3
Carbonic oxide
29.6
20°2
34.6
21.5
Carbonic acid
5'9
19'4
0'9
15'2
Carburetted hydrogen
(marsh gas)
Ι'Ο
0'3
4°2
Carburetted hydrogen
(olefiant gas)
1.8
Hydrogen
O'I
0'4
O'I
ΙΟ
Total
•
100'0
100'0
100'0
100'0
Specific
Pyrometric
0*000105
""
1255° C.
Absolute heating power o‘081
When employing hot blast, the ores are reduced and
smelted at a lower level of the furnace than when using cold
blast; the gases have, therefore, to pass a longer distance
Bgwkfd., viii., 450.
† Ibid., viii., 456.
0*06
0'077 0*162
0.000078 0.0001 0˚000211
1075° C. 1265° C. 1480° C.
FUEL.
417
before arriving at the top of the furnace, and are thus
enabled to communicate their temperature to the ore more
completely, and consequently they escape from the furnace.
in a cooler state.
Sefström states the relative temperatures of the different
parts of the furnace for hot and cold blasts, to be as follows:-
At the lower part of the furnace throat
At of the height of the furnace shaft
""
3
4
4
وو
In the furnace hearth
""
وو
Cold Blast.
800° C.
Hot Blast.
500° C.
800°
1000°
وو
1000°
1200°
""
1400°
>>
1400°
1600
1700°
""
When employing soft charcoal in blast furnaces the saving
in fuel amounts to from 15 to 20 per cent, whilst 40 per cent
of mineral coal is saved; according to Potter, this saving
of coal amounts in Scotland to 72 per cent.*
Blast furnaces which do not make use of the waste gases,
lose about one-half the temperature produced by the fuel
employed.t
Charcoal is sometimes partly replaced by coke in order to
reduce the cost of production.‡
In some cases (Ireland) || turf coal is also employed; at
Underwiller,§ 1 part of turf coal replaces I part of charcoal
whilst charging 580 kilos. of brown iron ore, 140 kilos. of
turf coal, 140 kilos. of charcoal, and 75 kilos. of fluxes;
I part of pig-iron consuming 15 parts of fuel. A small
amount of lime which is contained in the ash of the turf
coal allows a saving of limestone.
At Tangerhütte T equal parts of turf coal and charcoal
yielded a foundry pig-iron, as good as the iron produced with
pure charcoal.
*
VALERIUS, Eisenhüttenkunde, i., p. 284.
† Berggeist, 1858, pp. 135, 179.
+
KARST. Arch., 2 R., xii., 551. LEOB., Jahrb., 1861, x., 426; 1863, xii., 155.
Ailgem., B. u. h. Ztg., 1862, p. 461. TUNNER, Ber. über. d. Londoner In-
dustrie-Ausstellung, in 1862, p. 40. B. u. h.' Ztg., 1861, p. 335. Berggeist,
1863, No. II.
|| Oesterr. Ztschr., 1854, p. 330.
§ B. u. h. Ztg., 1862, p. 264.
¶ Preuss. Ztschr., ii., 172.
VOL. II.
B. u. h. Ztg., 1963, pp. 139, 148.
2 E
418
IRON.
Fuel in a Raw State is frequently employed with the
following advantages :-When carbonising wood in the usual
way in heaps only 18 or 19 per cent of the 40 per cent of
carbon contained in the air-dried wood is made available,
whilst from 29 to 30 per cent of charcoal is produced in the
blast furnace. A calculation shows that in the Prussian iron
works about 18 million cwts. of wood are lost in the usual
way of carbonising.
The iron works at Dowlais consume 45 cwts. of coal for
the production of 1 ton of pig-iron, whilst formerly 50 cwts.
were consumed when applying the coal in a coked state.
The reducing gases evolved at the carbonising in the blast
furnaces reduce some oxide of iron, thus saving fuel, as well
as the apparatus for the carbonisation and the cost of the
operation itself.
The application of raw fuel has, on the other hand, the
following disadvantages :—
The carbonisation of wood or turf in the blast furnace is
liable to cause an increase of the temperature of the upper
part of the furnace, whilst it is lowered in the lower parts,
thus giving rise to an imperfect reduction and carbonisation
of the iron, as well as to an irregular process. Therefore
wood and turf are only employed in admixture with charcoal
and coke.
The carriage of raw fuel is more expensive.
Mineral coal contains more sulphur than coke.
The carbonisation of raw fuel in blast furnaces sometimes
gives rise to explosions,* and as wood must be applied in
small pieces it causes a great deal of mechanical labour.
In order to prevent too high a pressure of the gases which
are abundantly evolved from the raw fuel in the upper part
of the furnace, and in consequence of which the combustion
before the tuyere is impeded, it is advisable to employ larger
furnaces, chiefly those having a larger furnace mouth. The
tension of the gasest exerts an influence upon the chemical
In coal
processes in the different parts of the furnace.
* Bgwkfd., v., 193. B. u. h. Ztg., 1862, p. 812. Ann. d. min., 3 sér., xvi.,
254; xix., 167.
TUNNER, in LEOB., Jahrb., 1860, ix., 159, 287.
FUEL.
419
furnaces an enlarged furnace mouth, nearly as wide as the
upper part of the boshes, allows a more uniform sinking of
the ore charges in the upper part of the furnace. As the
coal is coked in the blast furnace under a high pressure, and
thus reduced in volume, the coke will form too thin a layer
if the furnace is much wider at the boshes than at the
mouth; the enlargement of the mouth causes the charges
to pass slowly, wherefore the carbonisation takes place more
gradually, yielding a greater quantity of coke.* The furnaces
of Truran* and Rachettet have their furnace mouths larger
than their boshes, and give favourable results with regard to
the carbonisation of raw fuel.
A. Wood.‡-Air-dried wood was first applied at Soumboul
in Russia, and later in some iron works of North America
and other countries. As the carbonisation of wood takes
place at the low temperature of 150° C., and is finished at
from 400° to 500° C., at which the reduction of oxide of iron
is scarcely commenced, the gases evolved by this carbonisation.
only exert a slight reducing action, and the chief saving upon
carbonisation in blast furnaces is caused by the greater yield
of charcoal, particularly if the furnace is sufficiently high
and wide. The slower process also causes the formation of
a more compact charcoal retaining the volume of the wood.
According to experience, from to of air-dried wood may
be added to one charge of charcoal, and of artificially dried
wood. When using the waste gases of such furnaces to heat
the blast, roast the ore, &c., the saving of fuel amounts to
from 20 to 25 per cent, from 5 to 10 per cent of which may
be attributed to the slower carbonisation and the assistance
in reducing the iron ore.
B. Turf|| has the disadvantage of containing more or less
ash and sulphur and phosphorus, besides losing much in
+
* TRURAN, The Manufacture of Iron, &c., London, 1862.
Ibid., 1862, pp. 265, 392; 1863, p. 156. ·
TUNNER, in LEOB. Jahrb., iv., 210.
Allgem., B. u. h. Ztg., 1862, p. 409.
Oesterr. Ztschr., 1861, No. 26.
|| B. u. h. Ztg., 1843., pp. 441, 832; 1845, p. 297; 1851, p. 427.
iii., 189; iv., 412; vi., 285; viii., 263. Oesterr. Ztschr., 1854, p. 399.
Jahrb., 1851, p. 128; 1854, p. 236; 1857, p. 129; 1861, xi., 45, 455·
Bgwkfd.,
LEOB.,
2 E 2
420
IRON.
volume. The latter disadvantage may be lessened by com-
pressing or artificially drying the turf. Turf is usually
employed in admixture with charcoal. At Ransko in Bohemia,
one charge consisted of 70 per cent of dried turf and 30 per
cent of charcoal, this being the highest proportion of turf
that has ever been employed.
C. Mineral Coal.-Mineral coal was first experimentally
applied to iron blast furnaces in 1619, by Dudley,* in some
iron works in Worcestershire, but only successfully introduced
after the year 1740.
The employment of raw coal in blast furnaces has certain
disadvantages, which cause coke to be more frequently used;
they are the following:-
1. The iron pyrites contained in the coal. Part of the
sulphur contained in the iron pyrites is evolved in the upper
parts of the blast furnace without doing any damage; but
another part is liberated in that part of the furnace where
the iron is reduced, and it combines with the reduced iron.
The remaining magnetic pyrites enters the slag, in which the
sulphur may be retained by an abundant addition of lime-
stone and the application of hot blast; the sulphide of
calcium formed also combines with some sulphide of iron.
The resulting pig-iron will be richer in sulphur the more iron
pyrites the coal contains. Therefore, coke is employed for
the production of the best quality of iron, an admixture of
coke and coal for iron of medium quality; whilst pig-iron pro-
duced exclusively with raw coal, as in Scotland, has little
tenacity, and yields an inferior wrought-iron unless the pig-
iron is previously refined.
2. The physical properties of the coal. Non-caking coal
poor in ash which does not splinter is best suited for this
Bituminous coal is unfit for direct use in blast
process.
furnaces, even if the furnace mouth is wider than the widest
part of the furnace shaft. But non-caking coal also requires
an enlarged furnace mouth to retard the carbonisation;
otherwise caked lumps of coal will be formed which burn
* Schles. Wochenschr., 1861. Beilage, p. 59. B. u. h. Ztg., 1861, pp. 6,
448; 1862, pp. 169, 240, 255; 1863, p. 156.
FUEL.
421
with more difficulty, causing scaffolds, explosions, &c. To
prevent the upper part of the furnace from becoming too hot
the furnace mouth is closed and the gases are made to escape.
Truran states that I part of good coal is equal to 3 parts
when converted into coke.
3. The slower passage of the charges required for the
more perfect caking of the coal.
Truran* states that a charge requires 35 hours for passing
through a Scotch blast furnace, 45 hours in a Staffordshire
furnace, and 63 hours in a furnace at Dowlais. The velocity
per hour in the different parts of the furnace stand in the
following proportions:-
Scotch furnace
Staffordshire
Dowlais
""
In the region of the furnace-
For preparing the For Carbonising
Ore for Reduction. the Reduced Iron.
Inches.
For
Fusion.
Inches.
28
48
Inches.
18
36
31
73
ΙΟ
41
The decrease of the production by the slower descent of
the charges is fully compensated for by the enlarged dimen-
sions of the furnace, which increase the pressure and quantity
of blast.
4. The lowering of the temperature of the upper part of
the furnace.
The advantageous application of raw coal therefore requires
the use of a non-caking coal as free from sulphur as possible,
a basic ore mixture, hot blast of strong pressure, compact
non-friable ores, and furnaces of wide dimensions. The ash
of non-caking coal is liable to fill the hearth, and its removal
is facilitated by constructing the furnace with an open front
wall.
As coke requires a different construction of the furnace
and a different conduct of the process, an addition of coke
to raw coal is rather injurious than otherwise, supposing the
process be carefully conducted. It may be advantageous in
cases where the process is conducted with less care.
•
In this country non-caking small coal unfit for coking is
*
Allgem., B. u. h. Ztg., 1863, p. 24.
422
IRON.
frequently used in blast furnaces, which, however, is perhaps
a disagreeable necessity rather than an advantage. Thirty
years ago, when still using coke and cold blast in blast fur-
naces, 10 tons of coal were consumed in this country for the
production of I ton of wrought-iron (about 6'15 tons for the
blast furnace, and 3'85 tons for the conversion into wrought-
iron). Since applying hot blast and raw coal partially, from
5 to 6 tons only are consumed, and this consumption is
further reduced to from 4 to 5 tons when making use of the
waste gases. But in spite of this saving and of other improve-
ments, such as an increase of the production, the use of
superior blast engines, lifts, &c., and the ample application
of puddling and other iron slags, the cost of production is at
present no lower than it was thirty years ago, owing to the
higher price of smelting materials and labour.
In Wales from 2 to 2'5 parts of coal are used for the pro-
duction of 1 of pig-iron when making use of the waste gases;
in Staffordshire, 4; in Cleveland, from 2.8 to 3; in Scotland,
from 24 to 2.6; in France, from 2 to 2'2 of good coal when.
smelting ores containing 40 per cent of iron, and from 4 to
5 of inferior coal when smelting poor ores. In rare cases
only less than 2 parts of coal are used for 1 part of pig-iron.
D. Anthracite.-Owing to its greater purity, anthracite
yields a better pig-iron than coke and raw coal; but as it
is difficult to burn it requires a hot blast of high pressure.
The difficult combustion causes a slower descent of the
charges, and therefore a small production, which may be
increased by enlarging the furnace and admitting a large
quantity of blast uniformly divided by a great number (up
to 15) of tuyeres. When applying cold blast the pressure
required would be so great as to cool the furnace. The
higher temperature and the application of an amount of blast
nearly three times as great as that used in coke furnaces
admits of a more perfect separation of the foreign sub-
stances contained in the smelting materials, thus producing
a superior quality of iron. The undesirable property which
anthracite has of splintering upon sudden heating is a con-
sequence of its stratification, and is peculiar to anthracite
occurring in vertical layers. The anthracite from Wales,
"
FUEL.
423
which has a granular fracture, is a conglomerate of very pure
anthracite and bituminous coal which does not splinter in
the fire.
Anthracite is used for the production of iron chiefly in
Pennsylvania, Scotland,† and Wales. ‡
+
In 1838 and 1839, Thomas made the first experiments on
smelting iron with anthracite at Pittsville, in Pennsylvania,
but he was unsuccessful, as the pressure of the blast was
insufficient. The second trials, made in 1840, were more
successful, and since that time anthracite has come into
more general use. Some of the anthracite furnaces are
60 feet high, having a furnace mouth of from 8 to 9 feet in
diameter; they have from 12 to 15 water tuyeres, and receive
10,000 cubic feet of blast per minute of 300° C. and upwards,
and of 8 or 9 lbs. of pressure per square inch; the weekly
production is 200 tons.
The furnaces in Yniscedwin are only from 25 to 30 feet,
as then the splintering of the anthracite into small pieces
does less harm. Other anthracite furnaces in South Wales
are from 36 to 40 feet high, and the charges are now and
then allowed to descend low, when the small pieces of an-
thracite are blown out of the furnace mouth by an increased
blast. The usual pressure of the blast is from 4 to 6 lbs.,
its temperature from 320 to 550° C., and its quantity
6000 cubic feet per minute. The weekly production is
So tons, and 2 parts of anthracite are consumed for every
part of pig-iron produced. Below the tymp of the furnace
an open space, from o'08 to o'10 metre high, is left, allowing
the ash to be continually blown out. Unconsumed pieces of
anthracite, intermixed with slag, which are discharged from
the tapping hole are re-charged in the furnace. In English
coke and coal furnaces the charges are distributed on the
periphery of the furnace shaft, thus leaving in the middle an
B. u. h. Ztg., 1849, p. 314; 1852, p. 755; 1853, p. 905; 1854, p. 149; 1862,
P. 344.
† Ibid., 1850, p. 673; 1862, p. 253. Allgem., B. u. h. Ztg., 1863, p. 157.
KARST., Arch., 2 R., xxv., 579, 601.
+
B. u. h. Ztg., 1862, p. 429. HARTMANN, Fortschr., i., 213.
|| B. u. h. Ztg., 1863, p. 156; 1862, pp. 237, 429.
424
IRON.
open conical space in which the ore will collect. In anthra-
cite furnaces the charges are made on the centre of the fur-
nace, thus falling towards the periphery, and forming a
thinner layer in the middle. This mode of charging some-
what prevents the disadvantages caused by the decrepitation
of the anthracite into small pieces.
With the view of saving compact fuel, several attempts†
have lately been made to introduce gases produced from in-
ferior fuel (carbonic oxide, carburetted hydrogen) into the
blast furnaces by means of the tuyeres, but no favourable
results were obtained, as it proved impossible to sustain the
high temperature required in the lower parts of the furnace,
and as the reduced iron was not sufficiently carbonised.
Gurlt's proposal‡ to reduce the ores in a cupola furnace by
means of gases, and to fuse the carbonised iron in a rever-
beratory furnace, has been successful only in Spain when
treating pure rich iron ores.
THE BLAST FURNACE AND ITS ACCESSORIES.
Lifts. In hilly countries the blast furnaces are often
placed below the general level of the ground supplying the
ores and fuel, so that all the materials necessary for working,
may be delivered at the furnace top without any special ap-
pliances, whilst on flat ground mechanical lifts must be
resorted to for raising the materials.
The following are the lifts most generally employed:
Vertical Lifts by means of Steam Power.||-The ap-
paratus consists of a cage moving between vertical guides,
similar to those used in colleries, and is connected with a
small direct acting steam engine of high pressure, which is
placed either on the platform surrounding the furnace top,
*B. u. h. Ztg., 1862, p. 429.
Berggeist, 1856, p. 97.
Oesterr. Ztg., 1856, pp. 137, 165, 195; 1858, pp.
245, 292, 310; 1859, p. 2. B. u. h. Ztg., 1856, pp. 223, 235; 1859, pp. 196,
286, 319, 320, 415, 420; 1860, p. 504. Vordernberg Jahrb., ii., 153. LEOB.,
Jahrb., 1860, ix., 317; 1861, X., 356. Bericht üb. d. erste allgem. Versamm. v.
B. u. Hüttenmännern in Wien., 1859, p. 54. HARTM., Fortschr., i., 229; ii., 160.
‡ GURLT, die Roheisenerzeugung mit Gas, &c., Freiberg, 1857. Berggeist,
1857, p. 606; 1859, p. 469. B. u. h. Ztg., 1857, p. 12; 1860, p. 27.
HARTMANN'S Fortschr., v., 98.
Allgem., B. u. h. Ztg., 1861, p. 252.
LIFTS.
425
or on the ground below.
It is advisable to have no self-
acting valve gear, but to work the engine entirely by hand,
in order to prevent the chance of accidents from over-winding.
These vertical lifts have to a great extent replaced
The Inclined Planes which are mostly to be found in old
works. An inclined plane is shown by Fig. 82. The two
FIG. 82.

P
P
M
L
W'
B
IT
N
L
W
t
inclined lines of railway, s and s', carry platform waggons,
w and w', which are connected with each other by means of
a chain wound round a drum. The motive power is a steam
engine placed on the ground.
The inclination of these planes varies from 25 to 45°; the
platform waggons are sufficiently large to carry four or more
of the iron wheel-barrows used in charging, with their
loads.
At the Barrow Iron Works, in Lancashire, two inclined
planes are used for the supply of seven furnaces.
426
IRON.
The Jacob's Ladder, or Endless Chain System of
Lifts, requires much attention, and necessitates many re-
pairs, and is therefore but little used. Fig. 83 shows one
FIG. 83.

α
g
of these lifts. The chains, a, pass on the top and bottom
over wheels, and are provided with hooks, b, by means of
which they lift the barrows, g.
The lift represented by Fig. 84 is provided with cages of
FIG. 84.

b
a
iron plate, b, which are fixed to the chains, a, at certain dis-
tances, and which carry the barrows used in charging.
The Water Balance is frequently used in this country as
well as on the Continent, as being very simple and effective,
LIFTS.
427
and requiring very little attendance; though, on the other
hand, somewhat more power is required for pumping the
water to the furnace top than would be necessitated by a
direct lifting of the charges; also it is difficult to keep the
water boxes tight, and the lift houses are generally damp
and sloppy from leakage.
Fig. 85 represents one of these lifts.
a is a wooden box
Ъ
FIG. 85.

F
3
a
5
FI
which may be filled with water from the reservoir, b, placed
on the platform of the furnace top by means of the tube, d,
which is provided with the cock, c; e is a waggon used in
charging, placed on rails which are fixed on the top of the
box, a.
As soon as the box, a, is filled with a certain quan-
tity of water it will sink by its own weight together with the
empty waggon, e, being guided by f. On the side of this
apparatus is placed a second one of the same construction,
moving vertically and guided, and both are united by a rope
or chain, g', passing over a guide pulley. When the one
box filled with water, and carrying an empty waggon,
descends, the second without water and carrying a loaded
waggon will be lifted up. The loaded waggon is discharged
on the top of the furnace, and the empty box filled with
428
IRON.
water, whilst the lower box, on reaching the ground, strikes
against a catch with the projecting rod of the discharge
valve, h, thus opening a passage for the water which runs.
out, and the box is ready for another ascent when loaded.
The speed of the apparatus is regulated by a brake on the
guide pulley.
The lift constructed by John Fernie, and employed in the
Britannia foundry at Derby, is represented by Fig. 86. a is
FIG. 86.

g
h
I
d
12
6 0
INS.Lulluluk
2-
3
FI
a steam cylinder provided with the piston, b, of 3 feet stroke ;
c is an iron tube admitting steam above the piston. The
water in the cylinder, a, below the piston is emitted and
conducted by means of the tube, d, in the cylinder, h, below
the piston, e. To the bottom of the latter piston is screwed
a piece of leather serving as packing. The piston rod, ƒ, is
made of a wrought-iron tube of an inch thick in order that
it may be light, and on the top it is provided with a platform
consisting of the iron plate, g, for carrying the loaded
vessels.
The lid of the cylinder, a, is placed in the level of the sole
LIFTS.
429
of the foundry; the cylinder, h, is placed in a shaft, and its
upper point lies about I foot below the sole of the foundry.
The plate, g, is guided by the rods, i, and furnished with
cushions of caoutchouc for lessening the shocks.
To set the apparatus in motion, the tube, d, is filled with
water until the piston in the cylinder, a, is placed in its
highest position. The vessel used in charging is now placed
upon the iron plate, g, and steam is introduced upon the
piston by the tube, c, thus causing the ascent of the piston, f,
which takes 20 seconds; the piston is 10 feet long, and the
weight upon it amounts to 9 cwts. The piston is made to
descend in 30 seconds by allowing the steam to escape
through a cock with which the tube, c, is furnished.
At Johannishütte, near Duisburg, Westphalia, hydraulic
lifts are employed, constructed upon Armstrong's system, in
which the cage is connected to a water pressure engine by
means of a chain passing over a system of compound pulleys.
Pneumatic Lifts were first used in this country, and intro-
duced later into Germany. One of these lifts used at
Friedenshütte in Upper Silesia has the following construction,
which is shown in Fig. 87.
The blast is introduced upon the piston, b, by means of
the tube, a, which piston then descends in the cast-iron
cylinder, c. The piston rod, d, and the adjoined rod, e, move
along with the piston; the latter rod sets two cog wheels in
motion, thus causing the ascent of a platform upon which
the vessels used in charging are placed. In order to lower
the discharged vessels the valve, h, is opened by means of
the rope, g, which is fixed to the windlass, f; the speed is
regulated by the valve.
This apparatus lifts altogether 643 kilos., the platform and
empty vessel weighing 143 kilos. The ascent is completed
in 2 minutes, and as much time is required for the descent,
discharging and charging of the vessels.
The simplest form of pneumatic lift is that used in
England; it consists of a wrought-iron cylinder, open at the
bottom and closed at the top, about 6 or 8 feet in diameter,
BAUERMAN'S Treatise on the Metallurgy of Iron, London, 1868, p. 150.
430
IRON.
and somewhat longer than the height of the furnace; it is
suspended in a tank by counter-balance weights passing over
pulleys in exactly the same manner as an ordinary gasometer.
FIG. 87.

e
с
đ
α
A pipe for the admission of air at 3 or 4 lbs. per square inch
above the atmospheric pressure is introduced through the
tank. The waggon to be lifted is carried on the top of the
LIFTS.
431
bell, and as the whole of the moving parts of the apparatus
are balanced, sufficient power only is required to raise the
additional load. For the return of the stroke, the air within.
the bell is allowed to escape by opening a valve communicating
with the atmosphere, the weight of the empty waggon being
sufficient to lower the bell in the tank.
Mr. Gjer's pneumatic lifts are much used in the newer
Cleveland Iron Works.* The lifts used at Messrs. Lloyd's
furnaces are constructed as follows:-The two lifts to the tops.
of the four furnaces, 68 feet high, consist each of a column
bored out to a diameter of 36 inches. Within this is a deep
cast-iron piston, almost fitting the bore, and weighing nearly
four tons. The platform surrounds the column, which is
secured to the stage at the top and forms the sole framing of
the hoist. The platform has a wire rope at each corner,
which passes over an 8 ft. pulley at the top-there being
four ropes and four pulleys-and thence down to the piston.
When the platform is loaded it weighs about 5 tons, or a ton
more than the piston; and when the empty barrows only are
coming down it weighs about 3 tons, or a ton less than the
piston. A pair of engines close by work a pair of air pumps,
which may be made, by a reversing lever, to act as forcing
or exhausting pumps. With 3 or 4 lbs. exhaustion per square
inch, the piston is brought down and the load raised; with
the same amount of pressure above the atmosphere forced
in beneath the piston, it is sent up again, and the empty
platform descends. The power is applied very gradually in
raising the platform, as the engine has first to exhaust from a
3 feet column nearly 70 feet high, and, in any case, no sudden
strain is brought upon the ropes. The platform rises and
falls at the rate of 250 or 300 feet per minute; the motion is
perfectly regular, so that, save for the unavoidable sensation
of rising or sinking, any one upon the platform would be
quite unconscious of the motion. The platform descends to
fixed supports at the bottom and rests upon them. When
the platform reaches the top, the piston reaches the bottom
of the column upon which it rests, preventing any further
Engineering, Nov. 30, 1866; Dec. 7, 1866.
:
432
IRON.
motion, and the piston is made to cushion in the air as it
reaches the bottom, so as to avoid all shock in stopping.
Blast Furnaces.
A furnace for the production of pig-iron from iron ores
must be able gradually to perform the following operations:-
namely, a preparation and reduction of the oxide of iron at
a temperature below the melting point; next, the carbonisa-
tion of the reduced iron at an increased temperature but still
without fusion; then the melting of the slag-forming com-
ponents and the pig-iron; and, finally, the separation of slag
and iron according to their specific gravities in a sufficiently
hot space.
These operations can only be performed in cupola furnaces,
and not in reverberatory or hearth furnaces.
The first treatment* of iron ores, chiefly pure spathic ores,
for the production of iron was effected in small hearths
similar to a common smith's forge, in which alternate
layers of crushed iron ore and charcoal were smelted by the
application of blast from bellows until a sufficiently large
lump of iron had collected on the bottom of the hearth, but
the resulting iron was so little carbonised as to be malleable.
In order to save fuel and to obtain a better yield of iron,
these hearths were transformed later into low cupola furnaces.
(Stücköfen, Wolfsöfen), which produced a steel-like iron,
containing more carbon, but not fusible.
About the middle of the 16th century the furnaces were
enlarged, so as to be upwards of 15 feet high, when they
produced a more highly carbonised and fusible iron (pig-iron).
The furnaces at present in common use are sometimes
upwards of 60 feet high, and are constructed with an open
front wall, from the opening of which the slag continually
runs out. In a few places the present blast furnaces of the
smaller size, and intended for a smaller production, are con-
structed with a closed front wall, requiring the slag to be
tapped off from time to time together with the iron.
*
Ann. d. min.,
Berggeist, 1860, No. 15. B. u. h. Ztg., 1860, p. 502.
livr., de 1858, p. 564. Jahrb. d. k. k. Geol. Reichsanst., 1850, i., 199.
CONSTRUCTION OF THE FURNACES.
433
Construction of the Furnaces.
One of the chief requirements in the construction of blast
furnaces is the counteraction of the corroding effect of the high
temperature, which both fuses and expands the furnace
walling.
The fusion of the furnace lining may be partly prevented
by applying materials as refractory as possible; artificial
stones are generally more refractory than native rocks. The
outside of the lower parts of the furnace is usually furnished
with cast-iron boxes or blocks having a serpentine wrought-
iron pipe inside, through which a current of water is kept
flowing in order to protect the brick-work from the intense
heat to which it is exposed, and from the corrosive action of
the molten slag which is constantly flowing through it.
The expansion of the furnace walling may be limited by a
suitable exterior form of the furnace, and by grappling irons.
A suitable connection of the interior furnace lining with the
exterior walling also greatly influences the expansion and
destruction. Furnaces having the hearth and boshes without
an inclosing walling have proved to be very durable, and it is
advisable to fill the open space between the inside lining of
the furnace shaft and the outside walling with a pumice-like
slag, as it permits the expansion of the inside lining better
than sand, and as slag, if used, will cake and close any
opening.
The construction of blast furnaces varies considerably with
regard to size, shape, the proportion of the parts to each
other, &c.
1. The oldest furnaces are shaped outside like a truncated
four-sided pyramid; they are still in use to a great extent
both in this country and on the Continent, where they are
known as Belgian furnaces. Nearly all charcoal furnaces
are of this construction. These furnaces are shown by Figs.
87 to 96.
a is the foundation formed of concrete masonry of quarry
stones; b is a channel formed of fire bricks, and serving to
dry the foundation and the lower parts of the furnace; c, the
VOL. II.
2 F
434
IRON.
fire-place; d is a cross channel in the foundation, called
Andreaskreuz, connected with the flues, f, f, by means of the
channels, e, e, and is covered with the iron plates, g, upon
which the brick walling, h, rests; i is the bottom stone (of
pudding stone of Marchin), resting upon a layer of sand; k is
FIG. 88.

ШИ
2005 0 0 52002001a6a1doäämoäoco0 05020 1000 ORSO)
다
​n'
11
*
1000000
༡
1
the lining of the furnace hearth (B, F), also formed of pudding
stone, and containing the tuyere openings, r', r'; l are the
boshes, the lower parts of which (G and H) are also formed
of pudding stones and the upper part of fire bricks; m is an
CONSTRUCTION OF THE FURNACES.
435
open space between the furnace linings, and filled with pieces.
of fire-brick; n is the lining of the furnace shaft, formed of
fire bricks; o is the second lining placed outside the first, and
formed of fire-bricks of second quality; this lining is sur-
rounded inside and outside by the filling, m; p is the exterior
FIG. 89.

q
P'
G'
F.-.-
-T'
Y
E
К
--E
D-
-D'
a
casing formed of common bricks, except the facing bricks, 9,
which are fire bricks; r are cast-iron bearings in the
quadratic form resting upon the annular iron, u, to which
they are joined; s are pillars of masonry; t are iron beams
and alternate layers of fire-bricks; u, annular iron bearings
2 F 2
436
IRON.
supporting the inner lining of the furnace shaft; the iron
bearings, v, serve the same purpose; w are likewise iron
bearings; x is the arch in the front of the furnace; y are the
arches for the tuyeres; z, a lining of fire bricks. a' is the
tymp stone, its upper part being pudding stone, and protected
by the iron plate, b', which consists of two parts screwed
together, and its lower part of large fire bricks protected by
FIG. 90.

น
f
7
Ma
M
で
​V
f
the tymp-iron, c'; d' is a wall of pudding stone called the
dam; it is provided with the tapping hole, ƒ', and its front
part is protected by the dam plate, e'; g' is a cast-iron plate
placed vertically towards the dam plate, and fixed to the
masonry by means of the iron rods, h'; i' are openings in this
plate for receiving an iron bar, serving as a support for the
tools used for manipulating in the furnace hearth. The
space between this plate and the furnace pillar is gently
CONSTRUCTION OF THE FURNACES.
437
b
FIG. 91.

GFEA
B
M
f
sloped, and is filled with small coal in admixture with some
loam, which is called the cinder fall, as the slag is made to
run off upon it. k' are cast-iron pilasters partly supporting
FIG. 92.

A
*
C
だ
​во
.
a
ធ
口
​ם
ם
~
D
О
f
ㅁㅁ
​FIG. 93.

2
3 F
FT
lí
Scale to Figs. 92 and 93.
the bearings, and provided with several openings, l', for re-
ceiving iron rods, which are used for supporting the tools
438
IRON.
when working in the hearth, &c.; m' are flues in the rough
walling for the escape of moisture. They commence in the
arches, forming horizontal rows, one about 1 feet above the
other, and are connected with the flues, f, without openings
on the outside of the furnace. n' and o' are grappling irons.
FIG. 94.

ས་
ட்ட 5
10 15
20
25
Scale to Figs. 81 to 91 and 94 to 97.
placed in channels of the furnace walling; p' is a cylindrical
chimney, termed the tunnel head, the inside of which is lined
with fire bricks, whilst the outside is formed of common
bricks. It is furnished with grappling irons and provided
FIG. 95.

Y
A.
歐
​F
Y
A
d.
T
$
71
with rectangular openings, q', or filling holes. These holes
have projecting parts in their corners to prevent the charging
barrows from falling into the furnace. s' is the platform on
the furnace head.
CONSTRUCTION OF THE FURNACES.
439
2. A more modern construction of the blast furnace is
shown by Figs. 98 and 99, in which the rough walling
is replaced by a wrought-iron casing, and the old stake
pillars by cast-iron columns, thus exposing the hearth casing
all round, whilst in the old furnaces it is accessible at the
tuyeres only. The outer form of these furnaces is that of a
cylinder or a truncated cone.
FIG. 96.

У
น
AX
4
it
Y
I
Being of lighter construction, this furnace requires less
foundation; it dries easily, is less modified by expansion, and
is erected more speedily and cheaply. Also the hearth casing
expands less owing to its free position, which allows a con-
FIG. 97.

0'0
tinual cooling of its outside. It requires no grappling irons,
and may be easily repaired during the operation; it also
allows the application of many tuyeres, even at different
440
IRON.
levels. By surrounding the hearth casing on its outside with
hollow iron rods through which cold water passes, a still more
perfect cooling is obtained.
FIG. 98.

જ
S
た
​17:
ů
a
The furnace represented is erected in Hasslinghausen
(Westphalia), and is of the following construction :-
FIG. 99.

M
A
ՊՆ
C
B
6
i...i
12
24
FI
CONSTRUCTION OF THE FURNACES.
441
a is the foundation; b, bottom stone of the furnace ; c, the
dam; d, six openings for the tuyeres; e, the hearth casing;
f, the boshes; g, the inner furnace lining; h, the filling;
¿, outside lining; k is the iron casing of iron plates; l, cast-
iron pillars resting upon the foundation, m, and supporting
the linings of the furnace shaft by means of the cast-iron
rim, "; o, the platform on the furnace top resting upon the
linings and the supports, p; r is the furnace chimney, fur-
nished with the charging openings, s.
The newer furnaces at Ulverstone* have no filling between
the furnace linings, but an open space filled with sand is left
between the lining and the iron casing.
This class of furnace is known as the cupola furnace from
their resemblance to ironfounder's furnaces.
A third kindt of blast furnace is a combination of both the
furnaces 1 and 2, the furnace shaft linings being inclosed in
a large rough walling resting upon pillars of masonry, whilst
the hearth casing stands unattached to the pillars, but free
like the casing on the second kind of furnaces.
Concerning the erection of blast furnaces, the following
general rules may be given :-
1. The foundation is usually as broad as the furnace is
high; two-thirds of this breadth are sufficient if the furnaces.
are very high, and cupola furnaces require only half of the
breadth. If possible, the foundation is made on rocky ground;
soft ground necessitates the erection of a groined vault or
of pile work, upon which the foundation of the furnace is
placed. The foundation is always provided with channels
for drying and for the escape of moisture, more or less like
those shown in the furnace represented by Figs. 88 to 97.
Upon the foundation are now erected-
2. The corner pillars, s, up to the height where the arches
y and x are intended to commence. The tuyere niches or
tuyere houses are usually connected with each other by
passages through the pillars. The iron plates, u (Fig. 96),
Allgem., B. u. h. Ztg., 1863, p. 185. B. u. h. Ztg., 1862, p. 379. TUNNER'S
Bericht üb. d. Londoner Industrie-Ausstellung, in 1862, p. 31.
B. u. h. Ztg., 1862, p. 85.
442
IRON.
are then placed upon the pillars, and connected with strong
hollow iron beams, ; upon these, at t, three layers of fire-
bricks are placed; next iron beams braced together; again
three layers of bricks and iron beams on the top; and,
finally, three layers of bricks, which are covered with the
annular iron, v. The arches, x and y, are then formed either
by the bearings, w, or of masonry without bearings. These
arches must be high enough for the workmen to stand in
them. They are generally 13 feet high, 15 feet wide on the
outside, and from 9 to 9 feet wide inside; the arches are
convex by 20 or 22 inches. Above the iron beams, v, the
furnace is round instead of square. The layer of fire-bricks, t',
is placed upon the lowest iron bearing; and now
3. The rough walling, or outside casing, þ, is walled up to
the furnace top, care being taken in the construction of
the different channels and flues, f and m, as well as of the
grappling irons, n' and o'. From the point where the inner
side of the rough walling takes a conical shape, a fixed
templet, movable on its axis, is employed to obtain the proper
form.
4. The chimney, or tunnel head serving as a protection
against the wind, and about 9 or 10 feet high, is usually
erected when the inner linings of the furnace are finished.
The charging opening, q', of the tunnel head are usually
provided with doors. The tunnel head is constructed
either of iron plate lined with bricks or of masonry only.
As the tops of coke blast furnaces are not usually large
enough for the workmen to work upon, they are enlarged by
projecting platforms. Furnaces lying beside each other are
connected by bridges.
After having thoroughly dried the rough walling of the
furnace, which sometimes takes some months, the inner
shaft linings are walled up. In some iron works, one lining
separated from the rough walling by an open space is found
sufficient; whilst in other works two linings are employed.
Dry pieces of brick, pumice-like slag, &c., are preferable to
sand and clay for filling the open space between the inner
furnace walls or linings, as they allow for expansion, and
thus prevent a cracking of the walls. These walls are
CONSTRUCTION OF THE FURNACES.
443
built up, whilst their shape is regulated by the templet, u',
(Fig. 90).
The lower part of the furnace is formed either of natural
minerals or artificial fire bricks; it is sometimes composed
of one mass of a mixture of clay and sand.
A layer of sand or loam, 2 inches thick, is beaten down
upon the brick walling of the foundation, or sometimes three
layers of fire bricks and one layer of sand 1 inch thick; upon
When employing
this the bottom stone, i, is placed.
natural minerals (pudding-stone, sandstone, &c.), the bottom
stone is formed of one piece from 1 to 3 feet thick; when
employing artificial stones, and of a roundish form, they are
joined in concentric rings, and bound together with iron
hoops. This artificial bottom stone is 10 feet in diameter
and 3 feet thick, and prolonged towards the front of the
furnace, so as also to support the dam. The bottom stone
must be placed sufficiently high to allow the liquid iron
to run out to the sole of the smelting works, even after
it has been partly corroded. The bottom stone is sometimes
surrounded by a wall of fire bricks, and separated from it by
an open space, which is filled like those spaces of the
furnace shaft, except the front part, which is kept open.
Upon the back and sides of the bottom stone the hearth
casing, k, is erected, forming either a circle or an oblong,
whilst the lining z, is walled up outside, and the space between
it and the hearth casing is filled up with non-conducting sub-
stances. The joints between the stones forming the hearth
casing are carefully filled with refractory mortar; upon this
hearth casing the stones forming the boshes, l, are placed
with the same care.
Blast furnaces with an open front wall are now provided
with the tymp stones, a' and a.* The iron plate placed in
their front is sometimes hollow, and a stream of water passes
through it; the other parts of the hearth casing are also
cooled with water if they stand isolated from the furnace
pillars.
KARST., Arch., 2 R., iv., 419; v., 508; viii., 191. Karst. Eisenhüttenkunde,
v., 86.
444
IRON.
The Swedish furnaces have a hollow piece of iron casting
or wrought-iron, instead of the lower tymp stone, which is
cooled either with water or with air.
The lower hearth is then closed by the dam, d', after
having warmed the shaft linings and hearth casing. The
dam is provided with the tapping hole, f', 3 inches broad and
4 inches high, and protected outside by a cast-iron plate.
Dams which are flat at the top are also covered with an
iron plate.
Furnaces (chiefly charcoal furnaces) used for the produc-
tion of foundry iron, which is ladled out for the purpose of
casting, are sometimes provided with a second fore-hearth,
as shown in Fig 100. a, is the dam; b, the fore-hearth,
communicating with the second fore-hearth, d, by means of
the channel, c; e is the tymp stone; f, the inner furnace
hearth.
FIG. 100.

Ъ
Ranch germany
f
a
གཏམ་ཚགས་
73
d
Some of the smaller blast furnaces are constructed with
a closed front wall (Blauöfen) as shown in Figs. 101 and 102,
The stone, a, below the tymp, is provided with the tapping
hole, b, and the two tapping holes, c, c; from the latter the
slag will run out, whilst the iron is tapped off through b.
The hole, d, is opened during night time to make use of the
flame coming out of it. e is the tuyere.
The hearth casing, formed of an admixture of clay, sand,
&c., has several advantages, and is frequently used where
good refractory stones are not easily obtained. This casing
is not very liable to melt, decrepitate, contract, &c.; they
CONSTRUCTION OF THE FURNACES.
445
admit of any shape, which with stone is more expensive;
and as they are more uniformly used up, the hearth may be
constructed of wider dimensions from the very commence-
FIG. IOI.

f
ཁག་ལས་ བ་ མ་
།
**
b
ས་མ་
.
༤༥ཟར་
Li
5
FIG. 102.
10
FI

CL
446
IRON.
ment, thus economising fuel and more quickly admitting of
the highest ore charges; this last circumstance outbalances the
disadvantage of the formation of these casings, requiring a
longer time for construction, which causes the furnaces to
be out of blast for a longer time.
Figs. 103 and 104 represent the hearth casing. The
mass, a, for the inner parts of the furnace, consists of 4
volumes of quartz sand and I volume of clay; and for the
outer parts, of 5 volumes of quartz to I volume of clay; b, is
FIG. 103.

-
do
k
the rough walling; c, the foundation; d, the tuyere; e, the
lower part of the hearth; f, the dam; g, the tymp iron,
FIG. 104.

C
a
CONSTRUCTION OF THE FURNACES.
447
supporting the tymp plate, h; i, plates for the protection of
the hearth opening; k, are iron bars supporting the mass
behind the tymp iron; i, sandstone for fixing the tuyere.
3
4
When constructing a furnace hearth in this manner, the
components of the mass* are mixed and moistened, so as just
to form into balls when pressed; the mass is then mixed
with some dry material, and a layer 5 inches high is placed
on the foundation and stamped, first with the feet and
afterwards with warmed iron stampers, until it forms a
layer or I inch thick. The surface of this layer is
roughened, and some fresh mass beaten on it till the bottom
stone is sufficiently high. After planing, its centre is as-
certained by means of a plummet, and a model consisting
of single boards, and of the, shape of the interior of the
hearth, is placed on the bottom stone, and the mass is beaten
down round it, thus forming the hearth lining, boshes, and
tymps.
The mass for the bottom stone used at Königshütte
(Silesia), is composed of two parts clay and one part
ground fire bricks, and the mass forming the side parts of
the hearth of one part clay and two parts fire bricks.
The walling of the hearth may be preserved by employing
very refractory building materials, by cooling it on the outer
parts, and by smelting a suitable ore mixture. An acid
mixture requires a siliceous material, and a basic mixture
materials poor in silica. A refractory mixture smelted with
cold blast preserves the hearth walling, whilst easily fusible
ores smelted with hot blast attack it strongly.
The bottom stone, dam, and tymp are the most easily
corroded.
Collecting the Waste Gases.-In most iron works these
gases are now collected and used for various purposes.
The apparatus for collecting the gases must be constructed
so as to allow a perfect collection of the gases without
impeding the process of the blast furnace, and also the
LE BLANC, Eisenhüttenkunde, Deutsch v. Hartmann, 1839, i., 37. Preuss.
Ztschr., ii., 126. B. u. h. Ztg., 1860, 392; 1861, Taf. 12, Fig. 1 and 2. HARTM.,
Fortschr., v. 97. STUNKEL, Eisenbergwerke, u. Eisenhütten am Harz., 1803,
p. 125.
448
IRON.
apparatus must allow an appropriate mode of charging the
furnace. The materials must be charged on the periphery of
the furnace shaft, where the finer parts of the fuel and ore will
then remain, whilst the coarse pieces fall towards the centre.
It is furthermore necessary to collect the gases in the upper
part of the furnace throat when they have passed the whole
height of the furnace. Though these gases have less heating
power than gases collected at a lower height of the furnace,
it is more profitable to save as much fuel as possible in
the blast furnaces than to impair them by extracting more
effective gases, particularly as the objects for which waste
gases are applied may be obtained by an inferior fuel, and,
if required, inferior fuel and waste gases may be employed
at the same time.
The smaller charcoal furnaces yielding a large quantity of
reducing and carbonising gases and working with an open
throat, are provided with simple apparatus, almost all of
which answer equally well.
In a great many iron works the gases are collected as
shown in Fig. 105. A cylinder, a, of cast- or wrought-iron
and from 4 to 7 feet high is suspended in the furnace shaft,
FIG. 105.

f
L
Ap
C
Ъ
a
Ъ
d
leaving a space, b, from 4 to 9 inches wide, between it and
the furnace walling, and in which the gases collect. The
CONSTRUCTION OF THE FURNACES.
449
tube, c, takes off the gases, and the surplus gases escape
through the tube, d; e is a sliding valve for regulating the
gases in the tube, c; f is a safety valve. A lid serves to
close the furnace mouth.
At the iron works in Siegen and in the Upper Hartz the
gases are taken off the furnace by means of small channels in
FIG. 106.

Ъ
α
C
Ъ
the furnace walling, as is shown in Figs. 106 and 107. a is a
cast-iron ring 8 inches broad, 4 inches high, and provided
FIG. 107.

-
α
C
டம்
6
9
12
J FI
with the projection, b, 8 inches broad, and 2 or 3 inches
high, and furnished with the openings, c; d is the annular
space for receiving the gases; e, gas pipe; f are cast-iron
plates. This arrangement is simpler, more durable, and only
half as costly as that with a suspended cylinder.
Amongst the many apparatus which have been proposed and
sometimes used for collecting the whole of waste gases, and
which, therefore, necessitate a closing of the furnace throat,
the following are those which have given good results:-
VOL. II.
2 G
450
IRON.
Fig. 108 shows the cup and cone charger first applied by
Mr. Parry in Ebbw Vale, and is that most generally used.
a is an inverted conical cast-iron funnel fixed to the top of
the furnace. Below this funnel or cup an upright cast-iron
cone, b, is placed in the furnace and suspended to a chain, c,
FIG. 109.
FIG. IIO.
FIG. III.
1
t
!
1
3
}
i2'
13- 3
11.6
79-
FIG. 112.

20-1
2=60
8
8
20-
FIG. 108.
attached to its apex so that it may be raised and lowered at
pleasure by means of a counter-weight or a lever. When
in a raised position the cone bears against the bottom of the
funnel and forms an air-tight stopper, thus preventing the
gases from escaping through the furnace mouth, and making
them pass through the adjoined pipe. The charges are made
upon the cone, and they drop into the furnace towards its cir-
cumference, whence the large lumps roll towards the centre.
APPARATUS FOR COLLECTING THE WASTE GASES. 451
Von Hoff* has modified this apparatus by fixing the gas
tube to the apex of the movable cone, and has provided
the cone with lids, at the same time keeping the joints tight
by means of water to prevent the escape of gases.
At Siegburg on the Rhine, Langen has constructed the
following apparatus:-The furnace mouth is closed with a
lid in the form of a bell, which may be raised and lowered
at pleasure for the purpose of charging the furnace. The
gas pipe is joined to the lid in such a manner that the lid slides
on the outside of the tube when being raised. This vertical
gas pipe is connected with a horizontal pipe, and both are
provided with safety valves. All the joints are made air-
tight by water.
At Hörde, Langen's apparatust has been improved as
follows:-
The furnace mouth, 9 feet wide, is closed with a flat lid,
which cannot be raised but may be turned on rollers, and is
kept tight by water; a gas pipe, 3 feet wide, is fixed to the
lid and kept tight in the same way. The lid is also provided
on its circumference with four openings, which are closed by
valves and kept tight with water. The charging of the furnace
is effected in quick succession by these four openings, and
before re-charging, the furnace lid is turned round 45,
thus uniformly dividing the materials on the circumference
of the furnace.
Whilst the waste gases of charcoal furnaces, when made use.
of at the level of the furnace mouth, have always greatly
economised fuel without impairing the smelting process, the
use of the gases of coke furnaces has entailed great incon-
venience, as they are mostly used at the level of the furnace
sole, and must, therefore, be sucked down by means of a
high chimney. This exhaustion of the gases is liable to raise
the smelting zone of the furnace to a higher level; con-
sequently, the ore charges pass the furnace without being
sufficiently reduced and carbonised, and the regular process.
ESCALLE, in Bullet. de la Soc. de l' ind. minér. ix., Sr. B. u. h. Ztg., 1864,
P. 131.
† DINGL., Bd. 165, p. 25. TUNNER'S Bericht über d. Londoner Industrie-
Ausstellung, in 1862, p. 38.
2 G 2
452
IRON.
can only be sustained by a larger consumption of coke; some
atmospheric air is occasionally sucked along with the gases
and causes explosions. Waste gases must be taken off the
furnace with a certain pressure in order to be profitably used;
they must not, therefore, be exhausted, but pressed. The
required pressure cannot be obtained if the furnace mouth is
too wide, therefore narrower throats have lately been employed.
In order to deposit the dust escaping along with the gases,
the pipe which takes them off should be vertical, or frequently
cleared if it should be horizontal; it is also beneficial to
conduct the gases over a long surface of water, thus con-
densing the steam contained in the gases, and depositing the
furnace smoke.
Experience has proved that all apparatus* for collecting
the waste gases are useful, whether they take the gases
from the centre, the circumference, or the whole surface of
the furnace throat; provided the apparatus are rationally
constructed, the gases of a certain pressure and the charging
of the furnace is done in such a manner that the larger lumps
are placed in the centre, and the finer parts on the circum-
ference of the furnace shaft.
Relations between the Different Parts of Iron Blast
Furnaces and the Circumstances Influencing these
Relations.
The Interior Form of Iron Blast Furnaces varies in
different countries, and even in every iron district; it has
no fixed rule, although the form undoubtedly influences
the consumption of fuel and the quality of the resulting
iron.
Without taking into consideration the state of aggre-
gation of the ore, two different forms of blast furnaces have
been adopted, one for ores which are easy to fuse, and one
* B. u. h. Ztg., 1858, pp. 37, 243; 1859, pp. 107, 400; 1860, pp. 298, 387;
HARTM., Fortschr., ii.,
1861, pp. 101, 404; 1862, pp. 245, 264; 1863, p. 211.
187; iv., 90; v., 160, 165. DINGL., Bd. 161, p. 355; Bd. 165, p. 25. Berggeist,
1861, p. 520.
Bericht. üb. d. 1 Vers. d. Berg. u. Hüttenm. in Wien., 1859,
p. 67.
THE SHAPE OF THE FURNACE SHAFT.
453
for more refractory ores. These forms are founded on rational
principles, whilst experience has led to the construction of
newer blast furnaces without strongly projecting boshes,
thus facilitating a more uniform descent of the smelting
materials.
1. The shape of the furnace shaft for easily fusible ore
mixtures, shown in Figs. 109 and 110, is described by
Tunner* as having a proportionally wide hearth, narrow
furnace mouth, and narrow belly (Kohlensack, i.e., the part
just above the boshes). Easily fusible ores, when intended
for the production of white pig-iron poor in silicon, require
the application of a low temperature, and, therefore, a wider
hearth. As the quick smelting of the easily fusible ore
mixture renders much of the heat latent, and since, as we have
stated, the temperature of the hearth is not very high, the
temperature required for reduction and carbonisation in the
upper parts of the furnace hearth is obtained by placing the
narrow belly on a lower point of the furnace, and by making
the furnace mouth proportionally narrow. The narrow fur-
nace mouth also increases the pressure of the ascending
gases and the disintegration of the smelting materials, thus
facilitating the reduction and carbonisation which the quick
descent of the charges renders necessary. On the other
hand, the narrow furnace mouth causes a less uniform descent
of the charges. When smelting easily fusible ores in fur-
naces with a narrow hearth, a fusion of the ores takes place
before they are sufficiently reduced and carbonised, and iron
is scorified at the same time; if attempts are made to prevent
this by charging ore and fuel in a suitable proportion, grey
pig-iron is produced instead of white iron.
The drawings 109 and 110 represent blast furnaces with
closed front wall (Blauöfen) in Austria; this form is also
given to furnaces in this country when producing grey forge
iron from clay ironstone, which is easy to smelt and reduce;
these furnaces, which at the same time have the advantage
of wider throats, are represented by Figs. III to 117. Fur-
naces treating somewhat less fusible iron ores have the belly
LEOB., Jahrb., 1860, p. 151.
454
IRON.
at a higher level and also a narrower hearth, as shown in
some of the drawings. Blackbands for the production of
5:0-
FIG. 113.
-90-
15.2.
7
4
17.4.
#
FIG. 114.
FIG. 115.
FIG. 116.

6
9.6.
0165
16
6
-10
12.6
9 6
darkish grey iron with the application of very hot blast are
also treated in furnaces of a similar shape, as shown in
Fig. 118 and in Fig. 98 on page 440. The application of
hot blast allows a wider hearth, and a proportionate enlarge-
ment of the throat is admissible if treating ores of easy
reducibility.
In furnaces of equal capacity with hearths and boshes of
the same height, and when smelting the same ore mixture, the
temperature in similar levels will stand in about the same
proportion as the furnace section of those levels.
2. The form of furnaces for ore mixtures difficult to fuse.
The furnaces most used for treating these ore mixtures for
the production of grey pig-iron have a narrow hearth, a wide
belly, and a wide throat. The smelting of these mixtures
requires a concentration of the heat in a narrow hearth,
whence the formed gases may ascend with so high a tem-
perature that the smelting mass in the space above the hearth
THE SHAPE OF THE FURNACE SHAFT.
455
would cake before being perfectly carbonised, if the belly were
not enlarged so as to distribute the heat more. As a mixture.
difficult to fuse descends more slowly in the furnace than a
mixture of easy fusibility, it has more time to be reduced
FIG. 117.
FIG. 118.
FIG. 119.

< 7.3 >
TH
20
32.9
O
and carbonised, and requires less pressure of the gases, which
escape at a higher temperature, consequently the furnace
throat may be wider. If the furnace mouth is too narrow
the gases in the lower part of the furnace will become of too
great a pressure and impede the combustion, whilst white
pig-iron is produced instead of grey.
Whilst half the fuel is consumed for fusion, and the
other half for reduction and carbonisation when treating
easily fusible ore mixtures, this proportion is as three-fourths
to one-fourth when treating ore mixtures difficult to fuse.
The furnaces used in this country for ore mixtures difficult
to fuse are shown in Figs. 119 to 122. Fig. 123 is a furnace
28/
456
IRON.
in Belgium; Fig. 124 is a furnace in France, and Figs. 125
to 128 furnaces in Germany and Sweden.
FIG. 120.
--10----
-12-
1
V
FIG. 121.
+14:0
1
·-16'-
-15
22.5″.
FIG. 122.


13
23
99
.-5'-0"-
Furnaces with flat boshes and high narrow hearths, like
that shown in Fig. 125, give rise to scaffolds, though they
allow a more perfect carbonisation of the iron, and conse-
quently the iron will combine with less silicon in the hearth;
but a purer iron is also produced in furnaces with lower and
broader hearths, and without the sharp angles of the boshes.
These furnaces have lately been employed, chiefly since the
application of hot blast. They are represented by Figs. 129
to 131, and by Figs. 114 and 115 on page 454.
The section of most blast furnaces is round; the circular
form, having the smallest circumference, offers the greatest
surface for the production, consequently less heat is lost whilst
it is uniformly distributed. Some smaller furnaces (charcoal
THE SHAPE OF THE FURNACE SHAFT.
457
furnaces) are constructed of square, oblong, or polygonal
section. Though circular furnaces may be proportionally
wide (7 feet and more), allowing a large production when
the blast is distributed by Sefström's principle, the blast will
not enter the centre of the furnace when using a slightly
FIG. 123.
FIG. 124.
FIG. 125.
14
136"
6
24'.
-29.
A
2.8"
45
combustible fuel (anthracite), or by enlarging the furnace
hearth still more, even if the pressure of the blast is very
high; and the gases will not ascend uniformly throughout
the whole section of the furnace, as the charges sink more
quickly in the centre than on the'circumference.
Owing to this last circumstance, Alger* and Abt† constructed
in 1850 blast furnaces of an elliptic section, Fig. 132, with
*
Polyt. Centr., 1858, p. 1137. B. u. h. Ztg., 1859, p. 156; 1860, pp. 143,
216. HARTM., Fortschr., ii., 206.
† Oesterr. Ztschr., 1859, No. 38. B. u. h. Ztg., 1860, p. 396.
-6-
61..
!



458
IRON.
་་
1.80.27
.25'
several tuyeres, a, on the long side of the hearth, and one, b,
on each of the short sides. This construction has lately
been improved upon and simplified by Von Rachette.* His
furnace, shown in Figs. 133 to 138, has two fore-hearths, a,
FIG. 126.
FIG. 128.
and an oblong section; the tuyeres, b, for the production of
a uniform temperature, are either placed crosswise on the
long sides, or the blast is introduced by a slide, c (Fig. 138).
The proportionally small furnace walls are also furnished
with a system of channels which are connected with the
* C. AUBEL, d. Rachette'sche System der Patent- Normal- u. Universal-
schachlöfen, Leipzig, 1863. TUNNER, Ber. üb. d. Lond. Industrie-Ausstellung,
in 1862, p. 34. DINGL., Bd. 165, p. 370. Oesterr. Ztschr., 1863, No. 14.
Berggeist, 1863, Nos. 18, 25. B. u. h. Ztg., 1862, pp. 265, 392; 1863, pp. 156,
270, 309, 349, 354. Allgem., B. u. h. Ztg., 1863, p. 253.
"1
35..
FIG. 127.




THE SHAPE OF THE FURNACE SHAFT.
459
channels, f, in the foundation (Fig. 136) and a fire-place
similar to that in the blast furnace on page 437. These
channels allow the quick erection of a furnace in any season,
by allowing warm air to circulate in them, and they serve to
FIG. 129.
FIG. 131.
FIG. 130.

5027
32"
ذا.
567
-12′
داد تار
51%
!
1
181
10.
>
24.
4..
cool the hearth by the circulation of cold air; and, when
closing the channels, they assist in the concentration of the
temperature, and make the erection of the furnace cheaper by
lessening the amount of masonry,
Schinz attributes the large production of Rachette's fur-
naces (30 tons of grey iron in 24 hours) to the suitable distri-
bution of the blast, and especially to the smaller transmission.
* DINGL., Bd. 169, p. 449. Berggeist, 1863, p. 331.
460
IRON.
of temperature by the furnace walling, owing to the air chan-
nels and to the general construction of the furnaces. They
FIG. 132.

་་་ ་
--
b
a
a
α
LA
י.
are also economical as regards fuel (15 to 20 per cent of
charcoal).
FIG. 133.

d
α
b
Ъ
a
a
日
​The enlarged furnace mouth which was before recom-
mended by Truran (Fig. 139 represents Truran's furnace)
allows a more uniform descent of the charges, and the fur-
naces may be somewhat lower, and pressure of the blast
RACHETTE'S FURNACE.
461
less, as the slowly ascending gases react upon the ores.
as strongly as a stronger draught in higher furnaces with
a narrow throat. According to the nature of the ores, their
FIG. 134.

E-
F
K
FIG. 135.

e
A
B
C
-D
462
IRON.
contact with carbonic oxide must take place for a longer or
shorter time, but this contact is independent of the section
of the furnace mouth, i.e., of the velocity of the gases; the
FIG. 136.

....
f
loss of temperature increases with the velocity of the gases
without facilitating the reduction. The resistance of the
FIG. 137.
maren

ascent of gases decreases with the velocity in quadratic pro-
portion, thus saving in motive power, and causing a more
uniform action of the blast, and fewer scaffolds.
FIG. 138.

ос
The smaller height of the furnaces lessens the building
cost, and the admissible lower pressure of the blast econo-
mises so much steam power as to counterbalance the more
perfect utilisation of the temperature which a higher furnace
admits.
TRURAN'S FURNACE.
463
Rachette's furnace has given satisfactory results in the
iron works on the Ural when smelting magnetic iron ores
difficult to reduce and fuse, for the production of grey pig-
iron; and the theory shows no reason why it should not be
applied for the production of white pig-iron from easily
FIG. 139.

--
fusible ores. The temperature in the smelting zone may be
lowered by various means; a certain pressure of the gases is
obtained by their slower ascent, whilst the smelting materials
exert great pressure; the lower part of the furnace may be
cooled by partly or wholly opening the flues in the foundation
and in the rough walling, and the upper part of the furnace
is preserved from cooling by the ascending gases required
for the reduction and carbonisation.
At Königshütte (Silesia) Dilla* has constructed blast fur-
naces with shafts in the form of a staircase for the purpose
of keeping the ascending gases longer in contact with the
ores, and of conducting them from the furnace walling to
the centre. These furnaces require less fuel, consuming
0'77 cubic foot of coke per cwt. of pig-iron less than the
usual furnaces.
The Absolute Size of the interior of a blast furnace de-
pends on the nature of the smelting materials, and chiefly on
the intended amount of the production, which is less influenced
*
B. u. h. Ztg., 1860, pp. 7, 327, 395, 440. HARTM., Fortschr.. iv., 92.
464
IRON.
by the height than by the width of the furnace. When in-
creasing the volume of a furnace shaft by increasing its
height, the pressure of the smelting mass will be greater and
the ascent of the gases impeded. The volumes of the different
parts of the furnace, chiefly those from the boshes to the
upper part of the hearth, must stand in a certain propor-
tion to each other, so that the different chemical reactions
may take place regularly at the lowest possible consumption
of fuel; for example, a large smelting zone requires a cor-
respondingly large zone for carbonisation. Mistakes are
more frequently made in this proportion than in the propor-
tions of boshes and shafts. The shaft is usually more than
sufficient for preparing and reducing the ores; this is only
disadvantageous if the smelting mixture forms a mass so
compact as to impede the ascent of the gases.
Distinct theoretical rules for fixing these proportions
cannot be given on account of the great number of influencing
circumstances, and empirical rules can only be deduced from
the comparison of the cubic contents of different blast fur-
naces which allow a normal process. The different dimen-
sions of the furnace are usually regulated by the diameter of
the belly, which has the greatest influence upon the pro-
duction.
Meyer* has lately proposed a different system; he prefers
several small furnaces to one large one, and deduces the
dimensions of the furnace from the size of the smelting
focus, which depends on the size of the tuyeres and the pres-
sure of the blast. He supposes that the relative consumption
of fuel depends on this circumstance only. But present ex-
perience makes a larger furnace preferable to several smaller
ones on account of the utilisation of temperature and the
saving in masonry.
The following rules, founded upon experience, may some-
times be found useful in the construction of blast furnaces,
although not applicable in every case :-
1. The Furnace Belly.-Its diameter and the height
above the bottom stone are of chief consequence.
* Oesterr. Ztschr., 1862, No. 4.
THE FURNACE BELLY.
465
In order to fix the diameter of the belly, French metallur-
gists* formerly stated the following facts as the result of their
experience :-
a. For Charcoal Furnaces.-160 kilos. of charcoal are
consumed for 100 kilos. of pig-iron; 11:56 cubic metres
of air of o° C. and o'76 metre pressure per minute and per
square metre of the largest section; 90 kilos. of charcoal
per hour and per square metre of the largest section.
b. For Coke Furnaces.—235 kilos. of coke for 100 kilos.
of pig-iron; 6'18 cubic metres of air per minute and per
square metre of the largest section; 49 kilos. of coke per
hour and per square metre of the largest section.
=
Now, in order to deduce the largest diameter D (in
metres) of a blast furnace, the three following ratios are
given, namely :-The absolute production of pig-iron (kilos.)
in 24 hours = E; the consumption of fuel in weight for the
production of 100 kilos. of pig-iron K; the quantity of
fuelm, in weight, which may be consumed per hour and
per square metre of the largest section.
The equation of the section may therefore be expressed by
π D₂
4
2
× 24111
KE
100
or D=0'02302/
ΚΕ
112
Now replacing in this equation the above stated ratios for
K and m, the result will be-
For charcoal furnaces. . D = 0°03070 E
""
coke
. D 0*05045/E
Example.-What diameter must a charcoal furnace have if
it is intended to produce in 24 hours 4000 kilos. of pig-iron
from ores of moderate fusibility, containing from 40 to 50 per
cent of iron?
Ores of this quality necessitate the consumption of
160 kilos. of charcoal for 100 kilos. of pig-iron, conse-
quently the consumption in 24 hours will be 6400 kilos.,
or in 1 hour 266.6 kilos. As the consumption of charcoal
*LE BLANC and WALTER, Eisenhüttenkunde, i., 59. B. u. h. Ztg., 1858,
P. 228.
VOL. II.
2 H
466
IRON.
}
per hour and per square metre of the largest section amounts
to 90 kilos., 2'96 square metres will be the section of the
belly according to the proportion-
90: I:: 266·6 : x, and
from this section we may calculate the diameter of the belly
to be 1'94 metres.
The same result will be obtained by placing 4000 instead
of E in the former equation, namely—
D 0'0307 4000 = 1′94 metres,
0*03074000
which undoubtedly is too small for the supposed production.
Von Mayrhofer's formula denotes by M the quantity of
blast which a blast furnace requires per minute, and conse-
quently the smallest diameter of the belly may be calculated
from the formula—
D3 — 34D₂
and the largest diameter-
D3
34D2
o'910 M,
1*287 M.
Example. When carrying on a blast furnace with soft
charcoal, hot blast of 250° C., at a pressure of 18 inches of
water, and with two tuyeres, each 2 inches in diameter, the
atmospheric air of o° temperature required per minute would
be 420 cubic feet, wherefore the smallest diameter of a fur-
nace in question would be D = 3 feet, and the largest,
D = 4 feet, which is also too little.
Lindauer's Formula*.-Finding that the two preceding
modes of calculation do not give mathematically exact results,
as in the first mode, the conditions influencing the dimensions
of blast furnaces being too general, and the second mode
taking into consideration nothing but the quantity of blast,
Lindauer investigated those elements required for a more per-
fect solution of the present question; he deduced the following
formulæ :
For charcoal furnaces-
*
3
k
D = 0·8448
08448+0+E
y'
24
DINGL., Bd. 136, p. 227. B. u. h. Ztg., 2855, pp. 227, 242.
THE FURNACE BELLY.
467
For coke furnaces-
D = 0·8883
0.8883
For coal furnaces-
3
I
-100 y'
+100+ c ] 2 E
24
+] //
Z
q'y
3
100 y'
q'y
D = 09728 +100+2
24
In these formulæ the letters have the following value, and
the figures give the average obtained in practice :-
D
-
diameter of the belly in Austrian feet.
E the production of pig-iron in 24 hours in Austrian
pounds.
=
Z the time which one charge requires to pass the furnace.
For charcoal 16, for coke 40, and for coal 48 hours when
た
​producing grey pig-iron. Three quarters of the time are
required when producing white pig-iron either with
charcoal or with coke.
consumption of fuel per 100 lbs. of pig-iron, namely—
With Cold Blast. With Hot Blast.
Charcoal .
Coke
Coal
160
280
330
130
210
250
C
y
y'
the addition of limestone to 100 lbs. of ore mixture,
15 lbs. on an average when using charcoal, and 35 lbs.
when using coke and coal.
I
weight of 1 cubic foot of mixture, on an average 90 lbs.
with ores yielding from 20 to 40 per cent of iron.
weight of 1 cubic foot of fuel, 7 lbs. for soft charcoal,
12 lbs. for hard charcoal, 20 lbs. for coke, and 40 lbs.
Austrian for coal; the weight of the mixed fuel is ob-
tained by calculation.
average amount of iron contained in the mixed ores
after deducting the loss in smelting.
Upon transposing the average values of Z, c, k, y and y' in
the above formula, the following diameters for the absolute
productions and for the different kinds of fuel will result:-
2 H 2
468
IRON.
0*4500,
For Cold Blast.
For Hot Blast.
3
For soft charcoal D = 0'4766/E
دو
D = 0°4125
3
VE
3
hard
E
3
coke .
. D = 0*6054
0*6054/E
3
E
3
وو
mineral coal . D
3
0*6249 E
0*3920
0*5657/E
3
0*5917/E
Differences, which may be noticed by calculating the dia-
meter of the belly of blast furnaces, may be caused by em-
ploying an unusual quantity or pressure of the blast, thus
altering the time in which a charge passes through the fur-
"nace; or the ores smelted may be so good as to greatly facili-
tate the production of grey pig.
When employing mixed fuel a particular formula must be
deduced.
Example. The construction of a blast furnace for a weekly
production of 500 cwts. of pig-iron is intended for the use of
two-thirds of soft and one-third of hard charcoal, a cubic foot.
of the soft coal weighing 7 lbs., and one of hard coal 10 lbs.
The ores yielding 30 per cent of iron will probably require
an addition of 12 per cent of limestone. One cubic foot of
admixture weighs 86 lbs., and the relative consumption of
fuel will amount to 120 lbs. when employing hot blast of
250° C.
According to these statements, the average weight of one
cubic foot of the mixed fuel will be—
k
100 y'
y' = }
27 + 10 = 8 lbs.,
and (the quantity of fuel required for the production of
I lb. of iron) + 100+ (the number of cubic feet of mixture
q'+y'
C
corresponding to 1 lb. of iron) =
120
800
+
II2
86 × 30
and supposing Z = 16 hours, it follows that
3
D = 0·84480*1974. . 50000
7
0'1974;
8.256 feet.
According to what follows, the capacity of the furnace
will therefore amount to-
J 8·2563 × 16702 = 940 cubic feet.
THE POSITION OF THE BELLY.
469
The other dimensions of blast furnaces usually stand in a
certain proportion to the diameter of the belly. Lindauer
states the average ratios to be-
Charcoal.
Coke.
Mineral Coal.
D.
D.
D.
Diameter of the furnace mouth
0'400
0'500
0*600
""
,,
upper hearth . 0350
0*250
O'250
between the tuyeres
0*250
0'210
0*250
Height of the hearth (boshes
included
0'740
0.667
O'250
"
upper hearth
•
0'490
0*457
0*146
belly.
0°292
O'113
0*506
boshes
""
""
0'464
0·842
o'644
shaft
وو
3'004
2.378
I'200
Total height of the furnace.
4'500
4'000
2.600
55°
65°
60°
Angle of the boshes
From these numbers of the ratio the capacity of the dif-
ferent parts of the furnace may be calculated as follows :—
Capacity of the shaft
وو
Charcoal.
Coke. Mineral Coal.
D3.
D3.
D3.
I'2270 I'0890
0*6158
belly
boshes
0*2293
o'0890
0*3972
0'1789 0*2298
0*0659
upper hearth. 0*0350
O'0190
0'0072
J
1*6702 1*4268 1*0861
Supposing the capacity of the whole furnace to be = 1, the
result will be-
Capacity of the shaft
",
""
Charcoal.
Coke. Mineral Coal.
J.
J.
J.
0*7347
0*7633
0*5670
belly
O'1373
0'0624
0*3658
boshes
O'1071
O'IбII
о обоб
upper hearth.
00209
0'0132
0'0066
J = 1'0000
I'0000
I'0000
The Position of the Belly.-The lower the belly is the
sooner will the temperature of the rising gases decrease;
therefore ores easy to reduce and fuse require a lower position
of the belly than more refractory ores, as they would otherwise
melt too soon. But when smelting these ores in furnaces.
470
IRON.
without real boshes the position of the belly is determined
by the fusibility and reducibility of the ores. A high position.
of the belly is not favourable to reduction. Therefore when
smelting ores of difficult reducibility in furnaces with a wide
shaft the belly should be placed lower, and a boshes-like pro-
jection should be formed by connecting the hearth of the
furnace with the belly by means of a double curve.
As at equal levels the temperatures are higher in coke
furnaces than in charcoal furnaces, the belly is placed higher
in coke furnaees than in charcoal furnaces.
In coke furnaces the height of the belly above the bottom
stone amounts to 0.33 or 0'40 of the height of the shaft,
sometimes even to o'64. In charcoal furnaces the usual
proportions are a quarter, two-sevenths, and one-third; the
former when an easily fusible mixture is smelted with soft
coal, and the latter when treating more refractory ores with
hard coal. When constructing the belly in a cylindrical
form the cylinder may be 18 feet high (Sweden), but it is
usually from 2 to 6 feet high; at Eisenerz it is 1 foot, and at
St. Stephan* in Styria, 4 feet.
The Height of the Furnace.-Though Rachette's furnace
has proved that a furnace may be less high if it is otherwise
suitably constructed, the older furnaces with a narrower
throat require greater height in order more completely
to utilise the enormous quantity of heat carried upwards by
the gaseous products of combustion. As upon combustion
coke produces a higher temperature than charcoal, and con-
sequently the ascending gases are hotter, coke furnaces
require a greater height than charcoal furnaces. The character
of the ores and fuel, as regards their power of resisting crushing,
considerably influences the height of the furnace. Soft fuel,
such as charcoal or coal, or pulverulent ores, necessitate lower
furnaces; lower furnaces are also required when using anthra-
cite as fuel. Anthracite and charcoal furnaces are from 30
to 40 feet high; furnaces using coal for fuel are usually from
35 to 45 feet high; in Wales most of these furnaces are from
LEOB., Jahrb., 1860, x., 285.
THE HEIGHT OF THE FURNACES.
47I
42 to 46 feet high; in Scotland they are from 43 to 48, some-
times 55 feet. Coke furnaces, on the other hand, are seldom
less than 50 feet, and often above 60 feet. The newer fur-
naces in and about Middlesboro' are from 68 to 80 feet high,
and the furnaces of the Rosedale and Ferryhill Company
even 102 feet high, owing to the extremely hard character
of the coke employed.
But the production of the furnaces is increased whilst
consuming less fuel, not only by the greater height of the
furnaces, but chiefly by increasing their width and by using
larger quantities of blast of a higher temperature.
Lower furnaces are less expensive, and require less pressure
of the blast, particularly if they have an enlarged furnace
mouth (Rachette's furnace), and allow a quicker removal of
the irregularities of the process, as modifications of the mix-
tures and the charges are quicker in their effect owing to
their more rapid descent; the pressure and temperature of
the blast also react more quickly. The greater height of the
furnaces does not stand in proportion to the results obtained
when smelting pulverulent ores. The greater width of the
furnaces has been the chief cause of their increased volume;
in Scotland from 90 to 200, in Staffordshire from 50 to 125,
in Wales from 60 or 70 to 140 or 150 cubic metres. This
increase has influenced the quality of the pig-iron produced.*
In Scotland the production of 1 ton of grey or darkish grey
iron in 24 hours requires a space of 7 or 8 cubic metres of
the furnace shaft; grey forge iron in Staffordshire from
6.5 to 7; light grey or mottled iron in Cleveland, 6'5; white
forge iron in Wales from 5 to 6; charcoal spiegeleisen in
Styria and Carinthia from 2 to 3 cubic metres; and I cubic
metre only is used in Tuscany when smelting iron ores from
Elba; this, however, is at the expense both of fuel and the
yield.
About 12 cubic feet of space are used for the production of
I cwt. of grey foundry iron in the Belgian coke furnaces.
Charcoal furnaces 30 feet high and 8 feet in diameter in the
belly have a capacity of from 700 to 800 cubic feet; coke
* B. u. h. Ztg., 1862, p. 237.
472
IRON.
furnaces 40 feet high and 11 feet in diameter in the belly
from 1900 to 2000; and furnaces from 48 to 50 feet high and
13 or 14 feet in diameter from 3500 to 4000 cubic feet.
Truran has given many examples showing the proportion
of the size of blast furnaces to the quantity of the fused mass
and the produced pig-iron, to the production of slag and iron,
to the quantity of blast, to the relation between carbon and
the production of slag and iron, and to the velocity and time
occupied by the materials in their descent from the furnace
mouth to the hearth.
The Width of the Furnace Mouth.-A narrow mouth
concentrates the heat, increases the pressure of the gases, and
consequently the reducing and carbonising reaction; there-
fore a narrow mouth is advantageous when smelting easily
fusible ores in furnaces with a wide hearth (page 453), and
when smelting pulverulent ores, or using friable fuel. In
these furnaces, on the other hand, the descent* of the charge
is not uniform; as they do not expand with the dimensions
of the furnace shaft, the fuel is pressed by the ores towards
the circumference, whilst the ores roll towards the centre
into a funnel-like hollow. The gases then rise chiefly in the
mantle of the circumference, but slightly heating the ore in the
centre of the furnace, thus giving rise to an irregular process.
When using furnaces with a narrow hearth emitting hotter
gases their greater velocity is liable to raise the temperature
in the throat so high as to cause too early a caking of the ores.
These reasons have led to the adoption of wider furnace
mouths,† approaching the belly in diameter (Fig. 118 on
page 455), thus rendering the furnace shaft nearly cylin-
drical. These wider throats have been applied to charcoal
furnaces as well as to coke furnaces, chiefly when employing
raw fuel and smelting refractory ores in furnaces with narrow
hearths, or easily fusible ores at a higher temperature in
furnaces with wide hearths, for the production of grey iron.
The charges in these furnaces sink more uniformly, and are
* B. u. h. Ztg., 1856, pp. 143, 396; 1858, p. 210; 1861, pp. 271, 400. LEOB.,
Jahrb., 1860, ix., 306. DINGL., Bd., 158, p. 413. Berggeist, 1860, No. 6.
+ HARTM., Fortschr., ii., 213. B. u. h. Ztg., 1859, p. 293; 1860, p. 386;
1862, p. 237.
THE WIDTH OF THE FURNACE MOUTH.
473
more gradually heated; the gases ascend more slowly and
uniformly, allowing a longer contact with the smelting mass
and the reduction and carbonisation at a lower temperature.
The cylindrical circumference also has a smaller surface than
a conical one, supposing the furnaces to be of equal capacity,
as it radiates less heat. As the column of the charges
exerts a stronger pressure, causing a quicker descent of the
charges, the production will be increased and fuel econo-
mised. The diminished velocity of the gases lessens the escape
of dust from the furnace mouth, and the larger capacity
of the furnace somewhat increases the production.
The very wide mouths of the furnaces have lately been
again abandoned, as they increase the difficulty of charging
the materials; they are liable to conduct the gases towards
the circumference of the furnace; it is difficult to collect the
gases from the whole surface of the charges, and it is also
difficult to obtain the required pressure of the blast owing
to the strong pressure of the smelting materials in the usually
high furnaces.
Whilst in the old furnaces, and in some of the newer ones,
the diameter of the mouth is equal to one-fourth to one-third
or one-half of that of the belly, in the furnaces of the Upper
Hartz this diameter is judiciously increased to from one-half
to three-fourths, in Neustadt to two-thirds, in England to
five-sixths and more; but, for reasons we have stated, it is,
in Wales, for example, decreased to one-half or two-thirds.
Obtaining the highest production with the smallest con-
sumption of fuel, Parry makes the furnace mouth half as
wide as the belly, and places the belly as far below the
mouth as it is wide. The height of the belly above the
tuyeres must not be less than the diameter of the belly plus.
half the diameter of the level of the tuyeres.
Truran* recommended a greater width of the furnace
mouth for raw fuel, chiefly anthracite (5-4ths to 5-3rds of the
diameter of the belly) which has been applied, and Rachette
has tried this alteration with the best results. The chief
TRURAN, The Iron Manufacture of Great Britain, theoretically and practi-
cally considered: 2nd edition, revised by J. Arthur Phillips and W. H.
Dorman. London, 1862.
474
IRON.
advantages of Rachette's furnace are-The charges descend
uniformly, i.e., arrive successively before the tuyeres in a
horizontal position, thus utilising the fuel more thoroughly;
causing a quicker smelting, producing a smelting product of
better quality, and allowing lower furnaces, as the rising
gases distribute better in the smelting column. The ores
do not roll to the wrong places, and do not come too near to
the tuyeres if charged more in the centre.
The older furnaces at Nischne-Tagilsk (Ural) of a capacity
of 5420 cubic feet, in which one charge required 24 hours to
pass through the furnace, produced 607 cwts. (Prussian)
in 24 hours, whilst a Rachette furnace of 1950 cubic feet
capacity produced 738 cwts. in 24 hours, one charge requiring
seven hours to pass through the furnace.
The Boshes.-The boshes are chiefly destined for the car-
bonisation of the iron, and their inclination and height depends
chiefly on the nature of the fuel, the ore, and the quality of
the intended iron. Flat boshes retaining the smelting mass
for a longer time in the zone of carbonisation cause a more
perfect carbonisation, but they lessen the production, conse-
quently the produced iron combines with less silicon in the
hearth and fuses more easily, thus consuming less fuel. But
as flat boshes give rise to an irregular descent of the charges
and to a loss of gas and temperature, they have been mostly
done away with since the introduction of the hot blast and
wide hearth (chiefly in charcoal furnaces when treating re-
fractory ores), and an increased production of iron not inferior
in quality is obtained. When smelting easily fusible and
reducible ores with charcoal for the production of white pig-
iron, the boshes are also connected with the hearth by a
curve. This curve is also used when smelting easily fusible
ores with coke for the production of grey iron, as in coke
furnaces flat boshes would soon melt away, particularly
when using hot blast; but when smelting more refractory
ores for the production of grey pig, the lower part of the
furnace is constructed as shown in Fig. 123 on page 457.
The steep boshes of 60° or 70°, which are frequently
given to coke furnaces, easily conduct the smelting masses
into the hearth, thus enlarging the production, as the
THE BOSHES.
475
temperature between the boshes is higher; but, on the other
hand, they impede the carbonisation, causing the iron in the
hearth to combine with a larger amount of silicon, with
separation of graphite, although the iron may still combine
with carbon in the smelting zone. The greater height of
coke furnaces in comparison with charcoal furnaces allows
a sufficient preparation of the ores in the furnaces by
gradually heating them, although these furnaces are provided
with steep boshes.
The shape of blast furnaces which have been blown out
has frequently given the rule for constructing new furnaces,
chiefly with regard to the angle of the boshes. Figs. 140,
141, 142, 143, and 144 show the shape of blast furnaces out
of blast.
FIG. 140.
FIG. 141.


α
b
6
FT
1
d
II
The height of the boshes, i.c., their capacity, must stand
in proportion to the smelting space, and must be larger the
more difficult the ores are to carbonise, if, for instance, the
mixture is too compact. When the most suitable proportion
of the boshes to the hearth is ascertained, the production of
the furnace will not be increased by enlarging the capacity
476
IRON.
of the boshes. According to Scheerer,* the Belgian furnaces
give the best result if the volume of the boshes is thirty times
that of the upper hearth.
FIG. 142.
FIG. 143.


C
C-
C
d.
d
e.
h
d
d
d
FIG. 144.

8
The Hearth.-The dimensions (round, oblong, elliptic, or
polygonal) of the hearth greatly influence the amount of the
production and the quality of the pig-iron, and depend chiefly
on the nature of the intended pig-iron, on the fusibility of
the ore mixture, on the fuel and the blast.
Narrow and high upper hearths are employed when
* SCHEERER'S Metallurgie, ii., 94, 127.
THE HEARTH.
477
smelting refractory ores which are difficult to reduce, for the
production of grey iron, as the ores have to pass as long as
possible through glowing coals at a sufficiently high tem-
perature. Narrow hearths raise the temperature in the
smelting zone, as the smelting mass is kept for a certain
time in contact with a larger quantity of hot gases. Wide
and low hearths are suitable for easily reducible and fusible
ores, whether they are treated for the production of white or
of grey iron (pages 453 and 454) (blackband*), and coke
allows the use of proportionally wider hearths than charcoal,
chiefly if employing hot blast, as it produces higher tem-
peratures upon combustion. Higher hearths may also be em-
ployed when using coke as fuel, as owing to its greater
compactness, it burns more slowly, and the carbonic acid
formed is more slowly converted into carbonic oxide, conse-
quently the lowering of the temperature required for carbon-
isation takes place at a higher level. When treating easily
fusible ores in furnaces with narrow and high hearths, the
ores would melt in the zone of reduction and carbonisation,
unreduced iron enter the hearth, and blackband ores, for
example, would yield white iron instead of grey.
Although to produce a certain temperature in the hearth,
either for white or for grey iron, the quantity, pressure, and
temperature of the blast, and the relative quantity of fuel,
are essentially important; yet in order to obtain the best
results with regard to production of iron and consumption of
fuel, each ore mixture requires a certain width and height
of the hearth, which must be ascertained by experiments.
If furnaces intended for the production of grey iron have too
wide a hearth, mottled or white iron is easily formed, some-
times whilst the slag is like that of a regular process. When
producing white iron too wide a hearth causes scaffoldings
and an irregular process, owing to the imperfect reduction
and carbonisation of the iron.* Hearths which are too
narrow are more quickly attacked, and do not allow a large
production, owing to the lower pressure of the blast, which
is only admissible.
HARTMANN., Fortschr., i., 191.
† Oesterr. Ztschr., 1856, p. 22.
478
IRON.
The diameter of round hearths is limited by the difficulty of
conducting the blast into the centre, even if the tuyeres are
uniformly distributed on the circumference. This circum-
stance has led to the construction of elliptical and oblong
hearths. The great width of many English and Scotch
hearths for smelting the rich and easily fusible blackband
ores, and the application of high pressures of the blast has
not been imitated in Belgium, France, and Westphalia, as
these circumstances increase the difficulties of the smelting
process (charging, collecting the gases, the manipulations
in the hearth, &c.)
The enlargement of the hearth upwards so as to give it
the form of an inverse truncated pyramid, facilitates the
descent of the smelting materials, protects the boshes from
too strong a heat, and enlarges the smelting space, thus
keeping the temperature somewhat lower; it also enables
the hearth to retain a larger quantity of iron. Its adapt-
ability depends on the nature of the ore and fuel and on the
blast to be employed.
3
The height of the hearth depends on whether white or
grey iron is to be produced, accordingly it is one-seventh to
one-eighth of the total height of the furnace; namely, to
4 or 4 feet in charcoal furnaces of 22 or 25 feet in height,
from 5 to 6 feet in charcoal furnaces from 35 to 38 feet high,
and to 6 to 6 and 6 to 7 feet respectively in coke furnaces
41 to 44 and 48 to 51 feet high.
The Lower Hearth or Crucible (that part of the hearth
lying below the tuyeres) must be large enough to contain the
iron produced between two tappings off, and it must be
sufficiently preserved against cooling, therefore it should
be as small as possible on the points most exposed to
cooling. In the English furnaces with a large diameter of
the hearth, the hearth walls are contracted towards the
tymp and the fore hearth.
The Tymp.-In small charcoal furnaces the bases of the
tymps are from 12 to 15 inches thick, and in larger furnaces
from 21 to 23 inches, and the lower bases are placed at the
level of the tuyeres so as to allow the slag to run well off.
The tymps of coke furnaces are from 23 to 30 inches thick,
THE TUYERES.
479
and they are placed with their lower part from 1 to 4,
usually 2, inches above the level of the tuyeres, thus facili-
tating the running off of the slag and the manipulations
in the hearth which coke furnaces require for preventing
scaffolds. These tymps are also very liable to waste, owing
to the frequent change of temperature to which their lower
part is exposed. In very small furnaces and with thinly liquid
slags only is the tymp placed somewhat below the tuyeres.
In small charcoal furnaces the distance from the front of the
tymp to the dam is from II to 13 inches, in larger furnaces
from 15 to 16 inches, and in coke furnaces from 23 to 26
inches.
The Dam.-The side of the dam lying in the furnace
slopes about 60°, and in charcoal furnaces its upper part is
placed from 1 to 3 inches below the level of the tuyeres so
as to facilitate the running off of the slags. When producing
thinly liquid slags, such as are usually produced in coke fur-
naces, it is sometimes advisable to place the upper surface
of the dam even as high as 10 inches above the level of the
tuyeres, and to make the liquid slag run over the dam by
the influence of the blast; the hearth is thus better preserved,
the temperature kept together, and the pig-iron protected
from the oxidising influence of the blast. In English fur-
naces the dam is placed 5 or 6 inches above the tymp, chiefly
when smelting a mixture with a large addition of puddling
slags, &c.; the walls of the hearth are thus more protected,
but usually the dam is placed only 2 inches above the level
of the tuyeres.
The Tuyeres.-We have to consider their number, con-
struction, and position. The number of the tuyeres depends
on the size of the furnace and the quantity of blast to be in-
troduced. According to Tunner's investigations,* a separate
area of combustion is formed before every tuyere, extending
at the utmost 1 feet in the direction of the current, and
containing in its interior the hottest part or focus, comprising
about 6 inches. This space of combustion cannot be essen-
tially enlarged either by a stronger pressure or by enlarging
* LEOB., Jahrb., 1860, ix., 297.
480
IRON.
the tuyere. In order to produce a uniform temperature at
the level of the tuyeres, a greater or smaller number of
tuyeres must be employed according to the blast required.
Two tuyeres are used when requiring from 600 to 700 cubic
feet of blast per minute; three tuyeres when requiring
2000 or 2400 cubic feet, and so on. In this manner from
4 to 10 or more tuyeres, yielding as much as 10,000 cubic
feet of blast per minute, are employed in large furnaces with
a large hearth.
When employing one tuyere it is usually placed in one of
the side walls a little nearer to the back than to the front, in
order to prevent the formation of deposits on the back; these
deposits may be removed with tools, but it is difficult (Fig. 104
on page 446). Two tuyeres are placed opposite each other,
on two sides, but so far from each other that the blast does
not meet (Fig. 100 on page 444), or one of the tuyeres is
turned towards the back wall. If a third tuyere is employed
at the same time, it is placed on the back wall, but at a dif-
ferent level from the other two tuyeres; it is also narrower
to prevent a too quick descent of the charges on the back
wall, which would attack and corrode it. When employing
four tuyeres, two are placed in the back wall and one in each
of the side walls; sometimes a fifth tuyere is placed in the
front wall of the furnace, introducing cold blast, whilst the
other tuyeres carry hot blast (Dowlais).* When employing
more tuyeres (but seldom more than five or seven) in furnaces
with rectangular hearths, several tuyeres are placed in each
of the long sides. In round furnace hearths the tuyeres are
distributed on the circumference (Figs. 98 and 99 on page.
440), after Sefström's principle of conducting the blast
towards the centre of the furnace, or somewhat more towards
the front. This arrangement helps to preserve the hearth
by cooling its walling on different points, but, on the
other hand, the loss of blast increases with the number of
tuyeres; it is more difficult to control, more repairs are re-
quired, &c. In many places in England and Westphalia
three or four tuyeres of larger dimensions are used, and with
* B. u. h. Ztg., 1862, p. 428.
THE TUYERES.
481
an increased pressure of blast. When the blast is too much
divided and too thin, it will not sufficiently penetrate the
centre of the furnace. As, according to the observations
of Tunner, the focus of heat at each tuyere does not
exceed a certain extent, a certain amount of coal in the heart.
of the furnace may be excluded from combustion when Sef-
ström's principle is applied to large hearths, and in such
cases the construction of Rachette's furnace (page 460) would
be preferable.
Le Blanc and Walter, Karsten, Scheerer, and others have
tried to find rules* for ascertaining the quantity of blast
required by a blast furnace, but few of them agree with
what are found best in practice. Of all these rules
Scheerer's perhaps is nearest to the truth, which recommends
the construction of the blowing engine sufficiently powerful
to produce with a moderate pressure (2 or 3 inches mercury)
and the usual width of the tuyere, as much blast of atmo-
spheric density, per minute, as will equal the capacity of the
furnace shaft. In many cases the production of grey iron.
requires more blast. Tunner's practical rule of introducing
as many cubic feet of blast in the furnace as the furnace
produces cwts. of iron per week is frequently found useful.
Buschbeckt has lately ascertained the quantities of blast
effectively used in many different blast furnaces, at the
same time taking into consideration all the circumstances
(dimensions of the furnaces, nature of ores and fuel, tem-
perature and pressure of the blast, the construction of the
tuyeres, &c.), and from these data he deduces the require-
ment of blast per square foot of the section of the belly,
the blast being of atmospheric density, and of an average.
temperature of 10°:-
a. Charcoal Furnaces.-When producing spiegeleisen,
from 17 to 23 cubic feet of blast, with a pressure from 1 to
2 inches mercury, and a temperature of from 160° to 250° C.,
are required per square foot of the section of the belly
(Thuringia, Styria, Carinthia).
B. u. h. Ztg., 1861, p. 4.
+ Ibid.
VOL. II.
2 I
482
IRON.
When producing smaller quantities of white iron from a
mixture of moderate fusibility in furnaces of an equal belly,
from 8 to 13 cubic feet of blast (pressure, 1'5 to 3 inches;
temperature, from 140° to 225° C.) are required (Schleiz,
Concordiahütte, Mägdesprung).
When producing daily from 60 to 80 cwts. of grey iron
from a non-refractory mixture in furnaces of the most suitable
height to correspond to the reducibility of the ores, from 5′25
to 6.75 cubic feet of blast (pressure, 5-6ths or inch; tempera-
ture, 10° C.) are used (Jlsenburg, Sayn, St. Stephan); with
a lower furnace (Spremberg) and a slower descent of the
charges, though the ores may be easily fusible, 5'16 cubic
feet; and with a larger production, according to the pres-
sure, 6.58 cubic feet (Lauchhammer), and even 14 or 15 cubic
feet (Rothehütte in the Hartz).
When producing grey pig-iron from compact lying ores
which are difficult to fuse and to reduce in suitably high fur-
naces, and using a pressure of 2.5 inches, 15:25 cubic feet of
blast are used (Malapane), and at a higher pressure 14'5
cubic feet (Wengers-Gorka in Austrian Silesia). In lower
furnaces with a smaller production and a higher pressure
6.5 cubic feet (Berggieshübel in Saxony). The large con-
sumption of blast is noteworthy at Königsbronn in Wurtem-
berg (18.70 cubic feet), at Lünen in Westphalia (13.60 cubic
feet), and at Katzhütte in Thuringia (11'11 cubic feet).
b. Coke Furnaces.-When using a moderate pressure of
3 or 4 inches, not very refractory ores in pieces, with a
daily production of from 73 to 215 cwts., from 13 to 17 cubic
feet of blast are consumed (Schleiz, Sayn, Nivern); at lower
pressures of 1.5 inches (Marienhütte, near Zwickau), up to
23.7 cubic feet-with a large production, owing to a greater
height of the furnaces, the reducibility of the ores, and
moderate pressure of 3.5 inches, up to 24 cubic feet; and at
high pressure and large production, from 14 to 20 cubic feet
(Espérance and Séraing in Belgium).
Refractory ores which lie compactly in the furnace and do
not admit of a quick descent of the charges, require, with a
small belly of the furnace (Königshütte, Gleiwitz) from 15'5
BLAST REQUIRED FOR THE FURNACES.
483
to 18.04 cubic feet, and with a large belly, from 12:38 to
15 38 cubic feet only.
The Styrian Blauöfen with one blast pipe 13 inches in dia-
meter require from 250 to 300 cubic feet of blast per minute; the
largest furnaces, having five blast pipes 3 inches in diameter
and a pressure of 2 or 3 lbs., require about 7500 cubic feet.
The large Scotch furnaces require per ton of pig-iron from
5000 to 5500, per minute 90, and per cubic metre of the fur-
nace shaft 0'5 cubic metre of blast. The Welsh furnaces pro-
ducing forge iron per ton of iron 5000 to 5800, per minute from
80 to 95 cubic metres of blast. Some furnaces at Dowlais
(Figs. 112, 113, and 114 on pages 450, 454), having a capacity
of 230 cubic metres, consume 180 cubic metres of blast. The
furnaces in Staffordshire and Cleveland (Figs. 120 and 122
on page 456), smelting poorer ores than the Scotch furnaces
require from 6000 to 7000 cubic metres of hot blast per ton
of grey iron, or from 0'55 to o'60 cubic metre per 1 cubic
metre of capacity, and 8000 and o'70 cubic metres respec-
tively when using cold blast. In England, 25 tons of blast
are estimated to be required per ton of pig-iron from clay
ironstone, and 8 tons of blast per ton of pig-iron from black-
band. In the latter case 34 cwts. of pig-iron and slag result
per cubic yard of the furnace, and in the former 2.5 cwts. per
week; hence, 8 tons of blast are required for the production.
of 1 ton of iron and slag from clay ironstone, and 4 tons of
blast for a similar production from blackband.
Belgian coke furnaces receive from 2880 to 3640 cubic feet
of blast per minute. Different modes of calculating the blast.
will be given later on.
The proportionally small production of some large furnaces
in comparison with small furnaces is frequently caused by
an insufficiency of blast, as the quantity of oxygen required
for the charged fuel is then wanting, and injures the tem-
perature, the consumption of fuel, and the quality of the re-
sulting pig-iron.* At Ulverstonet the weekly production of
593 tons of pig-iron has been increased to 684 tons by
Allgem., B. u. h. Ztg., 1862, p. 401; 1863, pp. 65, 190.
† TUNNER, Ber. üb. d. Londoner Industrie-Ausstellung, in 1862, p. 33.
2 I 2
484
IRON.
increasing the blast from 9000 to 10,000 cubic feet per minute.
The introduction of too much blast into a furnace increases
the consumption of fuel without increasing the production,
but sometimes lessening it.* The slags are cooled by the
blast; the gases ascend with too great a velocity, injuring
the reduction and carbonisation; and owing to the quicker
process, the pig-iron produced will be less carbonised and
more difficult to fuse.
The same effect, caused by too large or too small a blast,
is produced by an increase or decrease of the fuel whilst
keeping the proper quantity of blast; the production will be
lessened, and the consumption of fuel increased.‡
The quantity of moisture introduced into the blast furnace
along with the blast depends on the quantity of blast, and
on the hygroscopic state of the atmosphere, and influences
the quality of the iron and the consumption of fuel owing to
its cooling effect. With regard to this circumstance the
yield of blast furnaces is more favourable in spring and winter
than in summer and autumn. In this country, where the
atmospheric air in common dry weather contains 142 per
cent of moisture, and the weekly production of 128 tons of
pig-iron requires 3000 tons of air (25 : 1), along with this
quantity of blast nearly 43 tons of water are introduced in
the furnace; this amount may be doubled in summer and
autumn. At the Dowlais Iron Works the pig-iron produced
in winter was shown by a comparison of several years to
FIG. 145.

E
་་་་
ს
d
a
* B. u. h. Ztg., 1860, p. 391.
† Allgem., B. u. h. Ztg., 1863, p. 190.
+
‡ Ibid., 1863, pp. 229, 231.
THE TUYERES.
485
be better by from 4 to 5 per cent than that produced in
summer; the other circumstances were the same in both
seasons.
Tuyeres.-The usual tuyeres for charcoal furnaces are
represented by Fig. 145, in which a, c, b represent the tuyere
made either of iron or of copper. The blowpipe, d, is inserted
loose in the tuyere.
The tuyeres for coke furnaces are constructed with double
walls, which are cooled by a current of water circulating
through the intermediate space.
Such tuyeres are shown by Figs. 146 and 147. b is the
tuyere and a the blow-pipe, movable by means of the wheel,
e, which turns the cog-wheel, f, on the rack, c.
The open
FIG. 146.

a
FIG. 147.
C

space between the blow-pipe and tuyere is closed by means
of the ring, d. Sometimes the intermediate space between the
tuyere and blow-pipe is stopped with clay. In some cases
the tube connected with the blowpipe is provided with a ball-
and-socket adjustment, by means of which the jet may be
made horizontal or inclined upwards or downwards. The
elbow-pipe connecting the blowpipe with the blast main is
usually perforated by a small hole, which is closed with a
486
IRON.
plate of glass or mica, and gives a view of the interior of the
hearth, or at least of a small part of it.
These water tuyeres are also made either of cast or
wrought-iron, or a combination of both, and frequently of
copper or bronze.
The higher pressures used in coke furnaces always require
the space between tuyere and blowpipe to be closed. The
pressure of the gases in the hearth will be lower the greater
the difference between the sections of the tuyere and the
blowpipe, and the farther the blowpipe lies to the back of the
tuyere. Leaky water tuyeres are not only productive of
great waste of fuel but are also likely to cause explosions.
The diameter of the blowpipe has a great influence upon
the pressure of the blast. The temperature in the blast
furnace and its production may be increased by an increase
of the blast and of the pressure, and the same temperature
may be better produced by a smaller quantity of blast
of high pressure than by a large quantity of blast of low
pressure. The former is preferable as regards economy
of fuel, but it can only be used to a certain limit, as
in consequence of the smaller consumption of fuel the
reducing and carbonising gases are not produced in sufficient
quantity.
1
4
The pressure in charcoal furnaces is usually from 3 to 21,
sometimes 3, and in rare cases only 4 inches and more of
mercury; that in coke furnaces usually from 3 to 6, seldom
less than 2, and sometimes 7 or 8 inches; the anthracite fur-
naces are from 6 to 8.25 inches. The width of the blowpipes
of charcoal furnaces with one tuyere is from 1 to 23 inches;
with two tuyeres, from 1 to 2 inches; that of coke furnaces
with one tuyere 3 or 4 inches, with two or three tuyeres from
225 to 4 inches, and with more tuyeres at a smaller pro-
duction 2, and at a greater production from 3 to 3 inches.
The large Siberian charcoal furnaces have blowpipes 5 inches
in diameter. In some cases the diameter of the different blow-
pipes varies.
An increase of the pressure of the blast may be required
if the air is impregnated with much steam, and in rare cases
only is steam intentionally introduced into the furnace along
THE POSITION OF THE TUYERES.
487
with the blast. A small amount of moisture in the blast
may facilitate the removal of sulphur from the smelting iron,
but usually a little steam has no effect, and much steam has
a cooling action.
The pressure of the blast is proper when the combustion
of the fuel takes place uniformly in the section at the level
of the tuyeres; too low or too high a pressure will cause the
ascent of undecomposed air in the hearth,when the tempera-
ture before the tuyeres will decrease and increase in the
upper part of the hearth, undesirably enlarging it.
The pressure of the blast corresponds with the velocity
and pressure of the escaping gases; this influences the re-
duction and carbonisation of the iron, and may be modified
by a suitable construction of the blast furnace, chiefly by the
width of the furnace mouth.
Tunner's experiments* have proved that measurements in
the furnace throat by a manometer are as useful as those at
the tuyere.
Experiments with an intermittent blast, in order to keep
the gases longer in contact with the smelting mass, have not
been successful.
With regard to the position of the tuyere we have to
consider its height above the bottom stone, its inclination,
and the distance of its mouth from the mouth of the blow-
pipe.
The tuyere may be placed higher above the bottom stone
the more easily the smelting mass is kept hot in the hearth,
the higher the temperature before the tuyere, and the greater
the necessity of protecting the iron in the lower hearth from
the influence of the blast; therefore the tuyeres are higher
above the bottom stone when producing foundry iron than
when producing forge iron.
The tuyeres usually have a horizontal position, particularly
when producing forge pig-iron, in which case they are often
somewhat inclined downwards, causing a partial decarbon-
isation of the pig-iron and the removal of injurious substances
(sulphur, arsenic, phosphorus, silicon, earthy metals). When
*
LEOB., Jahrb., 1860, ix., 160, 286.
488
IRON.
producing foundry iron the tuyeres sometimes have an up-
ward inclination of 3° and 5° and even 14°, thus causing a
quicker descent of the charges and an increase of the pro-
duction, and protecting the pig-iron in the hearth from the
reaction of the blast. But such an extension of the zone of
smelting lessens its temperature and confines the other zones
to a smaller compass; the ores are less gradually heated, and,
together with other circumstances, this position of the tuyere
easily gives rise to an irregular process.
Well roasted and refractory ores remain sufficiently long
in the furnace, and allow a greater upward inclination of the
tuyere than easily fusible ores.
The distance of the mouth of the blowpipe from that of
the tuyere is of more or less influence upon the pressure
of the gases in the furnace, on the preservation of the
tuyeres, &c.
Example of the Projection of a Coke Blast Furnace
according to Lindauer's Formula (page 466).
What dimensions must a coke blast furnace have which
is intended to produce daily 200 cwts. of grey foundry pig-
iron from the following iron ores ?-
Red Iron Ore. Brown Iron Ore.
Peroxide of iron.
Silica
Alumina
Lime
Magnesia.
Oxide of maganese.
Carbonic acid and water
60
40
30
25
20
6
4
5
ΙΟ
100
100
On page 387 we stated that when mixing the ores it was
advisable to form a mono-silicate slag containing not more
than 15 per cent of alumina, and this condition will in the
present case be best obtained by mixing 58 per cent of the
red iron ore with 42 per cent of the brown iron ore, when
the following mixture will result:-
EXAMPLE OF PROJECTING A BLAST FURNACE.
489
58 of Red
Iron Ore,
100 of
Mixture.
42 of Brown
Iron Ore.
Iron
24'4
11.8
36.2
Silica.
17'4
10'5
27.9
Alumina
•
8.4
8.4
Lime.
3'5
3'5
Magnesia.
2.3
2.3
Oxide of manganese.
2'I
2.I
Carbonic acid and water
4.2
4°2
Oxygen
10'4
5'0
15'4
58'0
42'0
100'0
Upon adding 32 lbs. of limestone, containing 17.9 lbs. of
lime, to 100 lbs. of this mixture, the resulting 132 of mixture
will contain 62°1 lbs. (or 100 lbs. of mixture, 47°1 per cent)
of slag-forming components, and 27°4 per cent of iron. And
supposing 1.5 per cent of the slag-forming components is lost
by decomposition of silicate, &c., not more than 45.6 per cent
of slag will be produced from the above mixture; the slag will
have about the composition of a mono-silicate as follows:-
Silica
43'0 containing 225 oxygen.
Alumina.
14'0
Lime.
Magnesia
35'5
4'0
Oxide of manganese 3'5
6.5
""
43'0-13'5",
As the smelting loss will be about 1.5 per cent of iron, and
as, on the other hand, the iron will combine with up to 7 per
cent of foreign substances, the 27.4 per cent of iron contained
in the mixture will yield 29 per cent of grey pig-iron, and the
proportion of pig-iron to slag will be as 100: 157, which is
sufficient if the blast employed is not of too high a pressure.
If the consumption of coke per 100 lbs. of pig-iron amounts
to 200 lbs.
k;
the addition of limestone to 100 lbs. of mix-
ture to 32 lbs. = c; the weight of 1 cubic foot of mixture
equal to 85 lbs. = y; the weight of 1 cubic foot of coke,
25 lbs. = y'; the percentage of iron in the mixture, after
having deducted the loss in smelting, to 35 = q'; the time
occupied by the charges in passing through the furnace to
38 hours Z; and the intended production in 24 hours to
E; then by Lindauer's formula-
=
20,000 lbs.
-
=
3
k
Z
D=0·8883 (10+100+) 2, E.
y'
24
490
IRON.
Upon replacing in this the stated values, the diameter of the
belly of the projected furnace must be 14'02 feet, which
corresponds with blast furnaces of equal production. The
width of the furnace mouth 0'5 D 0'5 x 14-7 feet, also
corresponds to a frequently adopted proportion, but may be
enlarged by some feet the better to obtain the advantages of
wide furnace throats. The height of the furnace shaft,
h = 2·378 D+0*113 D=2°5 × 14 = 35 feet, is too considerable
in comparison to other furnaces, and also to the following
calculation-In the Belgian and also many English furnaces
it has been found by experience that 12 cubic feet of shaft
capacity are required per cwt. of grey pig-iron, therefore a pro-
duction of 200 cwts. would require 2400 cubic feet of capacity
of the furnace shaft. The shaft has the shape of a truncated
cone of 7 feet upper and 14 feet lower diameter, and the height
of its side would measure 22.7 feet, supposing the cone to have
a capacity of 2400 cubic feet. That height of 22'7 feet may
be increased by a few feet when smelting raw ores, which
require a longer preparatory heating.
The height of the boshes is 0·84 D=11*8 feet, with an
inclination of 65°, and is nearly correct, perhaps I foot too
high.
The height of the hearth =0·667 D=9 feet may, perhaps,
be reduced to 8 feet to render it suitable for the production
of grey pig-iron from iron ores which are chiefly siliceous;
it would also be advisable to place the tuyeres 2 feet above
the bottom stone.
The upper diameter of the hearth = 0.25 D=3 feet 6 inches
is correct; when intending the production of a very hot
darkish grey foundry iron the hearth may be constructed of
a conical form down to the tuyeres, where it measures 2 feet
6 inches; the adjacent lower hearth has vertical walls, and
must have a capacity of 100 cwts. of pig-iron in 12 hours; and
as I cubic foot of pig-iron weighs 4'4 cwts., the capacity of
the lower hearth must amount to 23 cubic feet, or, according
to the given dimensions of the upper hearth, it must be 2 feet
high, 2.5 feet broad, and 4 feet 7 inches long. Therefore the
total height of the furnace from the bottom stone to the
mouth is 44 or 45 feet.
BLOWING MACHINES.
491
Blowing Machines.-Cylinder blast engines are now almost
exclusively used for producing the blast required by iron
furnaces. These cylinder engines may be classified into
two chief systems, namely, the horizontal and the vertical
blast cylinders. In the horizontal system steam and blast
cylinders are usually placed in one line, and the rod which
carries the two pistons goes through both covers of the blast
cylinder, and is guided on either side. The vertical cylinders
are connected in various ways, usually by a beam, and fre-
quently also by the piston rod, when one cylinder is placed
above the other (overhead engines). Both systems of blowing
engines work well when properly constructed and attended
to. The vertical engines are the most popular in this
country and in some parts of the Continent (Silesia), whilst
the horizontal engines are almost exclusively used in West-
phalia and on the Rhine, but they have also been in use for
a long time in some works in this country, at the Plymouth
Iron Works for example, and they have lately been adopted
in different works of the Cleveland iron district.
Engines with oscillating cylinders are also employed in
some iron works (Styria) and they answer well.
These engines vary much in size. According to Bauer-
man* the largest hitherto constructed are those of Dowlais
and of Ebbw Vale in South Wales; the former engine,
erected by the late Mr. Truran, has a cylinder 144 inches
in diameter, with the same length of stroke; the admission
valves have an area of 56 square feet and discharging valves of
16 square feet. The steam cylinder is 55 inches in diameter,
with a piston making a stroke of 13 feet and 20 strokes per
minute. The volume of blast delivered is about 51,000 cubic
feet, at a pressure of 3 lbs., sufficient for the supply of six
large blast furnaces and four refineries.
The quantity of blast delivered by a blast engine in a cer-
tain time, usually in a minute, may be calculated† from the
capacity of the blast cylinder (or apparatus) and from the
Metallurgy of Iron, p. 158.
↑ KARST., Archiv., 1 R.. ix., 451. Bgwkfd., v. 214;
and 19. HAUSMANN, Studien d. Götting. Ver., iv., 1.
745. TUNNER's Stabeisen u. Stahl-Bereitung, 1858, i.,
xi., 227; xvii., Nos. 18
B. u. h. Ztg., 1844, P.
243.
492
IRON.
number expressing the discharges of the cylinder during that
time; or it may be calculated from the section of the nozzle
of the blowpipe by multiplying it by the velocity of the
issuing blast and by 60.
The former method states thè quantity of blast which
the cylinder contains, and the latter the quantity which is
really introduced in the furnace. A comparison of both
methods shows that an unavoidable loss of blast always.
takes place, caused by the imperfect junctions of the blast-
conducting pipes, of the regulators and blast-heating stoves,
and also by the imperfections of the valves, by the friction,
&c. This loss in the best cylinder blast engines seldom
amounts to less than 20 or 25 per cent, and to still more
in engines of inferior construction. Whilst in practice the
quantity of blast entering the furnace is usually ascertained
from the section of the blowpipe nozzle and the velocity of
the blast, it is also sometimes required to calculate the blast
by the first method, in order to control the loss of blast and
to remove the causes of the loss if it exceeds the allowable
limit.
The section of the nozzle is found by the direct measure-
ment of its diameter and the calculations drawn from it.
The velocity of the blast must be calculated from its pressure
(pressure being the density of the blast in proportion to at-
mospheric air). The higher this pressure the more blast
will enter the furnace for equal sections of the nozzle, and
the higher will be the temperature produced, as the blast
enters more perfectly into the pores of the fuel, causing more
fuel to burn in a given volume in the same time.
The pressure of the blast is ascertained by means of a
manometer, from the height to which the compressed air will
raise a column of water or mercury in a glass tube.
Manometers are either water or mercury manometers.
The former is a hermetically closed vessel of wood or iron
half filled with water; it communicates with a vertical glass
tube about 3 feet long, on the side of which is fixed a scale.
divided into inches and lines; the zero point of this scale
* Literatur über Manometer in Schubarth, Repertor. d. techn. Literatur,
1856, p. 565.
MANOMETERS.
493
indicates the level of the water in the vessel. A second iron
tube passes through the bottom of the vessel and nearly
reaches its roof. This tube is placed with its lower conical
part in a corresponding opening of the blast main, as near as
possible to the tuyere. The water originally standing at
zero will then be pressed so as to rise in the tube, and the
height of the water corresponds to the pressure of the blast.
A high pressure would necessitate very long glass tubes,
which are inconvenient and easily broken.
As the irregular current of blast usually makes these
manometers fluctuate considerably, the highest and lowest
state of the water in the glass tube must be several times
observed, the mean of each of these observations must
be calculated, and, finally, the arithmetical mean must
be taken of those calculations, which then represents the
pressure.
These fluctuations and the high pressures make mercury
manometers preferable.
Mercury manometers are constructed of a glass tube bent
so as to form three legs, two of which are fixed on a board con-
taining on both sides of the legs a scale divided into inches
and lines. These tubes are half filled with mercury, and the
zero point of the scale is placed at the level of the mercury.
The third leg is then fixed in the blast main by means of a
cork, and the pressure of the blast is measured by the differ-
ence in the level of the mercury columns in the two legs.
A scale of 3 inches above and below the zero point is suffi-
cient. The manometer must stand perfectly perpendicular,
or the column will rise too high. The fluctuations of the
mercury level are less in degree than those of the water
manometer.
When observing the pressure of hot blast, which is liable
to cause a volatilisation of mercury, the observations must
be made quickly, and the admission of the blast must be
regulated by a cock which is placed in the third tube-leg; or
this leg, communicating with the main blast, may be cooled
with water.
*
*
KARSTEN'S Eisenhüttenkunde, 3 Aufl., tabula, ix., Figs. 17 to 20.
494
IRON.
Owing to the vis inertia of the mercury, the fluctuations
of the mercury manometer are greater even than the irregu-
larities of the blast, and, in order to measure these irregu-
larities, Nordenskjöld* has narrowed the bent part of the
tube which connects the two legs containing the mercury,
either by a cock or otherwise, so that its section amounts to
between one-third and one-fourth of the section of the tube.
The exact proportion of the section must be ascertained by
experience, and the fluctuations of the mercury will then
correspond to those of the blast.
The height of the pressure of water manometers is usually
expressed in feet and inches, and that of mercury manometers.
in inches and lines. A water column, W, ascertained in feet.
may be reduced to inches of a mercury column, M, by the
formula-
M
W. 12
13'596
0.882 W.
In order to observe on a mercury manometer the height
of a corresponding water column, each inch of the scale must
be divided into 13'59 parts, one part of which, observed on
the mercury column, will then correspond to one inch of the
water column.
The scales of mercury manometers are sometimes divided
into inches, centimetres, &c., and are at the same time pro-
vided with a second division indicating the pounds and
ounces. The reduction of the weight into measurement may
be effected by the formula-
II I
9 32
P = 13.596.
2
Z. = 0'5193 Z
P
0'5193
In this formula P expresses the pressure of blast per square
inch in Prussian pounds, which is thus reduced to the height
of a mercury column in Prussian inches, Z.
Calculations from the height of the manometer columns.
by means of the modified formula for the fall of bodies do
not show the true velocity of the blast, as it is influenced
DINGL., Ixviii., 437.
Manometer: Aschauer in
Polyt. Centr., 1853, p. 529.
Polyt. Centr., 1858, p. 1191. Ueber verbesserte
Dingl., lxxix., 187. Bgwkfd., xi., 62, 737; iii., 431.
TUNNER, Stabeisen und Stahlber., 1858, i., 241.
THE TEMPERATURE OF THE BLAST.
495
by the temperature, by the state of the barometer and hygro-
meter, and by the contraction of the blowpipe nozzles.
The Temperature increases the velocity of the blast.
The common temperatures are ascertained by the thermo-
meter, but the higher temperatures of hot blast are ascer-
tained-
a. By a mercury thermometer showing temperatures up
to 300° C.
b. By a metal pyrometer for ascertaining moderately high
temperatures. Oechsle's pyrometer is frequently used;
it consists of a spiral spring formed of steel and brass;
the exterior extremity of the spring is fixed to the brass
casing of the apparatus, whilst the interior end sets a hand
in motion by means of a vertical axle; this hand indicates
the degrees on a dial.
c. By means of metals or alloys having a known point of
fusion, when the blast is heated to very high temperatures
(vide chapter on Fuel).
The state of the barometer influences the density of the
blast, by Mariotte's law.
The moisture in the air indicated by the state of the
hygrometer is less influential, and may usually be left out of
consideration when calculating the quantity of blast.
This moisture is more important as regards the consump-
tion of fuel and the smelting processt (page 484). Eckt has
ascertained that in Königshütte in Silesia a blast furnace
receives 150,000 cubic feet of blast per hour, and along with
the blast 1.35 cubic feet of water, as the average amount of
water in 1,000,000 cubic feet of air is 9 cubic feet.
According to Weisse's and Eck's investigations, the
Hygrometrishe Tabellen zur Anwendung bei Gebläsen u. Gradirwerken.
KARST., Metallurgie, iii., 273. KARST., Eisenhüttenkunde, ii., 586. KARSt.,
Arch., xxi., 49.
† KOHLER., Bergm. Journ., 2 Jahrg., ii., 93. Polyt. Centr., 1849., p. 1134.
B. u. h. Ztg., 1849, pp. 497, 700. DINGL., lxvi., 316. ERDM., Journ. f. ök. u.
techn. Chem., ii., 398, 409. LAMP., Fortschr., 1839, p. 43.
B. u. h. Ztg., 1849, p. 497.
|| Ibid., 1849, pp. 498, 700. Polyt. Centr., 1849, p. 1134. DINGL., lxvi., 316.
496
IRON.
months may be classified into the dry and moist months as
follows :—
January, December, February
November, March, April
October, May, June .
September, August, July
5: 5: 6
6: 7: 8
8: IO: II
II: 12: 13
The summer stands, therefore, in proportion to the winter
as 37 : 65, or 4:7; and experiments have proved that the
effect of the blast in summer and in winter is in the propor-
tion of 5: 6.
According to Rogers,* air of o° contains 1-600th of its
own weight of water, air of 15° C. 1-80th, of 30° C. 1-40th;
therefore in very hot weather the production of iron may be less
by 20 per cent. In England 1 cubic metre of air is supposed
to contain on an average 87 grammes of water. Balling
estimates the quantity of water which is introduced in
24 hours into a blast furnace of medium size by means of the
blast to be about 800 lbs.
The contraction of the current of blast when emitted from
the nozzles has some influence. As it is impossible to apply
the manometer close to the mouth of the nozzle, the air has
to overcome some resistance on its way from the manometer
to the tuyere, owing to the contraction of the nozzle, which
causes a decrease of the velocity of the issuing air. The
extent of that decrease has been determined by d'Aubuisson,
Schmidt, Koch, Buff, Weisbach, and others, and, according
to the latter, the co-efficient of contraction for the lower
states of the manometer of I centimetre amounts to 1'010,
and for the higher states of the manometer of 20 centimetres
to 0.928 or o°920 on an average.
Taking all these circumstances into consideration, the quan-
tity of blast emitted from the blowpipe nozzle at 28 inches
barometric pressure, and o° temperature per minute, may
easily be calculated from formulæ, the construction of which
may be learned from the hand-books on applied mathematics.
Karsten has suggested the following formula :—
0*079/h) Ah H+h
M = 204'9 (1
- 0°
*
Bgwkfd., x., 148.
H(1+0.0047(t+t')
FORMULÆ FOR CALCULATING THE QUANTITY OF BLAST. 497
In this formula M expresses the quantity of blast in cubic
feet; A, the section of the blowpipe nozzles in square feet;
H, the state of the barometer in feet; h, the state of the
water manometer in feet; t, the temperature of the blast
before the blowpipe in degrees of Reaumur; ť, the tempera-
ture of the atmospheric air.
Scheerer has suggested the following formula :-
2481 (1 — 0*084/M) D (1+.0003665)
DV
M (B+M)
w
Q
In this formula Q expresses the quantity of emitted blast
in Prussian cubic feet, the blast being not compressed, re-
duced to 28 Prussian inches barometer and o° hygrometer.
M, the state of the mercury manometer expressed in
Prussian inches.
B, the state of the barometer in Prussian inches.
D, the section of the blowpipe nozzle in Prussian square
feet.
t, the temperature of the blast in centigrade degrees.
w, the co-efficient of expansion with regard to the humidity
of the blast. This co-efficient can usually be omitted without
causing much error.
These formulæ prove that a smaller quantity of blast, when
it is hot, enters the furnace than blast of common tempera-
ture, supposing all other circumstances to be equal. When
the same quantity of hot blast is to be introduced, the
diameter of the nozzles must be enlarged; the amount of
enlargement may be easily calculated from the formula.
Herter* employs the following formula :-
Q'
=
(b+m) m
16.787 (1 00084 m) I+at
Q'expresses the quantity of blast in Rhenish cubic feet,
reduced to a temperature of o° and the pressure of one atmo-
sphere, which is emitted per minute, the mouth of the blow-
pipe nozzle being 1 square inch.
m is the state of the manometer in inches of mercury.
b, the state of the barometer in inches.
t, the temperature of the blast expressed in centigrade
degrees.
SCHEERER'S Metallurgie, i., 465. B. u. h. Ztg., 1858, p. 227.
+ Berggeist, 1860, p. 829.
VOL. II.
2 K
498
IRON.
<
a is the co-efficient of expansion of the blast per centi-
mesal degree = 0.003665.
In the derivation of all these formulæ the rather consider-
able resistance has not been taken into consideration to which
the blast is exposed by the smelting materials in the furnace,
thus causing a loss. Scaffolds before the tuyeres may reduce
their section to a smaller size than that of the blowpipe
nozzles, thus causing a higher pressure, and the formulae will
show a greater quantity of blast than really enters the furnace.
This loss is not easily estimated, and is perhaps small if the
tuyeres are clear.*
In order to avoid the frequent circumstantial calculation
of the quantity of blast, tablest have been formed according
to one of the formulæ, showing the quantity of blast for the
various diameters of the blowpipe nozzles, at the various.
pressures and temperatures of the blast, &c; but these tables
are deficient, as all the influential circumstances (for instance,
temperature of the air, state of the barometer) cannot be
taken into consideration, and therefore the values resulting
from the tables still require revision.
V. Schwind‡ has invented a convenient contrivance similar
to that by which arithmetical questions are mechanically
solved, to calculate the quantity of blast for every diameter
of the nozzles, for all pressures and temperatures and every
state of the barometer. It is based upon a logarithmic cal-
culation, and consists essentially of a rule 32 centimetres
long and 3.2 centimetres broad, on which two scales are
fixed, and which besides is provided with four movable scales.
fixed on two sliding pieces.
* B. u. h. Ztg., 1858, p. 227.
TUNNER, Stabeisen u. Stahlber., 1858, Bd. iii., p. 214. RITTINGER'S Centri-
fugal-Ventilatoren, 1858, p. 64. KRAUS'S Jahrb., 1852, p. 96. Berggeist, 1860,
No. 102.
Oesterr. Ztschr., 1854, No. 48; 1856, No. 30; 1858, No. 37. B. u. h. Ztg.,
1856, No. 39; 1858, No. 13. KRAUS'S Oesterr. Jahrb., 1855, p. 1; 1852, p. 172.
TUNNER'S Stabeisen u. Stahlber., 1858, Bd. 3, p. 245. This contrivance may
be bought of the Mechaniker Redtenbacher in Jschl (Austria) for 5 gulden
(about 10 shillings), together with a pamphlet explaining the application of the
apparatus. V. Schwind. das Aichmaäss für Gebläseluft, Wien, 1856.
|| SELLA, Theorica e Pratica del Regolo Calcolatore, Torino, 1859.
FORMULÆ FOR CALCULATING THE QUANTITY OF BLAST. 499
These formulæ for calculating the quantity of blast are
based upon Mariotte's law. Weisbach* has lately applied
Mariotte's law or Poisson's law to calculating the amount of
blast, and proved its exactness by experiments. This law is
based upon the fact that the condensation of gases produces
warmth and their expansion cold; therefore the temperature
is lowered on the emission of the compressed blast from
the tuyeres; and this, on the other hand, influences the
pressure. Whilst by Mariotte's law the tensions of gases.
at the same temperature stand in the same proportion as
their specific gravities, Poisson's law teaches that the elas-
ticity decreases or increases in a quicker proportion than the
specific gravity if the volume is modified by mechanical
compression or expansion, thus causing a modification of
the temperature.
The following is Weisbach's new formula which takes this
law into consideration :-
b+h 0*2953
b
Q=395μF.(~~) 3*3866(1+0*003677) (1 — [6] 0'2953). (1).
(I +h.
Q is the quantity of air per second in cubic metres, mea-
sured as standing under the exterior pressure.
F, section of the blowpipe nozzle in square metres.
, co-efficient of contraction (0.910-0930).
b, state of the barometer in metres.
h, state of the manometer in metres.
7, temperature in centigrade degrees.
This formula may be expressed in cubic feet as follows:—
Q
=
b+
b+
8.2 // 10 (6+4) 03 [(6+4) 0·3_1] cubic feet.
3
(II).
Q expresses in Prussian cubic feet the quantity of blast
emitted from I square inch of the section of the nozzles; its
temperature supposed to be 10° C., and the state of the baro-
meter 29 inches. h is the state of the manometer, and b the
state of the barometer in inches.
Bornemann upon Weisbach's formula (1) has founded a
WEISBACH, Ingen. u. Maschinen-Mechanik, 3 Aufl., Bd., i., § 422, p. 821.
PLATTNER'S Vorles., v. Richter herausgegeben, 1859, Bd. 1, p. 274. B. u. h.
Ztg., 1861, p. 179.
† Civil Ingenieur, N. F., Bd. v., Hft., I.
+ Ibid., Bd. vii., Hft. 2. B. u. h. Ztg., 1861, p. 178.
2 K 2
500
IRON.
graphic table, which comprises the quantities of blast passing
through nozzles from 15 to 200 millimetres wide at from
4 to 180 millimetres of mercury column. This is sufficiently
exact for practice, and is preferable to the other numerous
tables, as it admits of three entrances, whilst the others have
only two. The graphic table therefore allows the direct
solution of inverted problems, and may be made useful for
every system of measurement by merely adding the desired
scale.
Fig. 1 on the following table represents a diagram covered
with a network of horizontal, vertical, and oblique lines, and
provided at the sides with different scales; namely, Prussian
and Austrian inches (they coincide, owing to the smallness
of the scale employed), English inches, and millimetres.
The lower portion of the diagram contains the pressures
of from o'005 to 0.25 atmospheres, and the upper portion a
corresponding scale, stating the pressure in from 4 to 180
millimetres of mercury column; above these are scales stating
the pressure of the blast measured by a water column in
English, Prussian, and Austrian inches.
The scale on the left border of the diagram gives the dia-
meter of the blowpipe nozzles in millimetres from 15 to 200,
also in English, Prussian, and Austrian lines.
The scale on the right-hand border of the diagram corres-
ponds to the oblique line of the figure, stating the quantity
of blast emitted per second in cubic metres, and English,
Prussian, and Austrian cubic feet, supposing a temperature
of 10° C. and the state of the barometer to be o'76 metre.
The following examples will explain the application of the
graphic table :-
A. On the Supposition that the Temperature of the Air
is 10° C. and the Height of the Barometer o 76 Metres.
Problem 1.-The quantity of blast in Prussian cubic feet
is to be determined which issues per second from a nozzle
30 Prussian lines in diameter at a pressure of 7 Prussian
inches water column.
The pressure of the blast must be observed on the upper
border of the figure, and the diameter of the nozzle on the
-Endian Lines
Millimeter
1,6
1,5
1,4
Fig. 3.
13
1,2
1,1
zoL
90
80
18bi
780
160
70
140
60
60
120
50
150
100
90
40
80
70
30
130
Diameter of the blowpipe-nozzles.
60
50
20
40
18
16
14
12
14
30
25
10
70
20
9
-29
178-
8
8 17
4
2
A
3
"
Fig. 1.
Pressure or the blast in inches water column
of
10
2-
Fig. 2.
160
Table 1. To face page 500.

154
4000
1.66
164
162
801
60
401
20
1,52
1,50
1,48
1,46
144

142
300
40
5
1
+
40
60
50 5 60 5 70
sot
80
90
100 inches Pr. or Austr
140
70
80
90
po" Engl.
+80
90 100
120
140 180 180mm 300
эро
80
B. P. O. Cubic feet
138
150 150
150
60
1,36
6
3
5
20t
8 20 2 4 6 8 30
+
22
28 120
150
Pressure of the blast in millimeters me
millimeters mercury
ercury column.
+50
60
R
essure in Atmos
Atmosphere.
0007 0.008 0.009
1001
0006
13.0007
0012 0014 0016 0018 0.00009
0103
004
-9001
0,16 0,17 0,18
0,19
0,2
1
2 3
4
6
7
8 9- 0,3
2
4
6 8
0,05
-0,5
04 2
0,06
007 008 009 01
0,6
0,72
014
0002
0,7
4.8
4 6 8 0,5
04
4
05
6 8 0.6
5
CE
0.3
02
015-
16
16 00055 006
26 0.0056
2005
Lek English
Cubic feet
"Pr...
Austrian-
Quantity of blast per Second.
London: Longmans & Co.
про
018 02
5 08
་
0,7
Second.
Cubic meter
sto
0/2
Поро
(4
022
0,003
09
08
910
2
15
to
10
200 200
100
90
+86
80
60
50
75
40
134
132
20
130
200
1.28
1,26
80
Quantity of blast per Second.
ure of the blast in centesimal degrees.
Temperature
124
Coefficients of Reduction.
60
122
120
40
118
20
116
114
100
112
80
210
60
50
30
20068
070
072
074
300
310
076
108
106
104
102
200
078
68 Met
350
355 Par Lin.
320
330 5 340
State of the Barometer.
H.Adlard Sculp.
፡
!
APPLICATION OF BORNEMANN'S TAble.
501
left-hand border. From the first point a vertical line is
drawn, and from the latter a horizontal, and from the point
where these two lines meet a rule is laid parallel to the
system of oblique lines, when the quantity of blast may be
found in the scale on the right-hand side, marked P. In this
case the quantity of blast will be 5'4 cubic feet. The deci-
mal figures must be estimated.
Problem 2.-What pressure of blast is required if a certain
quantity-for example, o'2 cubic metre per second-is to
issue from a blowpipe 60 millimetres in diameter? The point
of the scale on the left-hand side, marked 60, must be observed,
and the point o°2 on the right-hand border. From this point
the oblique line upwards must be followed till it crosses the
horizontal line commencing at the point 60, and from that
junction, going vertically upwards, we shall find in the scale
of the pressures that a mercury column of 28′2 millimetres
is required.
Problem 3.-The diameter of a blowpipe nozzle is to be found
which emits 10 Austrian cubic feet of blast per second, the
pressure being 6 inches of water. The quantity of blast, 10,
must be looked for in the scale marked O, on the right-hand
border, and the point 6 in the scale on the upper border;
an oblique line is now drawn from 10 and a vertical line
from 6, and a horizontal line drawn from the junction of
the two former lines towards the scale on the left-hand,
which will then show 43 lines as the required diameter of
the nozzle.
B. On the Supposition of Different Temperatures of the
Blast and Different States of the Barometer.
Problem 1.-When finding, the quantities of blast for other
states of the barometer and thermometer (as o'76 metres and
10° C. respectively), the numbers resulting from Fig. I
require to be corrected, and Fig. 2 contains the co-efficients
of correction. The lower border of Fig. 2 forms a scale
containing the height of the barometer from o‘68 to o‘S
metres, and the left-hand border contains a scale of tem-
peratures from 20° to 400° C. Supposing the thermometer
and barometer to be respectively 300° C. and 0'72 metre,
502
IRON.
these two numbers must be looked for in the two scales of
the lower and left-hand border, and from the two respective
points we must follow the horizontal and vertical lines till
they cross each other; from this junction the oblique lines
must be followed upwards to the scale containing the co-
efficients of reduction on the right-hand side.
The co-
efficient in the present case will then be found to be 1'494.
Problem 2.—What pressure of blast is required to emit
0'3 cubic metre through a nozzle 60 millimetres wide, the
blast to be of 300° C., and the height of the barometer 0'72
metre? The co-efficient for a blast of 300° and 0.72 metre
height of the barometer, according to Fig. 2, is
0'3
I'494
1'494;
the quantity of blast o'3 must be divided by that co-efficient
= 0.20 cubic metre, which reduced quantity must be
looked for in Fig. 1, taking into consideration the before-
stated suppositions, when a pressure of 28.2 millimetres will
be given; therefore the next thing to be done is to reduce
the quantity of blast to 10° C. and o'76 metre.
Fig. 3 affords a comparison of the different quantities of
blast which result when applying different formulæ, all
other circumstances being the same. As Weisbach has
proved, formula 1, expressed in the diagram by the curve.
No. 1, yields smaller results than the older hydraulic
formula (curve No. 3), and larger than the logarithmic
formula (curve No. 4), whilst Weisbach's formula of ap-
proximation stated in III. (curve No. 2) gives but slightly
higher results. In the diagram the proportion of interior
and exterior pressure is expressed on the left-hand side as
abscissæ, and the corresponding quantities of blast as
ordinates. If there is 1-20th more than the ordinary pres-
sure of the atmosphere (abscissa 1'05) the differences
become rather obvious.
Weisbach* has lately suggested a very simple formula
which gives, at low pressures, the same values of blast as the
older formulæ, and as formula 11.; when applying higher
pressures the results of this formula of approximation differ
more from the values resulting from the older formulæ, but
* B. u. h. Ztg., 1860, p. 201.
WEISBACH'S FORMULE.
503
they are nearly equal to the results of formula II., therefore
this formula of approximation yields results sufficiently
exact for practice.
The new formula is:
Q = 363 F
363 F. ✓ (1 + 0·00367 7) ½ cubic metre, or
=
Q 1158 F✓
I
T)
½
0'00367 7) cubic feet.
Substituting 10 instead of 7, it follows :
Q = 369 F
cubic metres, and
b
(III.)
(IV.)
(v.)
(VI.)
Q = 1179 F cubic feet
b
When employing hot blast, this expression must be multi-
plied by-
I⚫018
I + 0.00367. T'
and Q then expresses the volume of blast in Prussian cubic
feet, or in cubic metres per second, reduced to the mean
temperature of 10° C., and to 29 inches of the barometer,
whilst
F, represents the section of the blowpipe nozzle in square
feet or metres.
h, the state of the manometer in inches or metres of
mercury.
b, the state of the barometer in inches or metres of
mercury.
т, the temperature of the hot blast (C).
The formula (vI.) is reduced to the simplest form, when
taking F equal to 1 square inch :—
Q = 8.255
Q, expressing the quantity of blast, of 10° C. and 29 inches
of barometer, in Prussian cubic feet, which issues per second
through a nozzle of 1 square inch section. b, the state of
the barometer in inches, and h, the state of the manometer
in inches.
From this formula the following table of blast may be
calculated :-
504
IRON.
h
b
Q.
h
b
Q.
Ο ΟΙ
0.82
0°30
4'49
0'02
I'16
0*35
4.85
0°05
1.83
0°40
5°19
ΟΙΟ
2*59
0'45
5'50
O'15
3*18
0'50
5'80
0'20
3.67
0'55
6'08
0*25
4'10
0*60
6'35
When applying hot blast these values require to be multi-
plied by the above-stated co-efficient.
The following table* contains the weight, G, of 1 cubic
metre of air at the mean height of the barometer of o'76
metres, and at different temperatures centigrade :-
Temperature.
Centigrade Degrees. I Cubic Metre.
Weight of
Temperature.
Weight of
Centigrade Degrees. 1 Cubic Metre.
Kilos.
Kilos.
0°
I'299
110°
0'919
I
I'294
120
0.896
5
1*265
130
0*873
ΙΟ
I'252
140
0*851
15
I*229
150
0.831
20
I.208
160
0*811
25
I'188
170
0*793
30
I'177
180
0'775
35
1*148
190
0*758
40
1'130
200
0'742
45
I'III
225
0°704
50
I'094
250
0.670
55
I'076
275
0*639
бо
1'060
300.
0.611
65
I'044
325
0'585
70
I'029
350
0'561
75
1'013
375
0*539
80
o'995
400
0'519
85
o'985
425
0'500
90
0'971
450
0.483
95
o'957
475
0'467
ΙΟΟ
0°944
500
0*451
The following formula is applied in order to find the
weight G' of a cubic metre of air of different temperatures
and of any pressure:
G' = G+0*097. h. G.
* RITTINGER'S Centrifugal-Ventilatoren, 1858, p. 30.
BLAST REGULATORS.
505
in which G expresses the weight as stated in the preceding
table, and h the height of a water manometer.
The weight of 1 cubic metre of atmospheric air at 200° C.
(G = 0.742) and the height of the water manometer being
I metre, therefore-
G'
0°742 + 0*097.1. 0°742
0.814 kilog.
Blast Regulators. As the direct blast from the blowing
engines is an irregular current which would exert an un-
favourable influence upon the smelting process, it is necessary
to use some means of equalising the pressure. The equal-
isation is usually effected by conducting the blast into large
closed reservoirs, which are usually constructed of iron plates
in the form of a cylinder like a steam boiler; in some cases
these reservoirs or regulators have the shape of a globe;
regulators in masonry are seldom used.
When employing sufficiently long blast mains a regulator is
superfluous. If the mains from the engine to the regulator
are short, the regulator must be from 20 to 50 times as great
as the quantity of blast delivered per second, but it may be
much smaller if those mains are long. The regulator must
be provided with a safety valve and a man-hole, and is
usually placed in the open air upon pillars, and sometimes
underground.
The sizes of some of the regulators in operation are as
follows:
At Hörde (Westphalia) the regulator is 230 feet long,
6 feet in diameter, of a capacity of 6500 cubic feet, and
serving for four blast furnaces, therefore containing 3250
cubic feet for two blast furnaces. The blast cylinder for two
blast furnaces has a capacity of 301.6 cubic feet.
At Séraing the regulator is 90 feet long, 7 feet in diameter,
and is constructed of iron plates 4 millimetres thick.
In some rare cases, when a low pressure is used, so-called
movable regulators are applied. This is effected by con-
ducting the blast from the blast engine into a second cylinder
provided with a loaded piston, thus exerting a compressive
power. The same effect may be produced by a gasometer
floating in a water tank, but the contact of the blast with
506
IRON.
water is injurious, as water will be absorbed, which lessens
the yield, and gives rise to the production of white pig-iron.
Blast Heating Apparatus.
According to present experience the following may be con-
sidered as the advantages of hot blast.*
1. An Increase of the Temperature, as the blast entering
the furnace in a hot state does not absorb heat, which would
otherwise have a cooling effect; but the increase so effected
is not considerable, and is only of consequence if the blast is
heated by waste heat.
The greater affinity to carbon and hydrogen possessed by
the oxygen of hot blast, and the property which the hot blast
has of permeating the fuel more perfectly on account of its
diminished density, are of greater importance. For this
reason in the smallest space more combustible substance
burns at the same time than when employing cold blast, which
increases the intensity of the heat. The effect of this is so
advantageous that it is even profitable to heat the blast with
separate fuel. According to Pfort's and Buff's experiments.
cold air cannot be used for direct combustion; moreover, cold
air requires to be previously raised to a higher temperature,
which causes a greater contact with coal, and gives rise to
the formation of more carbonic oxide, preventing the pro-
duction of a very high temperature.
According to Scheerer,† charcoal at o° C. burned in air of
o produces a temperature of 2700° C., and coal of 400° C. a
temperature of 2735° C.; but when heating the air to 400° C.
the produced temperature will amount to 3065° upon burning
coal of o°, and to 3100° C. upon burning coal of 400° C.
Though the advantages are greater the hotter the blast is,
v.
* Theorien über die erhitzte Gebläseluft. Von BUFF and PFORT, in POGG.
Ann., Bd. 34, p. 173. Bgwkfd., iv.. 102. Studien des Götting. Ver. Bergm.
Freunde, iv., I. EBELMEN, in Valerius Roheisenerzeugung, Deutsch
HARTM., 1851, p. 281. Bgwkfd., ii., 470; viii., 456. SCHEERER'S Theorie, in
his Metallurgie, i., 479; ii, 140. B. u. h. Ztg., 1854, p. 244. Berggeist, 1858,
p. 161. Ann. d. Min., 3 sér., T. 18, p. 183.
† B. u. h. Ztg., 1860, p. 494.
BLAST HEATING APPARATUS.
507
there is also a certain limit in all cases caused by the
difficulty of keeping the apparatus tight at higher tempera-
tures, and also by its action upon the process, the furnace
walling, and the produced iron. The blast is not usually
heated above the melting point of lead (322° C.), but some-
times temperatures of 400° C. and higher are employed.
2. A Saving of Fuel.*-If the melting point of pig-iron
is accepted as 1200° C., and the blast be heated to 200° C.,
a direct saving will be effected of 1-6th of the fuel; but this
saving is much larger owing to the higher temperature in the
furnace, which admits of smaller coal charges and higher
ore charges. The hotter the blast is the less fuel will be
required for smelting; but here also is a fixed limit, as a
certain quantity of fuel is required for the formation of the
reducing and carbonising gases.
The thinner and hotter blast enters more quickly and
easily into the pores of the fuel, and when using a proper
pressure the hot blast permeates the fuel in a horizontal
direction at the level of the tuyeres, which forms another
reason for the saving of fuel. The reaction of the hot blast
at the level of the tuyeres also causes the reduction and
smelting zones to be at a deeper point of the furnace than
when employing cold blast; the hot gases have therefore to
pass a longer way to the furnace mouth, are enabled to com-
municate their temperature to the ore more perfectly, and
escape from the furnace mouth at a lower temperature;
besides, the ores are kept for a longer time in contact with
the reducing gases (page 417).
+
3. A Larger Production.t-If, under equal circum-
stances, cold blast is replaced by hot blast, a lower pressure
will be required, as the hot blast permeates the fuel more
easily, and consequently less oxygen enters the furnace in
the same time, less fuel is burned, and the charges descend
more slowly. But in spite of the fewer charges an increase
of the production will take place, as the higher temperature
* KRAUS'S Jahrb., 1852. p. 117.
† B. u. h. Ztg., 1858, p. 244.
.f
P. 47
RITTINGER'S Centrifugal-Ventilatoren, 1858, p. 74. KRAUS's Jahrb., 1852,
•
508
IRON.
of the furnace allows heavier ore charges, which more than
outbalance the smaller number. Whilst cold blast con-
centrates the heat in the hearth in a small compass, the hot
blast distributes the heat uniformly over the whole section
of the enlarged hearth, and also induces an increase of the
production, which sometimes amounts to as much as 50 per
cent.
Too strong and too low a pressure of the blast causes com-
bustion above the tuyere, when much heat will be lost by
the waste gases, and irregularities of the process may occur.
4. A More Regular Process,* as irregularities caused by
accidental cooling may frequently be more quickly regulated
by an increase of the temperature than by any other means.
Hot blast also facilitates and accelerates the blowing in of
blast furnaces, and the resulting slags are more thinly
liquid.
5. The Frequent Advantageous Smelting of Sulphur-
ous Ores, or Ores Difficult to Reduce and Easy to
Fuse. In the case of sulphurous ores smelted at a high
temperature, hot blast with an addition of limestone yield a
much better pig-iron than when smelted with cold blast at
a lower temperature. According to the experiments of Price
and Nicholson,‡ hot blast does not separate phosphorus.
Wrightson states that hot blast reduces more phosphoric
acid. As the temperature of the upper parts of blast furnaces
is lower with hot blast than with cold blast, the charges
pass more slowly in the furnace, allowing a better prepara-
tory heating of ores easily fusible and difficult to reduce
(puddling slags).
The hot blast may have the following disadvantages:-
1. Deterioration of the Pig-Iron.§-Pure iron ores
smelted with cold blast yield a pig iron freer from silicon
than when employing hot blast, as the higher temperature
facilitates the reduction of silicon, calcium, &c.
* B. u. h. Ztg., 1858, p. 243. LEONHARDT, Hüttenerzeugnisse, 1858, p. 101.
† B. u. h. Ztg., 1858, p. 227.
‡ Ibid., 1856, p. 73.
+
Ibid., 1850, p. 480.
§ Bgwkfd., ix., 257. WACHLER, Eisenerzeugung Oberschlesiens, Hft. 3,
p. 76.
POGG., B. 55, p. 485.
HOT BLAST APPARATUS.
509
*
2. Quicker Waste of the Furnace Walling, owing
to the higher temperatures, chiefly when smelting easily
fusible iron ores (blackbands) for a larger production.
3. An Increase of Zinciferous Deposits on the Furnace
Walling. When smelting zinciferous iron ores with cold
blast, most of the zinc volatilises, the furnace mouth gives
a strong flame and very few deposits are formed in the
throat, owing to the higher temperature there. And as, when
using hot blast, the upper parts of the furnace cool, more
deposits are formed even at the lower part of the furnace
throat; the contrary is said to have taken place at Geis-
lautern.‡
are
The advantages of the hot blast have led to its introduc-
tion into the iron works of all countries, and these advantages
the greater chiefly on account of the increase of
chemical affinity the higher the temperature required in
the furnace hearth, i.e., the more difficult to fuse are the ore
mixtures to be smelted. Cold blast is now only employed
in furnaces for the production of a very excellent and strong
foundry pig-iron (Swedish Dannemora iron) or a forge
iron as free as possible from silicon (Low Moor,§ Bowling,
Pontypool, Blaenavon, &c.), from very pure iron ores.
Owing to the increase of temperature, the hot blast further
facilitates the formation of pig-iron rich in silicon, the more
siliceous the ore mixture employed, and, therefore, the
temperature of the blast and the addition of lime must be
increased in order to unite with the silica; higher ore charges
are given, and the pressure of the blast is lessened in order
to lower the temperature in the smelting zone.
If those remedies are judiciously employed a pig-iron may
be produced approaching in quality pig-iron produced under
the same circumstances with cold blast; otherwise a pig-iron
rich in silicon, calcium, magnesium, aluminium, &c., will
* B. u. h. Ztg., 1858, p. 243. KRAUS, Oesterr. Jahrb., 1848, pp. 27, 30;
1855, pp. 436, 439.
+ B. u. h. Ztg., 1854, p. 245.
ماء
LEONHARDT, Hüttenerzeugnisse, 1858, p. 103.
TUNNER, das Eisenhüttenwesen Schwedens, 1858, pp. 14, 37. TUNNER'S
Ber. üb. d. Lond. Ind. Aust., in 1862, p. 27.
§ Preuss., Ztschr., iv., 217. B. u. h. Ztg., 1862, p. 246.
510
IRON.
result.* When using cold blast the combustion takes place
in a larger space, but is less intense than hot blast, which
produces a higher temperature in a smaller compass of the
upper hearth, the reduction therefore commences later on.
If the temperature is too high, it is somewhat lowered to
avoid hindrance of the reduction, and for this purpose an
enlargement of the hearth and lower pressure of the blast
is preferable to too heavy ore charges.
Therefore, whilst hot blast deteriorates the quality of
pig-iron when produced from good iron ores and fuel, it
improves the quality of pig-iron from impure (sulphurous) ores
and fuel, if the ore mixture is of basic composition. Some iron
ores yield, with cold blast, almost exclusively white pig-iron
even at a large consumption of fuel, whilst the same ore
produces a good grey foundry iron when using hot blast.
Hot blast is also frequently employed for the production
of white iron in order to save fuel, when the hearth is suit-
ably enlarged to prevent the temperature from rising so high
as to produce grey iron.
An injurious enlargement of the furnace hearth is induced
more by too hot a blast than by blast of too great a pressure.
In blown-out furnaces which have been carried on with hot
blast the hearth is chiefly enlarged, whilst in similar cold
blast furnaces the shaft shows an enlargement.
.
The blast for charcoal furnaces is usually heated to from
100° to 250° C., and for coke furnaces to from 250° to 350°; in
some of the iron works of this country hot blast of about
550 C. is used, whilst the coke furnaces in Upper Silesia and
Belgium sometimes employ temperatures of 60° or 80° only.
According to Mayrhofer, the percentage of fuel saved by
hot blast is equal to o'03264 t; t, expressing the number of
centigrade degrees to which the blast is heated.
The apparatus chiefly used for heating the blast are known
as hot blast ovens or stoves, and consist essentially of a series
of tubes, arranged in a chamber of fire-bricks, and heated
externally by a fire.
In order to heat a given quantity of blast to a certain
Berggeist, 1862, No. 26. B. u. h. Ztg., 1862, pp. 254, 326.
HOT BLAST APPARATUS.
5II
temperature the surface of the tubes must be of appropriate
dimensions. According to Weisbach's calculations, I square
metre of heating surface can heat about I cubic metre of
blast to 300° C., consuming about 1-15th kilo. of wood or
1-30th kilo. of coal.
As air is a bad conductor of heat the air must be exposed
to the heat in the thinnest possible currents, and the heating
will, therefore, be more advantageous the greater the pro-
portion of the circumference of the tubes to the contents of
their section, and also the longer the tubes are; for this reason.
the elliptical shape is preferred to the circular one for the
tubes, and a system of tubing is preferable to one tube in
the form of a spiral. As the resistance of friction increases
with the enlargement of the surface and the length of the
tubes, experience has taught a certain limit concerning the
size and length of the tubes. It is not advisable that the
velocity of the air in the tubes should exceed 10 or 12 metres
per second, as otherwise loss will ensue.
The hot blast ovens are heated either by a separate fire
or, as is usually the case, by the waste gases of the blast.
furnaces.
When space permits, hot blast stoves are placed at the
side of the furnace mouth, as is frequently done in char-
coal furnaces; the cold blast is then forced upwards to the fur-
nace top and the heated blast conducted down again. In order
to prevent the temperature of the blast from falling too
much (owing to the long conducting pipes) before entering
the tuyeres, these are either placed inside the rough walling
of the furnace or they are protected with bad conductors of
heat; but still the great distance of the heating apparatus from
the tuyere frequently causes a loss of temperature of from
40° to 70° and more.
In cases where the platform of the furnace top is too
small, or when too great a loss of temperature is to be appre-
hended on account of the greater height of the furnaces,
(necessitating very long conducting pipes), the hot blast
MERBACH, Erläuterungen der vorzüglichsten Apparate zur Erwärmung
der Gebläseluft, Freiberg, 1840, i., p. 80. WEISBACH, Ingenieur. u. Maschinen
Mechan., i., p. 1082.
512
IRON.
stoves on the sole of the blast furnaces are placed as close
as possible to the tuyeres, whilst the waste gases are con-
ducted down to the apparatus.
* These apparatus are
frequently provided with a spare fire-place in case the waste
gases should not be sufficiently effective.
Hot blast stoves heated with the waste gases are more
durable than stoves with a direct coal firing, owing to the
more uniform distribution of the heat. At the Ormsby Iron
Works two hot blast stoves with gas firing are more effective
than three hot blast stoves with coal firing.
The advantages of both modes of firing depend on the price
of coal and coke. The waste gases from a furnace with
closed mouth (for collecting the gases) have a temperature
of about 450° F., whilst a temperature of 1000 or 2000° F. is
produced upon burning the gases at the open furnace mouth,
consequently the smelting materials undergo a better pre-
paratory heating when the furnace mouth is kept open; and,
under the same circumstances, a furnace with a closed mouth
requires 10 per cent more fuel for the production of the same
quantity of iron. The advantages of the waste gases depend
on the value of the 10 per cent of coke in comparison with
the value of the coal required for heating the blast, &c. A
furnace producing weekly 200 tons of iron at a consumption
of 300 tons of coke yields a quantity of waste gas equivalent
to 150 tons of coal.
The hot blast stoves used chiefly on the Continent, and
named after the works at Wasseralfingen,* in Würtem-
temburg, where they were first introduced, are represented
by Figs. 148 and 149, which represent the apparatus in use
at Hasslinghausen (Westphalia). The 36 horizontal tubes, a,
each 9 feet long, 3 inches inside wide, and 13 inches high,
are arranged in the stove, and connected by the horse-shoe
tubes, b, as shown in the drawing. The cold blast enters
by c, and the hot blast issues by d; e are the fire-places;
ƒ are partition walls; g, flues. The stove is 22 feet high,
and has an area of only 8 feet by 10 = 80 square feet.
LEOB., Jahrb., 1855, v., p. 282.
† Ann. d. Min., 3 sér., iv., 77.
HOT BLAST APPARATUS.
513
Similar apparatus, of smaller dimensions, are also used for
most charcoal furnaces.
d
FIG. 148.

g
9
α
α
α
e
e
6
1
1
12 FT
The apparatus most common in this country are those
originally adopted at Calder in Lanarkshire. They are
shown by Figs. 150, 151, and 152. The cold blast enters
in the middle of the main, A, by means of c, and passing
through the horse-shoe tubes, B, issues at D' on the end of
the second main, B. The waste gases conducted from the
furnace mouth to the reservoir, D, enter the apparatus by f,
and are burned by the atmospheric air which is introduced
through the grate, g. The ash-pit is provided with two
doors. After playing round, the tubes of the first partition,
the flame passes over the partition wall, d, into the second
compartment, and the burned gases pass into the chimney, F,
through the channel, E. The dust slides down on the in-
clined planes, e. m are openings for cleansing the apparatus.
G is a slide door in the channel, E.
2 L
VOL. II.
514
IRON.
To increase the effect and to obviate various defects,
this apparatus is modified in many ways in different iron
works. The horse-shoe pipes are frequently made of an
elliptical shape; in other cases the vertical pipes, instead of
being arched at the top, are united by a short horizontal
pipe, the limbs being close together.
FIG. 149.

g
000000
000000
f
000000
000000
000
1000000
W W W W W W
e
Another modification, known as the pistol pipe, is used in
Scotland, Cleveland, and other districts in this country, and
also in Germany and France. The two vertical pipes are
replaced by a single one, divided by an internal partition
reaching nearly to the top. It is closed at the upper end, and
usually bent over into a semi-arch. One of the divisions is
connected with the entrance, and the other with the exit.
This hot blast apparatus has the disadvantage of causing
a considerable loss of temperature, as the heat is first com-
municated to the iron pipes, and from the iron to the blast,
and as the escaping flame still retains a high temperature;
the friction of the blast in the tubes also causes loss of pres-
sure; the temperature frequently penetrates the tubes a few
HOT BLAST APPARATUS.
515
inches only; the tubes are strongly attacked and with diffi-
culty kept tight, &c.
These disadvantages are avoided, and a more intense
heating of the blast is obtained, by strongly heating by a
flame a great number of uniformly arranged fire-bricks in a
FIG. 150.

D
.....
h
h
F
ރ
O
m
B
m
m
stove, then removing the flame and allowing the cold blast
to pass between the hot bricks till they are no longer red
hot. The flame is again applied, and at the same time the
blast is heating by a second stove. Siemens first applied
this principle to puddling furnaces, producing a temperature
as high as 4000°.
2 L 2
516
IRON.
Cowper has lately employed this so-called regenerative prin-
ciple of Siemens to heat the blast to very high temperatures,
FIG. 151.

D
m
9
འ ས ་
C
:11
mu
m
ジ
​m
A
B
D
FIG. 152.
D

E
LLA KURTI
B
B
m
D
ི་
་་་་་
མ་
m
HOT BLAST APPARATUS.
517
when employing gases and also direct firing. An apparatus
with direct firing is shown by Figs. 153, 154, and 155. A is
FIG. 153.

B
the fire-place; B, the regenerator; c, valve of the chimney;
D, door before the fire-place, filled inside with water; E, slide
FIG. 154.

F
E
B
A
valve in the tube for introducing the cold blast; F, a cast-
iron hemispherical valve for the hot blast. Whilst heating
518
IRON.
the furnace D and c are opened, and E and F closed.
D and
c are heated for two hours, then closed, and E and F opened.
The original flame has a temperature of about 1750° C., and
FIG. 155..

F
from 70° to 120° only when passing into the chimney, whilst
the hot blast has a temperature of 560° C. The loss of
pressure in this apparatus is not greater than in other ap-
paratus, and amounts to about 1-10th lb. per square inch.
The utilisation of the temperature is more perfect on account
of the great difference between the specific heat of the bricks
and that of the atmospheric air, for which reason a small
quantity of bricks are able to heat a great quantity of air.
These regenerative heating stoves may perhaps offer some
difficulties in practice, as the interstices between the bricks
become filled up with tar and dust from the furnace, and on
account of the number of valves, and the lowering of the tem-
perature when changing the stoves.
IRON BLAST FURNACE PRocess.
Theory of the Process.*-The treatment of an iron ore
mixture in blast furnaces aims at the reduction of the oxidised
EBELMEN, in B. u. h. Ztg., 1851, p. 321; 1844, p. 113. Bgwkfd., ii., 464;
ix., 55. SCHEERER'S Metallurgie, ii., 13. TUNNER'S Beitrage zur nähern
Kenntniss des Eisenhohofenprocesses durch directe Bestimmungen. Leob.,
Jahrb., 1860, ix., 281; 1861, x., 491; 1862, xl., 300. B. u. h. Ztg., 1860, pp.
207, 418; 1861, p. 315; 1862, p. 320. STAHLSCHMIDT, Untersuchungen über den
Cokesofenbetrieb. B. u. h. Ztg., 1860, p. 54. Von MAYRHOFER, Studien des
Hohöfners. LEOB., Jahrb., 1861, x., 277.
IRON BLAST FURNACE PROCESS.
519
iron, the carbonisation of the reduced iron, and the separa-
tion of the carbonised iron from the slag-forming components
by smelting. As iron is very refractory, but easily trans-
formed by an admission of oxygen, the ore mixture and the
fuel are charged in alternate layers in the furnace. By this
means a uniform reduction and smelting of the ore is ob-
tained by its direct contact with the coal and the ascending
reducing gases respectively, and also as small quantities only
are gradually smelted. If ore and coal were mixed together the
coal would prevent the combination of the slag-forming com-
ponents. And if the ore and fuel were charged in separate
vertical columns, as is done with lead, silver, copper, tin, and
other ores, the reduction would not take place uniformly, the
temperature in the smelting zone would be insufficient, and the
pig-iron before the tuyeres not sufficiently protected against
the blast. When charging in alternate layers, the lower
charge of fuel produces the heat sufficient for smelting the
ore charge lying above it.
The reduction of oxide of iron does not require an intimate
mixture of the iron ore with the fuel, as the reducing reaction
in blast furnaces must be attributed chiefly to the carbonic
oxide, which is formed partly direct by the combustion of
fuel before the tuyere, and partly by the reduction of carbonic
acid likewise produced by the combustion; some small quan-
tities of hydrogen, carburetted hydrogen, and cyanogen,
also exert a reducing reaction.
The chemical reactions in the different parts of blast fur-
naces have been chiefly elucidated by the analyses of Bunsen,
Ebelmen, and Playfair, of the gas of blast furnaces, as well
as by Tunner's direct investigations, and Stahlschmidt's cal-
culations; Scheerer has divided them into certain zones,
shown by Fig. 156. The capacity of these zones may be
modified by a number of circumstances, and these modifica-
tions influence the amount of the production and the quality
of the pig-iron.
1. The Preparatory Heating Zone (Vorwärmzone) is
formed by the upper part of the furnace. The smelting
materials emit volatile substances (water, carbonic acid,
sulphur, &c.,) which cause them to disintegrate and prepare
520
IRON.
them for the subsequent reduction, &c., the longer and the
more uniform is their contact with the ascending hot gases.
(Influence of narrow and wide furnace mouths, page 460).
FIG. 156.

1
400°
..C.
1000
1200°
d
α..
1600°
to
1700°
e-
2000
to
2650
STUD
ហា
Preparatory
heating zone.
Zone of
reduction.
d.
Zone of
carbonisation.
Smelting zone.
Zone of
combustion.
The sulphur volatilised in this zone from the ore or fuel is
transformed into sulphurous acid and sulphuretted hydrogen,
and is here innocuous. In this zone coke furnaces, as through-
out the furnace, are hotter than charcoal furnaces, but in
both kinds of furnace the escaping gases from hot blast are
of lower temperature than those from cold blast.
IRON BLAST FURNACE PROCESS.
521
Ebelmen* found the temperature at the mouth of cold
blast charcoal furnaces to be 112°, and 200° when the charges
sank somewhat lower; and in coke furnaces fully charged
from 228° to 330°, and from 366 to 430° C. with somewhat
descended charges. Le Blanc states that 2.5 metres below
the furnace mouth the temperature is 250° C.
Ebelingt states that if the temperature of the furnace
mouth is too high, owing to some defect in the construction
of the furnace and its accessory apparatus, or to an irregular
conduct of the process, the ascending carbonic acid will be
unprofitably reduced in the furnace mouth to carbonic oxide,
and the mixture will cake too early. These disadvantages are
most likely to occur if raw wood is employed instead of char-
coal; therefore, in wood furnaces, it is even more advisable to
collect the waste gases in order to cool the furnace mouth.
2. Zone of Reduction; this comprises the space from
the former zone down to the top of the boshes. When the
smelting mass enters the deeper and hotter parts of the
furnace the reduction of the oxide of iron by carbonic oxide
is begun; at first forming magnetic oxide, next protoxide
of iron, and, finally, metallic iron, the carbonic oxide being
converted into carbonic acid. If metallic iron is formed at
once, part of it may be re-oxidised by the excess of carbonic
acid, whilst the formation of metallic iron takes place by a
gradual reduction at a level of the belly poor in carbonic
acid. According to Gay-Lussac, the reduction of peroxide
of iron by hydrogen and by pure carbonic oxide takes place
at 400 C., but Tunner's direct investigations show that a
temperature of from 650° to 700° or even above 900° is required
for the reduction of peroxide of iron by furnace gases con-
taining a smaller amount of carbonic oxide only. The
temperature of the ascending gases is several hundred degrees
higher than that of the descending mixture.
In the reduction zone the limestone which was added to
the ore mixture loses the greater part of its carbonic acid,
and so also do raw spathic ores (at above 580° C.); the
*
Bgwkfd., viii., 440.
† B. u. h. Ztg., 1863, p. 307.
522
IRON.
carbonic acid then uselessly consumes part of the fuel and
absorbs heat. For this reason it is advisable to burn the
limestone and to roast ores containing carbonate of iron;
but a loss of fuel will always take place owing to the oxida-
tion of carbon by the carbonic acid which is formed by the
reduction of the oxide of iron, and at still higher levels in
coke furnaces than in charcoal furnaces. Iron pyrites con-
tained in the fuel loses part of its sulphur in this zone only,
and the sulphur then combines with the reduced iron; the
sulphates present (gypsum, heavy spar) are also trans-
formed into sulphides, whilst phosphates require a higher
temperature and are mostly reduced in the zone of carbon-
isation.
The reduction of zinciferous ores and the volatilisation of
zinc also begin here, whilst the zinc, entering higher levels
richer in carbonic acid, is re-oxidised and forms zinciferous.
deposits.
In order to facilitate the perfect reduction of the oxide of
iron previous to the smelting of the ores in an ore mixture
difficult to reduce, the mixture must be rendered sufficiently
refractory, thus causing a slower descent of the charges; the
ores must be suitably divided, and the furnaces constructed
with higher and wider shafts and higher and narrower
hearths; the use of hot blast is also advisable, but by these
means grey iron will always result. When the production of
white iron from an easily fusible ore mixture is intended, a wide
hearth and narrow belly and mouth are required, in which case,
according to Tunner's experiments, the reduction, requiring a
temperature of over 900°, will extend to the boshes and even
the upper part of the hearth, whilst at the production of grey
iron the reduction is usually finished in the belly. By a suit-
able mixture any desired degree of fusibility may be obtained,
but the reducibility cannot be thus regulated.
The mode of charging (page 450) and the dimensions of
the furnace shaft, chiefly the mouth, which causes a more
or less uniform descent of the charges, have most influence
on the reduction.
Ebelmen ascertained the temperature in a charcoal blast
furnace 8.4 metres below the furnace mouth, or o'63 metres
IRON BLAST FURNACE PROCESS.
523
above the belly, to stand between the fusion point of silver
and copper (1022° to 1170° C.) or above 1000 C., and the tem-
perature of a coke furnace in the belly to be about 1170° C.
3. Zone of Carbonisation.-This zone comprises about
the space between the boshes, when producing grey iron,
but it extends far into the upper hearth when producing
white pig-iron from easily fusible ores, as in that case in
equal levels no such high temperatures are produced as at
the production of grey pig-iron. According to Tunner, the
carbonisation commences at 1000° C.; at 1170° C., the fusion
point of copper, steel is produced (for instance when ce-
menting wrought-iron); and at about 1400° C. the formation
of pig-iron is finished. Tunner states the temperature in the
belly to be from 500° to 800', in the lower part of the boshes
from 1200° to 1500°, and in the upper hearth at 1500° to
2000°C. A more perfect carbonisation takes place if the well-
prepared and fused masses descend slowly into the hearth,
regardless whether the boshes are more or less steep and
the fusibility more or less easy. If the reduced iron contains
sulphur, arsenic, or phosphorus, an easily caking compound of
these substances will be formed with iron, which impedes the
carbonisation, and causes the formation of white pig-iron. An
imperfect carbonisation also causes the iron in the hearth
to combine afterwards with a larger quantity of silicon
(page 281), sulphur, and phosphorus. The carbonisation
renders the iron fusible, and protects it in the hearth from
oxidation and scorification.
The carbonising agents in the iron blast furnace process are
carburetted hydrogen and carbonic oxide (Stammer*), solid
carbon and gaseous carbon compounds (Laurent and
Despretzt), and also chiefly cyanogen and cyanogen com-
pounds, namely, cyanide of potassium in the form of gas; this
last cyanogen compound is formed from the nitrogen of the
atmosphere, the carbon of the fuel, and the potash of the
ash of the fuel or the other smelting materials, in the lower
part of the smelting zone.
DINGL., Bd. 120, p. 430.
† Polyt. Centr., 1849, p. 1343.
524
IRON.
Lossen's experiments have confirmed the theory that the
iron is first converted into steel and next into pig-iron.
Liebermeistert has founded upon this circumstance the
direct production of steel in cupola furnaces.
4. Smelting Zone. This zone reaches from the lower
part of the boshes to the level above the tuyere, where the
reduction of the originally formed carbonic acid into car-
bonic oxide is completed. If the ore is suitably reduced and
carbonised, grey or white pig-iron is produced according to
the volume of the smelting zone, which is chiefly regulated
by the proportion of the charges of ore and fuel. Although
when producing white pig-iron the greater part of the upper
hearth may be considered as belonging to the zone of
carbonisation, yet an after carbonisation also takes place in
this part of the furnace when producing grey iron, chiefly
from ores difficult to reduce in furnaces with a high upper
hearth, by the contact of the liquid iron with carbon (Mag-
netic iron ores, page 327). As the upper hearth maintains
a temperature of from 1600° to 1700° C., an ample amount of
carbon may enter the iron, and, upon slowly cooling, a grey
iron rich in graphite may be formed.
Coke furnaces chiefly facilitate the subsequent carbonisa-
tion in the smelting zone, on account of their high tem-
perature, but this temperature, on the other hand, facilitates
the formation of pig-iron rich in silicon and phosphorus.
Upon the descent of the smelting masses into the hearth,
and the continued carbonisation of the iron, the slag-forming
components react upon each other; they gradually soften,
until above and before the tuyere a perfect fusion takes place,
and the fused pig-iron and slags fall down in drops.
Smelting materials of low capacity for heat allow a more
perfect preparatory heating in the upper parts of the fur-
nace, consequently they smelt more readily later in the
smelting zone. The following is the capacity for heat of
different substances :-peroxide of iron, o'165; magnetic
oxide, o*168; quartz, o*188; felspar, o'191; steel, o'118;
* Ann. d. Chem. u. Pharm., Bd. 47, p. 150.
+ B. u. h. Ztg., 1861, pp. 137, 243.
IRON BLAST FURNACE PROCESS.
525
iron, o*114; coke, o'201; charcoal, o'242; and the ash of
fuel, 0'200; from this it follows that coke and charcoal, in
order to attain a certain temperature, must absorb a greater
quantity of heat from the ascending hot gases than the other
substances, which therefore enter the smelting zone better
prepared than coke and charcoal; coke again requires to
absorb less heat than charcoal. The high temperature
before and immediately above the tuyere causes part of the
silica to be reduced by carbon to silicon; the silicon then
enters the pig-iron and facilitates the separation of graphite
from the solidifying iron (page 297), like phosphorus (page
273). Iron also has a reducing reaction upon the silica, and
the more so the less perfect its carbonisation. According to
H. Hahn's investigations iron seems to reduce silica more
powerfully than carbon; the reduction of silica is facilitated
by a siliceous admixture (page 296), a siliceous ash of the
fuel (coke, mineral coal, page 399), high temperature (coke,
hot blast), and the great extent of the space in which the
originally formed carbonic acid is converted into carbonic
oxide (coke furnaces); the larger this space the more carbon
may be extracted from the pig-iron, chiefly if produced at
a low temperature, by the reaction of the carbonic acid. A
basic mixture rich in lime counteracts the reduction of silica,
but, on the other hand, owing to its difficult fusibility it
facilitates the reduction of the alkali and earthy metals
(chiefly aluminium, calcium, and magnesium), which like-
wise deteriorate the pig-iron. At a high temperature a certain
amount of lime combines with the sulphur contained in the
slag-forming components and the fuel, but reacts less upon
the sulphur which has already combined with iron, and
which is chiefly reacted upon by the manganese contained in
the ore mixture; this manganese also facilitates the removal
of silicon from the iron. Richter* states that upon roasting
sulphurous and manganiferous iron ores protosulphate of
iron and per- and prot-oxide of manganese are formed;
and when the temperature is raised these are converted
into peroxide of iron and sulphate of manganese. This
LEOB., Jahrb., 1862, p. 291. B. u. h. Ztg., 1862, p. 320.
526
IRON.
sulphate is transformed in the zone of reduction into sulphide
and protoxide of manganese; the protoxide of manganese
is reduced in the furnace hearth, and owing to its great
affinity to silicon forms silicide of manganese, which combines
with the sulphide of manganese and also with other metallic
sulphides contained in the liquid iron. The greater part of
the protoxide of manganese contained in the ore mixture is
scorified, owing to its slight reducibility; part is reduced by
the fuel, by the carbon of the pig-iron, and also by the iron
itself.* Whilst, according to List, the manganese in iron
amounts to 3.8 per cent only at the highest, it may be con-
siderably increased under certain circumstances (page 404).
Spiegeleisen from Siegen, for instance, contains from 8 to
14 per cent, and spiegeleisen from Theresienthal above 22 per
cent. Limet acts but little towards the removal of phos-
phorus from iron, and, according to Caron,‡ manganese does
not act at all; but phosphorus contained in the additional
limestone or in the gangue is transformed into phosphide of
calcium, and partly enters the slag, which must not then be
kept too long in contact with the liquid iron. Whilst higher
temperatures lessen the amount of sulphur and increase the
amount of silicon in iron, no such fixed rules exist for the
behaviour of manganese and phosphorus.
5. Zone of Combustion or Oxidation. This zone com-
prises the immediate space at the level of the tuyere, and as
far upwards as carbonic acid is present (page 411). It was
formerly supposed that in the level of the tuyere carbonic
acid was chiefly formed owing to the excess of oxygen, thus
causing a temperature of 2200° C. (Ebelmen), or even 2458° C.
(Scheerer), and that this temperature was lessened to 1310°
and 1670° C. respectively, when the carbonic acid was trans-
formed into carbonic oxide by contact with glowing coal above
the tuyeres in the upper part of the hearth. Tunner's direct
experiments have proved that even before the tuyere much
carbonic oxide is formed together with carbonic acid, and
↑
B. u. h. Ztg., 1860, p. 52.
HARTM., Fortschr., v., 125.
† DINGL., Bd. 168, p. 380.
|| SCHEERER'S Metallurgie, ii., 21. Bgwkfd., vii., 417.
IRON BLAST FURNACE PROcess.
527
therefore the temperature calculated by Scheerer may be con-
sidered as too high. Tunner ascertained the temperature
before the tuyere to be 2200° C., at the loop-hole of the fur-
nace in St. Stephan in Styria 1750° C., and at the production
of white pig-iron in Eisenerz, 1450° C.
+
+
According to Ebelmen, an iron wire, 0'03 metre thick, fused
before the tuyere in less than half an hour (the fusion point of
wrought-iron is from 1900 to 2100 C.); porcelain fused
almost immediately. Aubel* found that platinum melted in
the focus of a Rachette's furnace (page 458), which corres-
ponds to a temperature of from 2600° to 3000° C., and Aubel
denies that easily fusible carbide and silicide of platinum
were formed in his experiments, as Heraeus,† Richter, and
others had suggested. Becquerel|| has lately ascertained, by
means of a thermo-electric pyrometer, that the fusion point
of silver is 960°, and that of gold barely 1092° C., and by
photometric measurements he found the fusion point of plati-
num to be 1600° C., and the temperature of burning coal
2070° C. Whilst in Germany the fusion point of pig-iron is
considered to be between 1400° and 1600° C., this fusion point
has been determined by Pouillet to be from 1050 to 1100° C.,
and that of steel to be from 1300° to 1400° C. These degrees
are adopted by Schinz§ and Ziurek¶ (cast-iron, 1200°, steel
from 1300° to 1400°, wrought-iron, 1600°).
I
Tunner observed that before each tuyere a separate space
of combustion is formed, which extends in the direction of the
air current to 1 feet only at the highest, the hottest part of
the focus, about 6 inches in extent, being in the middle. The
space of combustion extends upwards about I foot, whence
it divides asunder. As an increased pressure of the blast
only slightly modifies the horizontal extension of the space of
combustion, the increase of temperature caused by narrowing
the hearth, and the use of several tuyeres (page 480) in larger
furnaces, will be obvious.
* B. u. h. Ztg., 1862, p. 392;
† B. u. h. Ztg., 1863, p. 256.
+ B. u. h. Ztg., 1863, p. 195.
+
1863, p. 272. DINGL., Bd. 165, p. 278.
DINGL., Bd. 167, p. 132.
LEOB., Jahrb., xii., 164.
|| Chem. Centr., 1863, No. 20. Allgem., B. u. h. Ztg., 1863, p. 252.
§DINGL., Bd. 169, p. 453.
ZIUREK'S technol. Tabellen, 1863, p. 344.
528
IRON.
As coke converts the originally formed carbonic acid into
carbonic oxide less easily than charcoal of the same volume,
owing to its compactness, a greater quantity of coke is required
to produce a certain quantity of pig-iron in order to obtain
a sufficient quantity of reducing and carbonising gases;
though, on the other hand, coke gives a greater heat. Grey
or white pig-iron will be formed according to whether the
smelting temperature before the tuyere is higher or lower
than the temperature required for the formation of the car-
bonised iron.
Whilst in charcoal furnaces carbonic oxide is only found
a little distance above the tuyere, the carbonic acid in coke
furnaces rises higher, and disappears at the top of higher
hearths; but in the zone of reduction it increases again. The
difference of the composition of charcoal and coke furnace
gases and their influence upon the reduction of oxide of iron
with regard to the pressure, and the transformation of car-
bonic acid into carbonic oxide by glowing coals have been
stated on page 416.
An oxidation of the iron before the tuyere by the oxygen
of the blast or carbonic acid exerts an influence upon the
temperature and the composition of the gases. If that oxi-
dation takes place by the oxygen of the blast, I litre of
oxygen produces 6216 units of temperature, or a temperature
of 2690° C., if calculated from the specific heat of the
products of combustion; consequently the ascending gases
attain a high temperature, whilst a corresponding decrease
of temperature takes place in the hearth. This explains how
the waste gases increase in temperature when iron before
the tuyere oxidises and scorifies; in this case white pig-iron
is usually formed, as the silicate of iron reacts upon the pig-
iron.
If on oxidising iron by carbonic acid 2 litres of this acid are
required, they yield 1 litre of oxygen, and produce 6216 units
of temperature; two litres of carbonic oxide are formed at
the same time, which, upon combustion, produce 6260 units
of temperature. These units must be absorbed by the de-
composition of the carbonic acid, and become latent. As
6216 units of temperature are produced, and 6260 are
IRON BLAST FURNACE PROCESS.
529
absorbed, usually no absorption of temperature (cooling) takes
place when decomposing carbonic acid by iron. The mixture
of carbonic oxide and nitrogen formed by this reaction retains
nearly the same temperature as the carbonic acid formed.
On the other hand, when oxidising iron by means of carbonic
acid, the proportion of carbonic oxide in the gases will de-
crease the more iron is oxidised before the tuyere. The car-
bonic acid decomposed by iron forms only an equal volume
of carbonic oxide, whilst one volume of carbonic acid yields
two volumes of carbonic oxide when decomposed by carbon.
This explains the fact that in an irregular process, in which
iron is oxidised at the expense of carbonic acid, the waste
gases are wanting in combustible carbonic oxide, and are no
longer fit to produce a certain temperature, which the gases
of the regular process can produce.
6. The Lower Hearth; Crucible.-In the space below
the tuyeres slag and pig-iron separate according to their
specific gravities. As the iron enters the hearth in a super-
heated state (in coke furnaces still more so than in charcoal
furnaces), having a temperature of from 1800 to 2000° C.,
whilst its fusion point is from 1600 to 1700° C., and even less,
it remains liquid for a longer time, and larger quantities may
be kept in the crucible, if an extraordinary cooling (for in-
stance, a too frequent ladling out) is avoided; otherwise the
grey iron will turn to white. Foundry pig-iron must be
more protected from the influence of the blast the stronger
the pressure of the blast (page 402); on the other hand, a
surplus of acid slag may introduce silicon into the iron if
kept too long in contact with it.
A certain loss of iron by scorification takes place even in a
well conducted process with a judiciously formed mixture.
At the best this loss amounts to from 1 to 2 per cent,
and is frequently as much as 4 per cent, and in rare cases
even over 14 per cent*-at the smelting of a mixture
rich in iron slags for the production of white iron, for in-
stance. Under equal circumstances the loss in poor mix-
tures is larger than in rich mixtures, and the chief causes
VOL. II.
*
Allgem. B. u. h. Ztg., 1863, pp. 151, 156.
2 M
530
IRON.
of it are the fusibility of the slags and their inclination to
chill.
The Reasons of the Fluctuation of the Temperature
in Blast Furnaces.
From the preceding explanations it will be obvious that the
temperature of blast furnaces chiefly originates in the zone
of combustion, and but little in the zone of reduction. The
current of gas carries the temperature from the zone of com-
bustion upwards, whilst the charges of ore and fuel move in
the opposite direction, absorbing heat and cooling the current
of gas; therefore every modification of the temperature
in the zone of combustion reacts upwards to the furnace
mouth, as every modification of temperature in other zones
spreads in both directions, in some cases lowering, and in
others raising, the temperature.
Scheerert considers the following to be reasons of an in-
crease of temperature:-
a. Reaction from Below Upwards.-Fuel of high pyrometric
effect; great relative quantity of fuel; highly condensed and
heated blast; purity of the blast; narrowness and height of
the furnaces.
b. Reaction from Above Downwards.-Dry and roasted ore;
dry fluxes and burned limestone; perfectly carbonised and
dry fuel; suitable degree of division of ore mixture and fuel,
as well as the density of the fuel; a perfect condition of the
furnace walling, and no collection of waste gases.
A decrease of the temperature takes place under other cir-
cumstances. A modification of the temperature in zones of
similar description by the action of the above-named circum-
stances causes a variable extension of the zones, and both
together exert an essential influence upon the amount of the
production and the quality of the pig-iron; this influence has
been clearly explained by Scheerer. The size of the smelting
zone chiefly affects the largeness of the production, and
the remedies for increasing the zone, within certain limits,
are the relative quantity of fuel, the temperature, pressure,
and quantity of blast, and the construction of the furnace.
SCHEERER, Metallurgie, ii., 42.
IRON BLAST FURNACE PROCESS.
531
When treating ores poor in sulphur and phosphorus, a
grey pig-iron rich in graphite and silicon is produced, if,
owing to the coincidence of circumstances mentioned under
a and b, the zones of combustion and smelting are very large,
the zone for preparatory heating very small, whilst the zone
of reduction retains a certain volume. If a decrease of tem-
perature takes place from opposite reasons, the zones will
stand in inverse proportion, and a white pig-iron poor in
silicon and carbon, with an amount of sulphur and phos-
phorus, large or small, according to the purity of the ores,
will be formed. If the circumstances stated under a for an
increase of temperature act together with the opposite cir-
cumstances which would occasion a decrease of temperature,
stated under b, the zones of combustion, smelting, and pre-
paratory heating will be increased, the zone of carbonisation
will be decreased, and the zone of reduction will remain
stationary; a grey pig-iron will then be formed poorer in
graphite and silicon than in the former case, and containing
more phosphorus but less sulphur than the former white
iron.
And, finally, if the reasons enumerated under b for in-
creasing the temperature act together with the opposite
reasons of a for lowering the temperature, the zone of car-
bonisation will attain its largest extent, and all the other
zones, except the zone of reduction, which remains as before,
will be reduced to their smallest compass. Pure, easily
fusible ores will then yield spiegeleisen; if they are less pure,
a white or mottled pig-iron results somewhat poorer in carbon
and richer in silicon, but its contents of sulphur will be
smaller than in the second case, and also its phosphorus if
the phosphorus is not contained in the ore itself. Ores diffi-
cult to reduce, which impede the zone of reduction, yield
white pig-iron richer in carbon than in the second case.
The time occupied by the charges in passing through
the furnace essentially influences the quality of the pig-
iron, as the reduction and carbonisation of the iron depend
on it. The velocity of the charges increases with the easy
fusibility of the mixture and the temperature in the furnace
hearth, and also if less compact and quickly burning fuel is
2 M 2
532
IRON.
employed. When smelting easily fusible ores with charcoal
(Styria) the charges are in the furnace 3-6-12 hours, and
up to 24 hours when smelting ores of slight fusibility. The
charges are kept in coke and coal furnaces as long as 63 hours
(page 421).
THE MANIPULATIONS IN THE BLAST FURNACE
PROCESS.
1. Drying and Warming the Furnace. For long opera-
tions the furnace walling must be very carefully dried and
previously warmed; the fuel expended is amply paid for by
the subsequent good action of the furnace.
The interior of the furnace is next to be kept about the
same temperature as a well heated dwelling-room for a longer
time (from one to several months, according to the size of
the furnace).
For this end a fire-place is sometimes placed in the
foundation of the furnace below the sole of the erection,
and is kept going, whilst the moisture is expelled through
horizontal and vertical channels, which are connected and
lead into the open air. When warming Rachette's furnaces
the temperature is gradually raised until coals thrown upon
the bottom stone will light by themselves. Furnaces, such
as those represented by Figs. 88 to 97 on pages 434 to 439,
are dried for about three months, but the length of the
warming operation depends entirely on circumstances.
For the purpose of drying, a sort of air furnace is some-
times erected before the front of the furnace at the height of
the dam, and from this air furnace the products of combus-
tion are conducted into the blast furnace, whilst the tuyeres
are closed and the furnace mouth loosely covered.
Where there is opportunity, the hot gases of other furnaces.
(reverberatory furnaces) are conducted into newly-erected
blast furnaces to warm them.*
Sometimes, also, a small fire is kept up in the fore-hearth
of the blast furnace itself, upon a temporary grate.
In order to protect the hearth walling from immediate
*LE BLANC, Eisenhüttenkunde, i., 102.
MANIPULATIONS IN THE BLAST FURNACE PROCESS. 533
contact with strong heat, it is lined with common bricks, and
sometimes with iron plates, which prevent the flying of the
stone work.
In charcoal furnaces, as soon as steam ceases to escape from
the furnace mouth, and the furnace walls have become per-
ceptibly warm to the hand, the furnace is filled with charcoal,
partly through the hearth and partly through the furnace
mouth, until nearly full. The fuel is then lighted from below,
and the tuyeres, fore-hearth, and furnace mouth are closed;
the furnace mouth is opened from time to time for the escape
of moisture. Sometimes the furnace is gradually filled with
fuel, and new charges are given when the fire has risen to
the surface. To the later charges of charcoal are added some
finely divided blast furnace slag for the purpose of glazing
the furnace hearth walling. After about two days an air-
hole about half an inch in diameter is made in the lid of the
furnace mouth and above the dam ; on the following day the
hole in the lid is enlarged. One day later a tube half an
inch in diameter is thrust through the tuyere reaching to the
middle of the hearth, and the day following it is removed,
leaving an opening in the tuyere, and the furnace lid is at
the same time somewhat opened. The ash formed is removed
from the fore-hearth about every 12 hours, whilst the charges
which have descended are again replaced; and when on the
fifth or sixth day fire appears on the surface, the first ore
charges are given, often about five or five and a half weeks
from the commencement of the warming.
Coke furnaces usually require still greater care, a longer
warming, and a more frequent removal of the ash than char-
coal furnaces. The fuel in the furnace is sometimes lighted
by means of a channel which communicates with the fore-
hearth, whilst the tuyeres are closed and the furnace mouth
covered (Belgium). The temperature and the exit of the
vapours are regulated by a damper with which the furnace
lid is provided, and in such a manner that in a few days the
fire from the channel reaches the fore-hearth, and four or five
days later the surface of the fuel. The ash and cinders are
now removed from the hearth, for which purpose a grating
of bars, carried by an external bearer laid across the rack
534
IRON.
plates of the hearth, is placed within the furnace a little below
the level of the tymp. The column of fuel is thus supported,
and the air passing through the grate bars revives the com-
bustion above. As soon as the lower part of the fuel is
glowing, the bars are withdrawn and the ignited fuel allowed
to fall into the hearth. For the first week this arrangement
of grates is repeated every 24 hours, afterwards every 12
hours, and at last every 8 or 6 hours, till the hearth and
bottom stone are of an orange-red colour without dark spots.
The first ore charge is then made.
2. Blowing in.-This is the period when the warmed fur-
nace requires to obtain the highest charge of ore and the
maximum of blast. This operation also demands much care
and time, and many expedients to shorten it have led to
damage to the furnace and a decrease of the production;
otherwise the time required for blowing in a furnace is de-
termined by the size and the materials of which it is built,
the fusibility of the ores, the quality of the fuel, &c.
At the commencement small quantities of easily fusible
slags or ore mixtures* are charged, or about a quarter of the
normal mixture, which is gradually increased to about half
till the first slag appears before the tuyere. Sometimes
limestone is given as the first charge, in order to indi-
cate the approaching ore charges, and also to scorify ash
and stones which may have separated from the hearth
lining.
In charcoal furnaces the blast is not put in action during
this time, and the air required is introduced through the hole
for the tuyere and the fore-hearth. The charges therefore
descend slowly, and the first ore charge enters the hearth in
about eight days. The hearth is now perfectly cleaned,
the bottom stone covered with small coal, the tuyeres and
dam are put in their place, cold blast is put on with low
pressure (about four lines), and the coals are drawn in the
fore hearth and covered with a layer of clay. The tuyeres
and fore-hearth are then cleaned every two or three hours
from deposits and crusts. When the crusts become more
* B. u. h. Ztg., 1862, p. 234.
MANIPULATIONS IN THE BLAST FURNACE PROCESS.
535
liquid the coals are drawn from the hearth, and the
sufficiently hot slag is allowed to drop on the bottom stone and
to collect in the hearth. The first iron, which is usually
white, may then be tapped off in 12 or 16 hours. From this
point the ore charges, and the temperature, quantity, and
pressure of the blast, are gradually increased, the pig-iron is
tapped off every 8 or 12 hours, whilst the slags are run off,
until after some weeks the furnace has attained its normal
ore charges.
At Königshütte on the Hartz, for instance, the warming
of the charcoal furnace occupies about four weeks. Three
days after filling the furnace with coal a slow descent of the
coal column takes place, and after five days the first ore
charge of 5 cubic feet is made. In ten days, during which
time 28 ore charges of 5 cubic feet each are given, slags
appear in the hearth, which is then cleaned several times
daily, frequently five times. The blast is next put on with a
pressure of four lines, and the pressure and the ore charges
are gradually increased in such a manner as to attain the
normal charge of 83 cubic feet in ten days more.
The blowing in of coke furnaces is effected in a similar
manner, with a more frequent application of temporary
grates. The larger furnaces require from eight to fourteen
days for the descent of the first charges down to the tuyeres;
and at the commencement blowpipe nozzles of from 1 to 13
inches, and at last of 3 inches or more in width, are employed,
at the same time the pressure and the number of tuyeres are
increased. When using too low a pressure (chiefly when the
tuyeres are extra wide at the commencement), the smelting
point becomes placed too near to the hearth walling.
The blowing in of furnaces in this country is described in
Ure's "Dictionary" as follows:-
"A fire of wood is first lighted on the hearth; upon this is
placed a quantity of coke, and when the whole is well ignited,
the furnace is filled to the throat with regular charges of
calcined ore, limestone, and coke, and the blast, which at
first is moderate, is turned on. At the works round Merthyr
* URE's Dictionary of Arts, &c., ii., p. 700.
536
IRON.
Tydvil the first charges generally consist of 5 cwts. of calcined
argillaceous ore and 1 cwts. of limestone to 4 cwts. of rich
coke. This burden is kept on for about ten days; it is then
increased to 6 cwts. of calcined ore and 2 cwts. of limestone
(Truran). The cinders usually make their appearance in
about 12 hours after blowing; the metal flows in about
10 hours after, collecting in the hearth to the amount of
3 or 3 tons in 60 hours after blowing. If all goes on well,
about 22 tons of metal will be produced in the first week,
38 tons in the second, 55 in the third, and nearly 80 in the
fourth; after ten or twelve weeks the produce will average
110 tons. By forcing the furnace in its infancy a much
greater produce of iron may be obtained, but to the injury of
the subsequent working. Mr. Truran relates the following
case in point:
'A furnace was blown in at the Abersychan Works with
such volumes of blast and rich burden of materials that a
cast of several tons was obtained within 14 hours after
applying the blast. The first week's blowing produced 200
tons, at which rate it continued for two or three weeks, when
it rapidly diminished, falling as low as 19 tons for one week's
make. From this deplorable state it was made to produce
26 tons, and after a considerable time 100 tons, but with a
large increase in the charge of materials over that of the other
furnaces.'
The latter instance clearly proves that the greatest care
is necessary in blowing in furnaces.
Upon applying first cold and afterwards hot blast the
ore charges may be increased in a shorter time. It is best
when blowing in to form easily fusible bisilicate slags (page
387), which, in coke furnaces, must be afterwards changed
to basic slags. Upon increasing the ore charges too
quickly, a lowering of the temperature takes place, and slags
rich in protoxide of iron are formed, which attack the fur-
nace lining; the introduction of a great volume of blast of
high pressure at the commencement of the blowing in raises.
the temperature too quickly; consequently, the hearth and
shaft wallings are injured.
MANIPULATIONS IN THE BLAST FURNACE PROCESS. 537
Dufournet* hastens the blowing in, and effects a consider-
able economy in fuel and time, by introducing the blast two
inches above the bottom stone by means of tuyeres, after
having sufficiently dried the furnace. When the furnace
hearth attains a white heat the tuyeres on the bottom are
closed with clay and the blast is introduced by the common
tuyeres situated higher.
3. Charging Ore and Fuel.-A constant supply of fuel is
usually added, whilst the charge of ore mixture is increased
or decreased according to the requirements of the process,
the quality of the fuel, &c. Only when charging by weight
is it possible to keep up a constant proportion between ore
and fuel, supposing the fuel to be uniform in quality; when
employing wet coal, the charging may also be done by the
volume, as the atmospheric influences modify the volume
less than the weight. With charcoal furnaces the fuel is
usually charged by weight and the ore mixture by measure-
ment, after having ascertained the weight of one cubic foot
of mixture.
#
in
Concerning the quantity to be taken for a charge of fuel
the rule is that the fuel charge must be sufficient to smelt
the ore charge lying immediately above, and to form a
sufficiently high layer in the furnace mouth, so that when
it spreads out in the belly the layer of charcoal will be
from 3 to 5 inches thick, and the layer of coke 4 inches.
Von Mayrhofer expresses the smallest quantity of coal or
coke required for a charge, by the formula '157 D2 (µ + 6)
which D expresses the diameter of the belly in feet, and ʼn
the weight of one cubic foot of fuel in lbs. Generally, the
size of the fuel charges depends on the dimensions of the
furnace and the quality of fuel and ore; for instance, re-
fractory ores in large lumps require larger charges than
easily fusible small ore; coke, owing to its difficult combusti-
bility, requires smaller charges than charcoal, hard charcoal
smaller charges than soft charcoal, &c. Excessive charges
of fuel require a deeper descent in order to make room
for the next charge, which thus cools the furnace throat too
* B. u. h. Ztg., 1862, p. 342.
538
IRON.
much. If the charges are too small, ore and fuel mix.
together, causing a deficient smelting; but in coke furnaces
this is not always the case, as, for instance, at Meppen*
where the charge of coke is reduced from 3000 to 1000 lbs.
= 60 cubic feet, with the best result; though, owing to the
size of the single pieces of coke, the charge of coke scarcely
forms a sufficient substratum for the ore in the furnace
mouth, 12 feet wide and having an area of 104 square
feet.
According to Le Blanc, the French blast furnaces, working
with a strong blast, use charges of from 20 to 40, or, at the
highest, 60 cubic feet of fuel in charcoal furnaces from 25 to
40 feet high, and charges of from 20 to 35 cubic feet of fuel.
in coke furnaces from 40 to 50 feet high.
Karsten states the volume of these charges to be from 15
to 30, and 12 to 20 cubic feet respectively; but in practice.
these numbers become modified in various ways.
The size of the ore charges varies according to the actual
process, which depends on the circumstances influencing the
raising or lowering of the temperature, which we have given
on page 530, and also on the intended nature of the pig-iron.
Grey iron requires lighter ore charges than white iron. If,
during the operation, the production of grey foundry iron is
to be changed to that of white forge iron, the ore charges
are gradually increased; at Malapanet when mottled iron
is to be produced for casting rolls they first form grey pig
in the hearth and then white pig by an increase of the
ore charges, decrease of fuel, and stronger pressure. For
the conversion of thickly liquid pig-iron rich in graphite,
produced by a continued regular process, into a good grey
iron, a few heavy ore charges are also given.
The operation of charging requires great care, as it
materially influences the regularity of the blast furnace
process.
The charging must be effected quickly, and therefore heavy
burdens ought to be made at once, thus preventing the
charges of coal from descending and changing their position.
* HARTM., Fortschr., v., 129.
† Berggeist, 1860, No. 99. B. u. h. Ztg., 1861, p. 438.
MANIPULATIONS IN THE BLAST FURNACE PROCESS.
539
before the ore charges are finished. Therefore, for the
operation of charging, large vessels are preferable to small
ones.
The fuel must be suitably distributed; charcoal is charged
with the smaller pieces placed upon the larger ones, in order
to prevent the lumps of ore from rolling over, to which
they have a tendency. Even with the most careful charging
this rolling cannot always be avoided as, upon the descent of
the smelting column, the heavier lumps of ore push the coal
out of its place, so that it rolls and arrives before the tuyere
when only one-third or one-fifth of the time required for its
descent has elapsed. This rolling over of the ore is least
liable to take place when the lumps of ore and fuel are of
nearly equal size, and most liable to occur when smelting
finely divided heavy ore (oolitic and pulverulent brown iron
ore) with coal in large pieces; when not kept within certain
limits this rolling may give rise to dangerous irregularities.
of the process. With fine ore (oolitic ore at Thiergarten*)
the rolling is sometimes lessened by introducing the coal
charge in a convex form, and the ore charge in a corresponding
concave manner, and also by employing the cup and cone
charger (page 450) in the furnace mouth, thus distributing
the iron ores over the circumference of the throat. For the
same purpose four-wheeled cars are sometimes employed;
these cars have a movable bottom of conical shape, which
serves as a distributor, as the conical bottom is lowered
after placing the car on rails above the furnace mouth. The
ore seems to roll chiefly in the upper part of the furnace,
where the difference between the specific gravity of ore and
fuel is not yet outbalanced by the weight of the superin-
cumbent column of smelting materials.
Usually the mixture must be so charged that the small
ore is distributed over the circumference of the furnace,
whilst the larger pieces of ore fall in the centre (page 448).
In smaller charcoal furnaces this purpose is gained by suit-
ably raking the ores, when the ore on the circumference is
beaten down. In furnaces with a wide mouth the charging
* Preuss. Ztschr., iii., A., 190.
540
IRON.
is effected in the manner stated on page 451, the most
simple of which is the charging on different points of the
circumference. At Ulverstone, in order to keep the furnace
lining clearer, the ore and fuel are distributed irregularly on
the different charging places. The mouth of the furnace
used at Ulverstone is shown by Fig. 118, on page 455. About
8 feet below the furnace mouth a skeleton of six simple
arches, a, of brickwork is erected, the centre of which is
supported by a cylinder of iron plate, b, 4 feet wide, and sur-
rounded with a refractory mantle. This cylinder communi-
cates above the furnace mouth with the tube, c, which is
made of iron plate 4 feet wide, and supported by the chimney,ƒ;
it is provided with a slide valve for regulating the current
of gas. The charging is effected by means of wooden barrows.
through the six openings, d, in the chimney, and through
the interstices of the arches.
In Rachette's furnace the ore is charged more in the
centre of the furnace, leaving the border freer in order to
prevent the mixture from coming too near to the tuyeres
when descending in the furnace.
When leveling the charges of charcoal furnaces part of
the charge is sometimes taken from the side of the tuyere
and placed on the side of the tymp, to cause the slag to pass
very hot before the tuyere, and thus prevent the formation
of deposits; on the other hand, the blast will then blow less
easily below the tymp out of the furnace, as the smelting
mass there is less thinly liquid. Deficient places in the
furnace shaft or hearth are sometimes repaired by placing
more of the ore mixture, and sometimes even quartz, on that
side of the furnace.
The regularity of the charging is of most importance with
lower furnaces and pulverulent iron ore. If the charges
sink too deep in the furnace, the new charges cool the fur-
nace, and the gases will burn too strongly, wasting the fuel;
whilst the ores have not a sufficient preparatory heating, and
when falling from a considerable height are so compressed
as to impede the ascent of the furnace gases.
In some iron works the depth to which the charges are
allowed to sink is controlled by means of an iron rod bent
MANIPULATIONS IN THE BLAST FURNACE PROCESS. 541
at a right angle, or an apparatus is applied which gives
notice as soon as the charges have descended to the necessary
depth.
We have already given the time required by the charges
for passing through the furnace, and we have also seen that
the charges descend more quickly in the centre of the fur-
nace than at the sides, which causes the ore mixture and
fuel at a certain depth below the furnace mouth to be more
or less mixed.
The different ores and appropriate fluxes for charcoal
furnaces are placed, in suitable proportions, in alternate
layers one above another, and from this heap the charges
required for the furnace are taken in a vertical direction.
In coke furnaces, ore and limestone are usually charged
separately, or the limestone is placed on the bottom of the
charging vessels, and upon it the different ores in alternate
layers (Vorwärtshütte).
4. Manipulations in the Hearth.-These are chiefly :-
a, the removal of the slag. The viscous bi-silicate slag
in charcoal furnaces with an open front wall is raked out if
the hearth is filled with it, and if it does not flow by itself
over the dam on the floor round the furnace; water is poured
over it and it is removed, whilst small coal is thrown in the
fore-hearth. Thinly liquid slag, such as that usually pro-
duced in Blauöfen furnaces with closed front walls, is either
tapped off together with the pig-iron, then cooled with water
and removed from the iron with iron rakes, or it is tapped
off separately by the slag-hole into a trough-shaped hearth
of small coal mixed with loam. An iron rod with a ring is
then put into the liquid slag, which is lifted, after cooling,
by means of a crane and chain upon an iron waggon and
carried away* (Styria). The slag of some coke furnaces
(Concordiahütte, Dudley, &c.), is treated in the same manner;
some hours after the tapping off, when the slag has again
collected in the hearth until it rises close to the tuyeres, an
opening is made in the covering of the fore hearth through
which the slag is allowed to run off, filling alternately two
*KARST., Arch., 2 R., xxv., 609.
542
IRON.
pits made for the purpose. In most cases the slag runs over
the dam direct into iron slag waggons, the sides of which are
loose; these boards are removed when the slag in them has
cooled, forming blocks of from 15 to 20 cwts. The waggons
with the slag are carried on railways by horses to where
the slags are collected in large heaps. Owing to the very
great quantity of slags which are daily formed, ample space
for stowing the slags is of great importance to iron works,
and where space is deficient artificial means are sometimes
employed by piling the slag in very high heaps; at Séraing,
for example, the slag is carried to the top of the heaps on spiral
railways with an inclination of from 4 to 42 in a hundred.
The slag of charcoal furnaces is usually submitted to a
dressing for the separation of the mechanically enclosed
iron.
1
Coke furnace slags are sometimes devitrified by allowing
them to cool slowly under cover, when they assume a basaltic
appearance, become at the same time
very hard and
tenacious, and form a good material for paving roads;*
various other applications of blast furnace slags have been
proposed. Elsnert employs slags which gelatinise upon
treatment with acids for the manufacture of hydraulic lime;
and Guettiert uses them for making manure and refractory
bricks; Kumy,|| for cement and roof-tiles; Cunynghame,§
for the production of alum, mortar, cement, and manure,
and for the purification of pyro-ligneous acid.
In this country§ a company has been formed to employ
blast furnace slags for tiles, paving, &c. At Gartsherrie,
valleys which were filled with slags have been covered with
mould and successfully cultivated.
b, the removal of the pig-iron. Foundry pig-iron from
charcoal furnaces is usually ladled out with warmed ladles
coated with loam. When much iron (12 cwts.) is to be
Von CARNALL, Preuss. Ztschr., 1861, p. 44. Oesterr. Ztschr., 1857, p. 324.
†DINGL., cvi., 321. Bgwkfd., xii., 273.
Polyt. Centr., 1839, p. 1309.
B. u. h. Ztg., 1846, p. 222.
Ibid., 1853, p. 538.
¶ Bgwkfd., xviii., 516.
MANIPULATIONS IN THE BLAST FURNACE PROCESS.
543
ladled out at once the blast is removed, and the hearth is
afterwards cleansed from deposits. Otherwise the iron is
tapped off and conducted into large pans, or into separate
hearths when intended for direct foundry purposes; or the
iron is conducted into water if it is to be granulated ;
when tapping off forge iron or foundry iron for a later re-
melting, the iron is conducted into sand or cast-iron moulds
placed in front of the furnace, whilst the blast is removed.
The tap hole is opened by driving into it a wrought-iron bar,
with a sledge hammer if necessary; the iron then runs into a
gutter, which is connected with the moulds by means of cross
gutters.
The moulds are arranged in parallel series on the side of
the chief gutter or feeder, and the transverse feeding gutters
are known as sows. Beginning nearest to the furnace the
moulds are filled with iron, one series after another, by suit-
ably shutting up the chief feeder with a sort of spade. An
iron spade is also placed in the chief feeder above the trans-
verse feeders, so as to allow the liquid metal to pass below,
whilst the slag is retained by it. The pieces of iron filling
the moulds are called pigs, and they usually have a D-shaped
section. The chief feeder immediately before the tapping
hole is beaten down 8 or 10 inches deep with refractory
sand; otherwise the iron will make its way into the sand
and easily damage the dam. Towards the end of the tapping
the blast is set in motion for some time, in order to empty
the hearth from iron and slag as much as possible; the less
often the furnace is tapped off the less the hearth will
be cooled. Iron moulds are usually used for forge iron
to chill the pigs and keep them free from sand, which adheres
to them when casting in sand moulds, and exerts an influ-
ence in the puddling process. The iron may be perfectly
whitened by throwing water upon it. Foundry pig-iron, on
the other hand, is cast in sand moulds, in order to prevent
chilling.
When tapping white pig-iron from Blauöfen the slag is
tapped off once or twice by itself, and the iron is run on to
the sand hearth immediately in front of the furnace, where it
forms plates 2 or 3 inches thick; grey pig-iron is made to run
544
IRON.
together with the slag into sand moulds, or in a pit 30 inches
in diameter formed in loose sand to prevent the iron from
boiling; the solidified crust of slag is then removed from
the liquid iron, which is formed into thin plates by throwing
water upon it, each time removing the solidified iron; the
larger the number of plates formed (usually 30 or 40 from
10 cwts. of iron) the more advantageous it is. Von Mayr-
hofer* states that iron covered with slag is freer from silicon,
and it is also supposed that pig-iron cast in thick iron
moulds, so as to form thin pigs, contains most of the silicon
on the surface, probably combined with manganese.
Upon throwing scraps of wrought-iron into the cast-iron
moulds silicon will be separated; the resulting pig-iron when
re-melted in cupola furnaces yields a more tenacious product
than when smelting a mixture of wrought and common pig-
iron in these furnaces. When the pig-iron is to have a
bright surface, it is tapped off together with a small quantity
of slag. When the surface of the iron is not to be bright,
the slag is tapped off previous to the tapping of the iron; its
quality indicates the nature of the pig-iron, and more of the
latter can be kept in the hearth and be partly decarbonised
by the influence of the blast.
To granulate pig-iron it must be conducted by means of a
sand gutter on to a concave pierced iron plate, which is placed
above a reservoir of water, and the iron run in thin jets
into the current of water; the iron in the reservoir is con-
tinually raked to prevent the formation of coarse grains.
When tapped, the pig-iron throws off a great number of
sparks, owing to an evolution of gas which is not quite ex-
plained, and to the oxidation of the particles thrown off;
and the larger the number of sparks thrown off the less.
carbonised the iron is.
White iron solidifies quickly; grey
iron remains liquid for a longer time, and throws off fewer
sparks than white iron.
After the tapping, the hearth is carefully cleaned whilst
the blast is not in motion; the tapping hole is then closed,
sometimes first with a plug of hot slag and then a plug of
*
LEOB., Jahrb., 1861, x., 327, 328, 330.
MANIPULATIONS OF THE BLAST FURNACE PROCESS. 545
sand. The fore hearth is filled with small coal and the blast
re-applied.
When the ores contain lead, part of it volatilises in the
hearth, but some of the metal finds its way through the joints
of the stones, and collects below the bottom stone, where it
condenses and is tapped off from time to time in one of the
channels of the furnace foundation; as, for example, at
Königshütte in Upper Silesia, and at Jchenberg, near
Eschweiler, where upwards of 600 cwts. of lead and about
1000 cwts. of zinciferous deposits are produced annually.
c. The cleaning of the hearth is effected by means of
straight and crooked iron bars, which in this country are
suitably supported by a movable crane suspended by a chain.
More deposits are formed in coke furnaces than in charcoal
furnaces. In charcoal furnaces, when the temperature rises
too high; the descent of the charges is accelerated by ladling
the furnace hearth empty, and by removing from the tuyere
and the back wall all deposits which may impede the sinking
of the smelting column; when there is an irregular process.
caused by the cooling of the hearth and the presence of ores
in it which have not been sufficiently prepared, the smelted
mass is kept in the hearth as long as possible, but without
allowing it to enter the tuyeres; and the quantity of blast
is also sometimes lessened. When deposits of forge iron
are formed, and the slag chills before the tuyere, the tem-
perature and pressure of the blast are increased, the de-
posits are removed, and the slag is taken from the tuyere to
prevent its becoming thoroughly cold on in contact with
the current of blast. Deposits of slag before the tuyere
which are easily removed indicate a regular process, as they
are poorer in iron and more refractory; deposits of forge
iron indicate an irregular process, and impede the free access
of the blast. A nose is only admissible if the hearth is very
much burned out.
If the smelted mass is hard and thickly liquid the forma-
tion of deposits renders the manipulations in the hearth
difficult, particularly the ladling of the iron. The formation of
forge iron frequently narrows a hearth which has been over-
widened, thus admitting of a normal process for a longer time.
VOL. II.
2 N
546
IRON.
Owing to the strong flame from below the tymp, coal and
anthracite furnaces require less manipulation in the hearth.
The manipulations are confined to shifting the smelting mass
with long thin iron rods, shovelling out the burned small coal
and slag by means of a supported shovel, partially closing
the hearth with coarsely grained slag, and cooling the hearth
with water if the flame becomes too lively; ash and fuel which
have been broken up by the manipulations are also removed
from the hearth by the blast escaping from below the tymp.
An overcharging of the furnace may cause dangerous irre-
gularities, which can only with difficulty be remedied.*
Zinciferous deposits in the upper parts of the furnace must
sometimes be removed with iron rods after having allowed.
the charges to descend sufficiently deep, and whilst putting
off the blast and closing the front wall and tuyeres.
d. Feeding the furnace through the tuyeres, with sub-
stances which either increase the temperature (coal), dis-
solve the deposits (puddling slags, fluor spar, iron blast fur-
nace slags), or partly decarbonise thickly liquid pig-iron rich
in graphite (iron ore, puddling slags). In such cases it is
advisable previously to remove all the old slag from the fur-
nace. As the introduction of cold substances is liable to
waste the hearth, they are usually charged by the furnace
mouth, or the ore charges are increased or decreased, but
the effects of this are only felt at a later time.
e. Repairs of the hearth (replacing the dam, tymp,
tuyeres, &c.) These operations are to be executed imme-
diately after the tapping off, which must be in a regular pro-
cess; and when the charges of fuel (some of which are made
without any addition of ore) enter the hearth, the blast is
removed and the furnace mouth closed. The red-hot stone
is cooled from the outside by means of blast, and new stones
which have been warmed are put in its place. In this
manner the whole of the hearth wallings may be replaced.
When replacing a tuyere, the corroded part of the tuyere
stone is filled up with clay, and the new tuyere is first made
to project further in the hearth; afterwards it is drawn back,
B. u. h. Ztg., 1859, p. 400.
BLOWING OUT.
547
which widens the hearth towards the opposite side. The
bottom stone* is likewise repaired by being beaten down with
a mixture of clay and sand; the boshes are also repaired in
like manner. Water tuyerest require repairs when they
become leaky, as they may give rise to violent explosions.
Pig-iron will also cause explosions if, owing to a wasting of the
bottom stone, it enters the lower channels for carrying off
moisture. The application of wood as fuel is also apt to
cause explosions.
5. Stopping the Furnace.-Certain circumstances, de-
ficiency of materials, for instance, may render it advisable
to stop a furnace. This is done by allowing the ore charges
to descend whilst re-filling the furnace with fuel only; the
iron is tapped off, the blast put off, and the furnace her-
metically closed after having removed the dam. The hearth is
cleaned from ash from time to time, and ore charges are given
in rapid succession when putting the furnace in blast again.
6. Blowing out. When it is necessary to blow out a fur-
nace, which is usually caused by the modified dimensions of
the furnace rendering the process irregular, and influencing
disadvantageously the quality of the pig-iron and the con-
sumption of fuel, the last ore charges are reduced and mixed
so as to fuse more easily, and the final ore charge is covered
with fuel. The smelting is then effected at the highest
possible temperature in the hearth, and at an increased pres-
sure of the blast. As soon as coal appears before the tuyeres
the pig-iron is tapped off, the blast is removed, the tuyeres
closed, the tymp removed, and the coal and remaining liquid
mass taken out of the furnace. When there is no great
hurry, the furnace is closed again and allowed to cool slowly,
and, after cooling, the portions of the furnace requiring
repairs are removed and replaced.
This method of blowing out causes the upper parts of the
furnace to be strongly attacked, as the heat increases with
the descent of the smelting column. The iron cylinders
placed in the furnace mouth for collecting the gases, and
other metallic fittings and gas tubes, are liable to melt, and
B. u. h. Ztg., 1859, p. 400.
+ Ibid., 1860, p. 38. Schles. Wochenschr., 1859. No. 22.
2 N 2
548
IRON.
must be removed as far as possible before the blowing out.
Tunner* has therefore modified this operation by keeping
the furnace full of coal until the last iron is tapped off, then
all openings leading to the furnace shaft are closed, and
after a sufficient cooling, the coals are removed by the front
wall, which is broken open for this purpose.
The length of an operationt varies according to the nature
of the furnace materials, &c., and is seldom less than one
or two years. Charcoal furnaces admit of operations of
one, two, or three, and some of them occasionally eight
years; those in Styria five or six years; blast furnaces in
Belgium from eight to ten years; and blast furnaces in this
country usually from seven to eight years, but frequently
twenty or even twenty-five and fifty years. The furnaces in
Sweden only admit of operations of three or six months,
owing to peculiar local circumstances; an iron-master, for
instance, will carry on different blast furnaces in different
localities, in order to avoid much carriage of charcoal, or on
account of deficient water power.‡
The shape of blown-out furnaces is shown by Figs. 140 to
144 on pages 475 and 476. Figs. 140 and 141 represent a
Blauofen furnace with closed front walls at Eisenerz after an
operation of five and one-third years. a is the first and b the
last position of the tuyeres; c the first and d the last tap-
ping-hole.
Figs. 142 to 144 show a furnace of Königshütte in the
Hartz which had been in operation for three and a half years.
The lower part of the furnaces is usually more enlarged
when hot blast is used, and cold blast enlarges the upper
part most.
CONDUCT OF THE BLAST FURNACE PROCESS, AND THE
INDICATIONS OF THE WORKING CONDITION OF THE
FURNACE.
*
Owing to the many different circumstances involved in the
Bgwkfd., viii., 14.
KARST., Arch., 2 R., xxi., 504. Bgwkfd., ix., 91; xi., 232. B. u. h. Ztg.,
ví., 129, 307; viii., 453; 1862, p. 414. Oesterr., Ztschr., 1853, pp. 249, 283,
367; 1856, p. 9; 1857, p. 279.
B. u. h. Ztg., 1863, p. 221.
CONDUCT OF THE BLAST FURNACE PROCESS.
549
blast furnace process, its conduct is not without difficulty.
If the process is regular, nearly all the iron is reduced, and a
perfectly liquid slag almost free from iron is formed. When
at the same time there is a correct proportion between fuel
and ore mixture so as to produce the intended pig-iron, the
regular process is called normal, and according to the nature
of the resulting iron, several modifications of the regular
process may be distinguished.
Opposite to the regular process stands the irregular process,
in which part of the iron remains unreduced, and becomes
scorified, and from which a white pig-iron of inferior quality
results. An abnormal process is caused either by an excess
of fuel or by a deficiency of slag.
Indications of the working condition of the furnace are
afforded by the nature of the pig-iron, the slag, the tuyeres,
and by the flame of the furnace mouth and of the tymp,
regardless of whether grey or white pig-iron is produced.
1. The Nature of the Pig-Iron as regards its colour in a
liquid and a cold state, its degree of fusibility, the appearance
of its fracture, the rapidity with which it solidifies, the aggre-
gations on the surface of the liquid pig-iron, the appearance of
the surface, the scintillations upon tapping, &c. The nature
of the iron may be ascertained by the appearance of the metal
in its liquid state (page 291), which serves as a rule for the
iron-founders. The liquid iron is poured into a hole made
in sand, and the hues which appear on the surface of the
metal are observed. Very hot iron shows a very light colour,
much movement of its surface, and remains liquid for a long
time. According to Schott,* the different forms of the figures
(aggregations and products of oxidation) caused by different
crystalline forms of the iron indicate its nature. He states
that iron of the regular process containing the highest
amount of carbon shows only figures with parallel lines; iron
containing less carbon, figures with lateral branches; mottled
iron produced with charcoal crosses, mottled coke iron
triangles; and thinly liquid iron containing the smallest
amount of carbon, small figures in the form of stars.
Bgwkfd., vi., 241.
550
IRON.
These figures are particularly distinct in iron containing a
small amount of sulphur. For instance, at Altenau on the
Hartz, the white surface of the iron immediately after
ladling is perfectly clear and mirror-like; the surface then
undulates, moving a small ring of graphite towards the
edges; in a few moments there appears a somewhat deepened
mirror, and afterwards the surface becomes covered with a
grey coating, in which the figures are formed till the iron
solidifies.
According to the impurity of the iron more or less scum
appears on the solidifying surface of the iron, either in the
form of small black points or in large spots with blisters
lying underneath. Iron (uebergaares) produced with an
excess of fuel shows no points at all, grey iron of the regular
process some few small points, mottled and whitish iron a
great number of black points, and white iron from an
irregular process shows large spots. These large spots move
to and fro on the liquid iron with great velocity, leaving
impressions on the convex surface of the solidifying iron;
these impressions have more or less sharp borders according
to the temperature at which the iron is produced. After solidi-
fying, iron produced with an excess of fuel shows a wrinkled
surface, owing to aggregations of graphite, and the borders are
higher than the centre; grey iron of the regular process and
mottled iron solidify with a more even surface; the surface
of mottled iron is somewhat concave, and dense castings
result from it.
Upon pouring off the liquid iron from below the chilled
surface of the sample, the inside of the solidified iron shows
different crystalline formations according to the quality of
the iron, which depends chiefly on the ore mixture; and the
same crystalline formations may be recognised on the fracture
of the iron by means of a microscope; these crystals essen-
tially influence the strength of the pig-iron.
According to Schott's* experiments, grey iron of the regular
process, containing the highest amount of carbon, shows
a crystalline formation of slight compactness intermingled
* B. u. h. Ztg., 1863, pp. 130, 334.
CONDUCT OF THE BLAST FURNACE PRocess.
551
with plates of graphite; the same iron, with less carbon, shows
a more regular crystalline composition, the crystals appear
more distinctly, intersecting each other, and containing but
little intermingled graphite; mottled iron is formed by still
more perfect crystals, with branches of a pine-like appear-
ance; thinly liquid iron with the smallest amount of carbon
shows the most perfect formation of small ramified crystals
with a scarcely perceptible intermingled graphite.
Grey Pig-Iron (page 282) is usually produced from
refractory ore mixtures in furnaces with a high, narrow
hearth (easily fusible ores require a wide hearth (page 454);
Rachette's furnace (page 461) with a proportionally wide
hearth also admits of the production of grey iron) if the ores
are first well heated and are not too rich in sulphur and phos-
phorus, if the fuel is dry and compact, if the hot blast is
employed, and other circumstances do not counteract the
production of a high temperature. These circumstances and
the large volume of the smelting zone, produced by an ample
proportion of fuel to ore and by a sufficient quantity and
pressure of the blast, are the chief conditions for the produc-
tion of grey iron, if, at the same time, the better and more
compact sorts of charcoal, or purer coke and coal poor in
sulphur and ash, are employed. As the temperature rises
(very hot blast of high pressure and compact coke, coal, or
anthracite) the impurities of the pig-iron increase, the more
siliceous the mixture employed.
Scheerer classifies as follows the differences in the regular
process which are caused by the joint reaction of the fuel
and the temperature of the blast :—
a. A lightish grey and somewhat bright iron poor in silicon
and not too rich in graphite is produced, by employing at the
smelting process plenty of fuel, chiefly soft charcoal, a cold
blast of not too high a pressure, and pure iron ores also
the smelting process must not be disturbed by ladling out
the iron. The soft charcoal employed soon becomes red
hot in the upper parts of the furnaces, facilitates reduction
and prevents the decarbonisation of the iron before the
tuyere, as it combines more easily with oxygen; it does not
tend to reduce the silicon as the temperature produced is not
552
IRON.
sufficiently high. Scheerer calls the process carried on under
the above-named conditions a "kaltgaaren gang." The
resulting iron is very strong and well fitted for forge purposes.
This process can only be carried out in countries where.
charcoal is cheap, and when quality is of more importance
than quantity. These circumstances also allow the produc-
tion of a good iron with the application of pure coke
(page 509).
6. In order to save fuel a strongly heated blast is employed,
producing a cheaper pig-iron; this is, however, almost
black, rich in graphite and silicon, and requires re-melting in
cupola or reverberatory furnaces to make it fit for castings; it
is little used for forge purposes. The lessened quantity of
fuel decreases the reducing gases, and the resulting iron is
therefore less carbonised and combines with more silicon,
owing to the high temperature of the furnace. This is the
case with the production of the Scotch dark grey pig-iron
with very hot blast and basic slags (page 341). If the blast
is not strongly heated white pig-iron will result. This pro-
cess is termed "heissgaarer gang."
*
The darkish grey Scotch pig-iron contains from 2 to 4
per cent of graphite, 0'4 to 0.8 per cent of chemically com-
bined carbon, from 2 to 3 per cent of silicon (the siliceous
ores contained 8 to 13 per cent) 0'03 to 0'07 per cent of
sulphur, and o'07 to 0'21 per cent of phosphorus. The
slags produced at the same time contain from 31 to 35 per
cent of lime, from 20 to 26 per cent of alumina, from 30 to
36 per cent of silica, and from 2.5 to 8 per cent of sulphide
of manganese.
The Normal Process stands between the two before-
mentioned. By employing a sufficient volume of blast not
too strongly heated, and with a rather larger consumption of
fuel (charcoal, coal, or coke), a moderately light grey pig-iron
containing from 3°5 to 4°5 per cent of carbon (2 or 3 per cent
of which are graphite) is produced; this iron may be used
at once for castings, and it is also fit for forge purposes,
either directly or after a previous refining when produced
* B. u. h. Ztg., 1862, p. 323.
CONDUCT OF THE BLAST FURNACE PROCESS.
553
from purer ores and fuel. The grey pig-iron most commonly
produced belongs to this variety (England, Belgium, Upper
Silesia, Westphalia, Hartz, &c.).
If the purity of the iron ores admits of it, a mottled iron
better adapted for forge and some foundry purposes may be
obtained by lowering the temperature in the zone of smelting;
for instance, by using a less hot blast or by increasing the
ore charges. Mottled iron may also be produced by smelting
roasted ores together with raw ores or iron forge slags (it is
less advantageous to feed the furnace with slags through
the tuyeres), or by producing grey and white iron alter-
nately, which are then mixed in the crucible; in Malapane*
both sorts of iron are produced separately and mixed in the
required proportion.
The iron produced in the normal process at a high tempe-
rature is usually very hot, but it may become thickly liquid
if it is exposed to cooling in an enlarged furnace hearth, &c.
If at a regular process (chiefly when smelting a too refrac-
tory ore mixture at a lower temperature), obstructions are
formed in the hearth, whilst, at the same time, a white
grained pig-iron results, and if refractory deposits accumu-
late before the tuyeres, the temperature must be raised by a
stronger heating of the blast, by a decrease of the ore
charges, and sometimes even by charging coal without ore.
If these remedies do not prove successful the furnace is fed
through the tuyeres with coal, forge cinders (page 546), or,
better, with fluor spar (page 546). If the temperature of the
furnace rises too high, owing to excessive charges of fuel-for
instance, if easily fusible ores suddenly have to be treated
instead of the refractory ores which have been used before, by
ill management of the process, or by the employment of too
great a volume of hot blast of high pressure—a darkish grey,
thickly liquid, coarse-grained pig-iron, rich in graphite (page
287) will be produced at an unnecessarily large consumption of
fuel. This iron is unfit either for foundry or for forge purposes
and the process which produces it is termed in German
"übergaar,” i.e., the normal process in excess. The iron
Berggeist, 1860, No. 99. B. u. h. Ztg., 1861, p. 438.
554
IRON.
may be freed from a part of its graphite by plunging burnt
iron into the furnace hearth, or by feeding the furnace
through the tuyeres with ironstone, fluor spar, or forge
cinders, or by giving a few heavier ore burdens (page 538),
and the further formation of such graphitic iron may be pre-
vented by a frequent removal of the slag, increasing the ore
charges, or by lowering the temperature of the blast. Upon
enlarging the quantity of blast and its pressure, and at the
same time lowering its temperature, a quicker descent of the
charges takes place, lowering the temperature in the hearth;
the same result may be obtained by ladling the crucible as
empty as possible, and removing all obstructions on the front
of the back wall, which retard the descent of the charges.
The separation of graphite is facilitated by stirring this pig-
iron with wooden poles.
Upon smelting an easily fusible ore mixture at a high
temperature in the hearth (blackband ironstone in a fur-
nace with narrow hearth, page 454), a ferruginous thinly
liquid slag is formed, owing to the imperfect reduction.
This slag strongly attacks the hearth lining, and decarbonises.
the formed pig-iron, thus causing the formation of a white
impure iron poor in carbon.
The remedies for these disadvantages are, lowering the
pressure and temperature of the blast, enlarging the hearth,
or employing an ore mixture more difficult to fuse.
A deficiency of slags in a furnace of high temperature
causes a decarbonisation of the pig-iron, as it is not sufficiently
protected against the blast, and deposits of forge iron are
formed on the tuyeres and the hearth walling. These deposits
are difficult to remove, and cause the charges to slip, thus
further increasing the obstructions. This may be observed
by the appearance of the tuyeres, and the flame escaping
from the furnace mouth and below the tymp.
To remedy this irregularity the blast and ore charges are
first lowered, the hearth and tuyeres cleaned, and the fur-
nace is fed through the tuyeres with forge cinders or fluor
spar, and easily fusible ores are added to the ore mixture.
If these expedients are of no effect, and too many obstruc-
tions are formed, the tymp is removed, and blast introduced.
THE PRODUCTION OF GREY AND MOTTLED IRON. 555
into the chilled masses through openings purposely made in
the furnace walling.
The following iron works are given as examples for the
production of mottled and grey iron. The subjoined refer-
ences show where the descriptions are to be found:
1. Charcoal in-
a. Blauöfen (page 444).-Mariazell,* St. Stephan,† and
Miesling in Styria, Witkowitz.||
I
b. Blast Furnaces with Open Front Walls.-Iron works in
the Upper Hartz,§ Ilsenburg,¶ Lauchhammer,*, Ludwigs-
hüttet, in Hesse, Eiserfeyt, on the Eifel, Wiesbaden, ||.
Veckerhagen,§, Gravenhorst, T, Theissholz,*, Thiergarten,†₂
Tangerhütte,‡₂ Upper Silesia,||2 Belgium, §, England, T₂
Sweden,*, Finland,t, Norway,‡, Polonia, Russia, §, &c.
2. Charcoal and Wood.-Iron works on the Bruns-
I
LEOB., Jahrb., 1842, p. 146; 1843, p. 96.
† Ibid., 1860, ix., 285, 301.
Oesterr. Ztschr.. 1856, p. 183.
+
Ibid., 1858, p. 193.
3
§ B. u. h. Ztg., 1851, p. 684; 1858, p. 129; 1861, p. 44;
duction, in B. u. h. Ztg., 1859, p. 113; 1860, pp. 44, 135;
pp. 44, 106; 1863, p. 13.
¶ SCHEEKER'S Metallurgie, ii., 104. Bgwkfd., vi., 241.
p. 129.
* SCHEERER'S Metallurgie, ii., 103.
I
tı Bgwkfd., ix., 198.
2
2
Pro-
1862, p. 77.
1861, p. 30; 1862,
B. u. h. Ztg., 1863,
+1 KARST., Arch., 1 R., vii., 9. SCHEERER'S Metallurgie, ii., 112.
I
I SCHEERER'S Metallurgie, ii., III.
Ι
§1 KARST., Arch., 2 R., Bd. 25, p. 235.
I
T 1 B. u. h. Ztg., 1853, p. 50. KARST., Arch., 1 R., xvii., 128.
I
*2 RITTINGER'S Erfahrungen, 1855.
t2 Preuss. Ztschr., iii., A., 190. HARTM., Fortschr., i., 166.
+ 2 Preuss. Ztschr., iii., A. 179.
||2 WACHLER, die Eisenerzeugung Oberschlesiens, Hft. 3, pp. 16, 41; Hft. 6,
pp. 86, 88.
§2 TUNNER'S Ber. über d. Londoner Industrie-Ausstellung, in 1862, p. 36.
VALERIUS, Roheisenfabrication, Deutch v. HARTMANN, 1851, pp. 100, 544.
¶½ B. u. h. Ztg., 1862, p. 169; 1863, p. 156.
2
+3
3 TUNNER, das Eisenhüttenwesen in Schweden, p. 1858.
†3 B. u. h. Ztg., 1861, p. 409.
+
+3
3 SCHEERER'S Metallurgie, ii., 117.
||3 Ann. d. Min., 1858, livr. 1, p. 89.
§3 SCHEERER'S Metallurgie, ii., 118. Bgwkfd., xx., No. 22.
556
IRON.
wickian Hartz, namely,-Zorge,* Rübeland,† Gittelde;‡
Schreckendorf|| in Glatz, Champigneulles§ near Nancy,
Witkowitz,¶ Hieflau,*, Russia,t,.
Ι
3. Charcoal and Turf.-Neustadt, I, Hieflau,, Tanger-
hutte,§, Thiergarten, T, Pillersee,*, Achthal,t, Irland,t, &c.
4. Charcoal and Coke.-Heinrichshütte, 2 near Loben-
stein, Upper Silesia, §, Siegen, T, Gaya*, in Moravia, &c.
2
2
5. Turf Coal.-Underwillert, in Switzerland, Tanger-
hütte,‡, &c.
6. Charcoal and Mineral Coal.-Wengerska Gorka||3 in
Moravia.
3
7. Coke.-Upper Silesia,§, Königshütte, T, Malapane,*,
Vorwärtshütte, Hubertshütte,‡, Hohenlohehütte, Glei-
4
B. u. h. Ztg., 1859, p. 441.
† Ibid., 1853, No. 1.
KARST., Arch., 2 R., xxv.
Preuss. Ztschr., i., B. 198.
§ B. u. h. Ztg., 1859, p. 282.
¶ Oesterr. Ztschr., 1861, p. 202.
*1 Ibid., 1861, p. 113.
ti Ibid.
I
+
+ I
1 B. u. h. Ztg., 1860, p. 331.
1 Oesterr. Ztschr., 1859, No. 41.
§1 Preuss. Ztschr., ii., 172.
I
¶ Ibid., iii., A., 190.
*
2 LEOB., Jahrb., 1854, p. 236.
t2 Ibid., xi., 45.
4
2 B. u. h. Ztg., 1863, pp. 139, 148.
+2
||2 Ibid., 1857, P. 325.
§2 Allgem., B. u. h. Ztg., 1861, p. 180.
¶½ TUNNER' SBericht. über d. Londoner Industrie-Ausstellung, in 1862, p. 37.
2
* B. u. h. Ztg., 1858, p. 149; 1860, p. 145.
3
tз Ibid., 1862, p. 264.
3
++ Preuss. Ztschr., ii., 172.
+3
||3 Oesterr. Ztschr,, 1857, p. 274.
§3 WACHLER, Eisenhohofenbetrieb in Oberschlesien: Oppeln, 1857. B. u. h.
Ztg., 1858, p. 193; 1859, pp. 204, 328; 1861, p. 169.
¶3 B. u. h. Ztg., 1860, pp. 244, 389, 399; 1861, p. 337.
*
4
WACHLER, Geschichte d. ersten. Jahrh. der Eisenhüttenwerke zu Mala-
pane: Glogau, 1857.
14 HARTM., Fortschr.. iii., 145; iv., 119; v., 105.
‡4 Ibid., v., 95, III.
||4 Ibid., iv., 115.
THE PRODUCTION OF GREY AND MOTTLED IRON. 557
witz,* Laurahütte,† Tarnowitz, Hasslinghausen,|| Hörde,§
Johannishütte, ¶ near Duisburg, Henrichshütte*, near Hat-
tingen, Porta Westphalica,†, Georg Marienhütte,‡, Mep-
pen, Mühlhofen,§, Kladno, T, Hof*, in Bavaria, Belgium,†₂
England, 12 Ulverstone, 2 Low Moor,§, Ystalyfera, T₂ Aber-
dare,*, &c. France, t, Creuzot, t, Decazeville, |, Maubeuge, §,
Fourchambault,¶¸ &c.
+
+2
3
3
8. Coal Alone, or in Mixture with Coke.-England, Dow-
lais, Aberdare,t, &c. Upper Silesia (experiments with
*
coal).
4
4
* HARTM., Fortschr., i., 187; iii., 144 ; vi., 139.
† Ibid., iii., 145; vi., 136.
† Ibid., iii., 145.
+
|| B. u. h. Ztg., 1858, p. 73; 1860, pp. 54, 284, 427.
§ HARTM., Fortschr., i., 188; iii., 149.
¶ Schönfelder, c., i., 1 Jahrg., 2 Lief., 1861.
* HARTM., Fortschr., i., 188; ii., 214. B. u. h. Ztg., 1858, pp. 106, 218.
I
I
†ı B. u. h. Ztg., 1859, p. 156.
Berggeist, 1859, No. 61,
+
+ I
1 HARTM., Fortschr., ii., 241.
1 Ibid., v. 103, 115.
§1 Allgem., B. u. h. Ztg., 1863, p. 59.
1 Ibid., 1863, p. 88.
I
2 Ibid., 1860, p. 293.
12 VALERIUS, Roheisenfabrikation, Deutch. v. HARTMANN, 1851, pp. 473,
498. KARST., Arch., 1 R., vii., 318; 2 R., xxi., 584; xxiii., 661.
+
2 VALERIUS, Roheisenfabrikation, Deutch. v. HARTMANN, 1857, p. 523.
ECK and CHUCHUL, in KARST., Arch., 2 R., xxv., 589. STENTZ, in Preuss.
Ztschr., iii., 81. B. u. h. Ztg., 1855, p. 278. TRURAN, in B. u. Ztg., 1856,
P. 335. Allgem. B. u. h. Ztg., 1863, p. 21. GRUNER and LAN, in B. u. h. Ztg.,
1862, p. 42. HARTMANN, Fortschr., vi., 140. JORDAN, in Allgem. B. u. h.
Ztg., 1863, p. 13. ROLLIN, in B. u. h. Ztg., 1862, p. 413.
||2 TUNNER, Ber. üb. d. Londoner Industrie-Austellung, in 1862, p. 32.
$2 Preuss. Ztschr., iv., B. 217. HARTM., Fortschr., i., 206. B. u. h. Ztg.,
1857, p. 142.
T2 B. u. h. Ztg., 1862, p. 429.
*3 Ibid., 1862, p. 413.
†3 Ibid., 1842, p. 80, 1860, p. 450.
3 KARST., Arch., 1 R., vii., 303.
3 B. u. h. Ztg., 1844, p. 417; 1845, p. ISI.
§3 SCHEERER'S Metallurgie, ii., 134.
93
B. u. h. Ztg., 1845, p. 154.
4
KARST., Arch., 2 R., xxv., 589. B. u. h. Ztg., 1845, PP. 481, 977; 1846,
pp. 226, 987; 1848, p. 5; 1862, p. 428. Preuss. Ztschr., iii., B. 86. TUNNER,
Ber. üb. d. Londoner Industrie-Austellung, in 1862, p. 52. KLOCHE, Dowlais
Works, Stettin, 1850.
†4 B. u. h. Ztg., 1862, p. 413.
+4 Berggeist, 1862, Nos. 18 to 25.
558
IRON.
9. Anthracite.-Scotland (Gartsherrie, Govan, Mock-
land, Dundyvan, Calder), South Wales,† (Yniscedwin,‡
Ystalyfera||), Pennsylvania.§
10. Anthracite and Coke.-Ystalyfera.¶
Similar examples are also to be found in the following
treatises on the metallurgy of iron, which may be recom-
mended for the study of the iron smelting process:-
JOHN PERCY, M.D., F.R.S., Metallurgy: the Art of Extract-
ing Metals from their Ores, and Adapting them to Various
Purposes of Manufacture. London: 1864.
W. FAIRBAIRN.
Iron its History, Properties, and Pro-
cesses of Manufacture. Edinburgh: 1861.
TRURAN. The Iron Manufacture of Great Britain, Theo-
retically and Practically considered, 2nd edition, revised
by J. Arthur Phillips and William H. Dorman. London:
1862.
G. WILKIE. The Manufacture of Iron in Great Britain.
London and Edinburgh: 1857.
KARSTEN. Handbuch der Eisenhüttenkunde.
nebst Atlas. Berlin: 1841.
5 Bände
LE BLANC UND WALTER. Practische Eisenhüttenkunde,
Deutch Bearbeitet von C. HARTMANN, 2 Bde., und 3 Sup-
plem. mit Atlas. Weimar: 1837-1841. Fortsetzung
unter dem Titel; Practische Eisenhüttenkunde von C.
HARTMANN. 3 u. 4 Band. Weimar: 1843—1846.
OVERMAN'S Treatise on Iron. New York 1851.
:
SCHEERER'S Lehrbuch der Metallurgie.
HARTMANN'S Grundriss der Eisenhüttenkunde.
1852.
Berlin :
* B. u. h. Ztg., 1843, p. 845; 1852, p. 99; 1854, pp. 144, 225; 1855, p. 278;
1862, p. 302.
Preuss. Ztschr., ii., Bd. 96; iii., Bd. 81. KARST., Arch., 2 R.,
xxv., 601. Polyt. Centr., 1856, p. 998. SCHEERER'S Metallurgie, ii., 135.
† B. u. h. Ztg., 1862, pp. 323, 413.
+ Ibid., 1862, p. 428.
**
|| Berggeist, 1858, p. 51. Preuss. Ztschr., iii., p. 81. B. u. h. Ztg., 1862,
P. 430.
§ Preuss. Ztschr., ii., p. xxii. Ann. d. Min., 5 sér., xix.,
¶ B. u. h. Ztg., 1862, p. 429.
490.
CONDUCT OF THE BLAST FURNACE PROCESS.
559
HARTMANN. Fortschritte der Eisenhüttenkunde mit 17
Tafeln. Berlin: 1851.
VALERIUS. Traité pratique de la Fabrication de la Fonte.
1851.
JULLIEN.
Theoretisch-practisches Handbuch der Eisen-
Deutch von Hartmann.
hüttenkunde, nebst Atlas.
Brüssel und Leipzig. 1861.
Die Fabrikation des Eisens. Von E. FLACHAT, D. BARRAULT,
und J. PETITE. Lüttich und Leipzig: 1851. Als syste-
matischer Text hierzu. HARTMANN: Practisches Hand-
buch d. Roh., und Stabeisen bereitung. Leipzig: 1857.
HARTMANN. Die Fortshritte des metallurgischen Hütten-
gewerbes. I.-VI. Leipzig: 1858—1863.
HARTMANN. Handb. d. Eisengewerbskunde. Leipzig:
1860.
WENIGER.
Practisher Schmelzmeister.
Carlsbad: 1860.
HARTMANN. Vademecum für den Practischen Eisenhütten-
mann, 3 Aufl. Hamm.: 1863.
White Pig-Iron (page 267).-This kind of iron is pro-
duced at a moderate temperature from easily fusible ores.
whilst employing either charcoal, coke, or mineral coal, and
either cold or moderately heated blast; the furnaces
employed have a more or less wide hearth, and a smelting
zone reduced to a small volume. As coke and hot blast
increase the temperature of the smelting zone, they neces-
sitate an enlargement of the hearth. White coke iron is
usually more impure than white charcoal iron, both being
produced from the same kind of pure iron ores; but white
charcoal iron produced from an impure ore mixture is usually
more impure than white coke pig-iron produced from purer
ores.
The process in the furnace must be modified according to
whether the production of one or the other sort of white iron
is intended; and more white, iron generally enters the slag,
even at a regular process, than grey iron.
Spiegeliesen (page 269) is nearly free from silicon, sul-
phur, and phosphorus, and is usually saturated with carbon.
It is produced from roasted, manganiferous, brown, or
spathic iron ores, free from sulphur and phosphorus, in
560
IRON.
furnaces with a wide hearth (large zone of carbonisation
and small smelting zone) and narrow mouth (page 453),
usually employing charcoal and hot blast (Styria); some-
times pure coke and cold or hot blast are used (Siegen,
Hörde, Hochdahl). The manganese contained in the iron
ores purifies the iron, facilitates its carbonisation, renders
the spathic ores of easy fusibility, whilst the porosity
of these ores facilitates the reduction and carbonisation,
assisted by the construction of the furnace. The applica-
tion of coke and hot blast probably causes the formation of
a pig-iron containing manganese and silicon (Siegen), and
Krupp prefers this iron for the manufacture of steel. A
certain amount of sulphur and phosphorus in the iron
prevents its saturation with carbon (page 301), and may
give rise to the separation of graphite. The production of
spiegeliesen from iron forge cinders and Franklinite has
been mentioned on pages 270 and 327.
Hohenegger* considers the formation of a mono-silicate
slag, which is rendered of easy fusibility by a certain amount
of protoxide of manganese, essential for the production of
spiegeleisen.
According to Rammelsberg's + latest investigations, the
suggestions of Karsten (page 271), and Gurlt (page 271),
concerning the constitution of the different kinds of pig-
iron are erroneous. Rammelsberg states that spiegeleisen
contains upwards of 15 per cent of graphite, and a variable
amount of chemically combined carbon, sometimes as little
as 3.1 per cent. He proves by analysis that neither Gurlt's
tetracarbide nor octocarbide have any existence, and con-
siders the pig-iron as an isomorphous mixture of its com-
ponents, thus explaining the change in its composition;
white and grey iron differ from each other, as iron and
carbon are contained in them in heteromorphous states.
The chief improvements in the production of spiegeliesen
(Styria, Siegen, &c.) consist in enlarging the furnaces and
the quantity of blast, in the employment of hot blast, in the
* Oesterr. Ztschr., 1863, p. 307.
† ERDM, J. f. pr. Chem., Bd. 89, p. 393.
+ LEOB., Jahrb., 1857, vi., 176. B. u. h. Ztg., 1857, p. 263.
+
CONDUCT OF THE BLAST FURNACE PROCESS.
561
improved methods of roasting, &c., and they have resulted
in an increased production and a saving of fuel.
*
Spiegeleisen is produced in Blauöfen in Styria (Vordern-
berg, Eisenerz, Hieflau, Turrach,|| Neuberg,§ Hütten-
berg, ¶ &c.), in Carinthia,*, (Lölling,t, Lavantthal,;, &c.)
Siegen,|| Thuringia.§, (Katzhütte,¶, &c.), Lombardy,*2
Tyrol, (Jenbach†₂).
I
At Gittelde‡2 on the Hartz, spiegeleisen is produced in
furnaces with open front walls.
B. Flowery and Porous White Iron (Blumige and
Luckige Flossen) (page 276).-These kinds of pig-iron are
produced under almost the same circumstances as spiegel-
eisen; they are formed if the temperature is somewhat lowered
by increased ore charges, &c., and if the very easily fusible
mixture remains for a shorter time in the zones of carbon-
isation and smelting, thus impeding the carbonisation.
LEOB., Jahrb., 1842, pp. 125, 146; 1843, p. 96; 1861, pp. 281, 402, 403, 404,
407; 1862, p. 301.
† B. u. h. Ztg., 1853, p. 789. LEOB., Jahrb., 1860, ix., p. 284. Oesterr.
Ztschr., 1855, P. 374.
B. u. h. Ztg., 1847, p. 753; 1853, pp. 223, 256; 1861, p. 5. Oesterr. Ztschr.,
1853, pp. 85, 120, 249, 283; 1855 p. 374; 1856, p. 21; 1858, pp. 87, 121.
LEOB., Jahrb., 1862, xi., 299. KRAUS, Jahrb., 1848, p. 30; 1849. p. 62; 1855,
pp. 63, 445.
|| B. u. h. Ztg., 1849, p. 401.
§ Oesterr. Ztschr., 1855, pp. 126, 241; 1858, p. 104.
¶ LEOB., Jahrb., 1841, p. 225; 1842, p. 77.
I
Preuss., Ztschr., 1857, v., Lief. 3. Bericht der K. K. Berghauptm.,
1858, p. 88. B. u. h. Ztg., 1857, p. 105. Oesterr. Ztschr., 1859, No. 22. B. u. h.
Ztg., 1860, p. 103.
I
tı LEOB., Jahrb., 1842, p. 220; 1843 and 1844, p. 82. MERBACH, Anwendung,
der erhitzten Gebläseluft, 1840, p. 6.
Ztg., 1861, p. 51.
Oesterr. Ztschr., pp. 87, 121. B. u. h.
+ I
+ 1 B. u. h. Ztg., 1862, p. 339.
1 Schles. Wochenschr., 1860, No. 18.
I
B. u. h. Ztg., 1856, p. 314; 1857, P.
408. HARTM., Fortschr., i., 166. JACOBI, das Berg- Hütten- und Gewerbe-
wesen im Regierungsbezirk Arnsberg, 1857, p. 37. TUNNER, Ber. über d.
Londoner Industrie-Ausstellung, in 1862, p. 37. Berggeist, 1863, No. 23.
§ı Ann. d. Min., 4 sér., ii., 231. Berggeist, 1860, No. 60.
T1 Berggeist, 1860, p. 432. B. u. h. Ztg., 1861, p. 5.
2 B. u. h. Ztg., 1844, p. 102. KARST. Arch., 2 R., xviii., 325.
+2 Oesterr. Ztschr., 1855, Nos. 47, 48; 1856, No. 2.
+
+2
KARST. Arch., 2 R., xxv., 261. KERL, Com. Unterharz. Freiberg, 1853,
p. 91. Polyt. Centr., 1858, p. 107. Preuss. Ztschr., ii., 126.
VOL. II.
2 0
562
IRON.
These kinds of iron contain 4 and 3 per cent of carbon re-
spectively. White and slightly porous iron (kleinluckiges)
has the smallest amount of carbon and is the most difficult to
fuse. Eggertz* found in such an iron 3.3 per cent of carbon.
If the small pores are dull looking, owing to the oxidation
(gekrauste Flossen) which usually takes place when the fur-
nace contains scaffolds, the iron is still poorer in carbon and
very difficult to fuse. As the formation of these varieties of
iron requires a comparatively low temperature, the process
for their production cannot be continued for any long time
unless the ores are of very easy fusibility, and the alternate
production of spiegeleisen is required. The production of
slightly porous white pig-iron chiefly gives rise to many
obstructions, and therefore charges of fuel are given from
time to time without the additional ore charges; this, how-
ever, is only effective in small furnaces. In higher fur-
naces the ore charges are changed systematically in the
following manner: the charges required for one tapping
of the furnace are increased by, say, 10 or 20 lbs. above the
normal charge, and the charges for the following tapping are
lessened again by 30 or 40 lbs., thus dissolving the old
obstructions and forming new ones. This sort of iron may
also be produced by feeding the furnace through the tuyeres.
In Blauöfen larger obstructions sometimes make it necessary
to raise the tuyeres and tapping hole to a higher level.
irregular process is remedied by frequently tapping the slags
and lowering the blast, and also by suddenly strongly heating
the blast. The slags formed in the production of white pig-
iron with small pores are richer in iron than slags formed in
the production of iron with large pores.
An
These kinds of iron are chiefly produced in Styria and
Carinthia.
White Pig-Iron Produced at a Regular Process
(page 279). This kind of iron, used for conversion into
wrought-iron, is produced from ores which are not perfectly
pure, and contain sulphur and phosphorus; coke and hot
* EGGERTZ, colorimetr. Bestimmung des Kohlenstoffs in B. u. h. Ztg.,
1863, p. 280.
CONDUCT OF THE BLAST FURNACE PROCESS.
563
blast are generally employed, thus producing a somewhat
higher temperature than is used for the production of
spiegeleisen, and the resulting iron is poorer in carbon but
richer in silicon, sulphur, and phosphorus. Iron containing
a larger amount of phosphorus is no longer of any use even
if a higher smelting temperature is employed; this also
increases the amount of silicon. According to Berthier, an
iron of this kind from Firmy contained-
C.
S.
Ꮲ .
Si.
I' + t
0°30
2°30
410
The easy fusibility of the ore mixture required is induced
either by manganese, which has a purifying action and
causes the formation of good slags poor in iron, or by
iron,* for instance, by adding a large quantity of iron forge
cinders to the ore mixture (production in Wales of white
iron of inferior quality intended for rails). In this case the
slag may contain 20 per cent of oxidised iron, causing a great
loss of iron (in Wales several thousand tons annually). An
attempt to produce white iron from refractory ore mixtures by
increasing the ore burdens would speedily give rise to
obstructions in the furnace. Sometimes, in a new furnace,
grey foundry iron is produced at the commencement and
white iron later on, after the hearth has been sufficiently
enlarged. This kind of iron is frequently produced in
England, Belgium, Westphalia, &c.
White Pig-Iron Produced at an Irregular Process
(page 280). Its production is caused by an imperfect
reduction and carbonisation, consequently oxidised iron
enters the hearth, part remaining in the slag and part
decarbonising carbide of iron which has been formed;
the resulting iron is then poor in carbon, and also rich in
sulphur and phosphorus if produced from impure ores.
Eggertz found 2.7 per cent of carbon, in such an iron, which
had been decarbonised by the thinly liquid ferruginous slag,
and was even to a certain degree ductile.
*
B. u. h. Ztg., 1862, pp. 209. 254, 323, 413.
202
564
IRON.
The insufficient preparatory heating of the ores, which in-
duces the irregular process, may be caused by many circum-
stances, such, for instance, as the following:-a lowering of
the temperature by too heavy and too moist ore charges; the
application of an inferior, moist, friable, and small fuel rich
in ash; if the ore mixtures are either too rich or too poor,
either too easy or too difficult to fuse, or if they are not of
uniform fusibility and reducibility; if the ores are over or
under roasted, if they are wet and friable with much
loam, thus forming lumps, or if they contain silicates of
iron, such as iron forge cinders, which accelerate the descent
of the charges; an irregular ascent of the gases caused by
too great a division of the ore or fuel; an imperfect reduction
if the ore pieces are not sufficiently broken up; inappropriate
dimensions of the furnace, chiefly if it is too wide, too narrow
or too low, if it has too high or low a hearth, too straight
or flat boshes, &c.; wrong proportions of the blast with
regard to quantity, pressure, and temperature; irregular
charging, &c. Von Mayrhofer* has given a comprehensive
review of all the circumstances which may induce an
irregular process, together with the expedients for restoring
regularity.
The consequences of an irregular process are, chiefly, an
inferior white pig-iron (not to be confounded with the white.
chilled pig-iron, page 281), a small production, loss of iron,
the formation of a thinly liquid slag which strongly attacks
the furnace lining and does not efficiently protect the iron in
the hearth against the influence of the blast, a decrease of
the waste gases, so as to render them insufficient for heating
the blast (page 529), and, above all, the formation of obstruc-
tions or scaffolds and deposits, chiefly of malleable iron,
causing a slipping or retaining of the charges, sometimes
even sufficient to completely choke the furnace (especially
if the boshes are too flat, the ores and fuel too small, and
the pressure of the blast too high, so as to cause the iron
to cake).
* Bgwkfd., ix. 67. B. u. h. Ztg., 1843, pp. 688, 720; 1845, p. 689. KRAUS,
Jahrb., iii., I. HARTM., Vade-mecum, f. d. pr. Eisenhüttenm., 1863, p. 227.
† Bgwkfd., viii., 455.
THE NATURE OF THE SLAGS.
565
In order to find the best means of remedying an irregular
process, the causes of the irregularity must be each time ascer-
tained by noticing the accompanying appearances, and
whether the irregular process be periodical or continued. If,
for instance, the temperature in the hearth has decreased, the
inferior iron and all deposits and obstructions must be quickly
removed from the crucible; lighter ore charges must be
given, the temperature of the blast increased, and its
quantity decreased by employing smaller tuyere nozzles,
and moderating the motion of the blast engine; sometimes
the pressure must be diminished to retard the descent of the
charges, the furnace is fed with coal through the tuyeres,
much slag is kept in the hearth, the ladling of iron from the
furnace is avoided, &c. During these manipulations care
must be taken to prevent the iron from entering the tuyeres,
as it will quickly solidify, and exclude the blast from the
furnace. In such cases the pressure and temperature of the
blast is increased, and it is passed through other tuyeres that
have not been blocked up, and if this proves insufficient, the
blocked-up tuyere is opened by force. If, owing to an over
refractory ore mixture, the process is not benefited by these
remedies, and continues irregular for a longer time without
improvement, the pressure of the blast must be increased,
the mixture rendered more fusible, or a too easily fusible
mixture is rendered more refractory. If the ores are not of
uniform reducibility, a smaller pressure of the blast is
advisable, or the mixture must be rendered more refractory.
If white iron is produced instead of grey, owing to too
great an enlargement of the furnace hearth, the hearth may
frequently be narrowed by charging quartz on the enlarged
side of the furnace. When the charges are retained in
their descent by obstructions, the blast is immediately
shut off by closing the tuyeres, and sometimes extraordinary
measures are required, such as raising the tuyeres, and
introducing cast-iron pipes to conduct the gases.*
2. The Nature of the Slags with regard to their con-
B. ur. h. Ztg., 1859, PP. 294. 399.
566
IRON.
sistency in the hearth, and when running above the dam, as
well as their appearance after cooling.
a. Slags resulting from the regular process at the produc-
tion of grey iron.
These slags from charcoal furnaces are mostly bisilicates;
when liquid they are viscous, they cool slowly, and are lighter
in colour the more regularly the process is carried on. Some
of these slags also are coloured, chiefly grey and blue, and
the grey show a violet tint when drawn into thin threads
(iron works on the Upper Hartz). Upon quickly cooling
they remain vitreous with a conchoidal fracture, and are
more or less transparent at the edges; when in contact with
water they take a pumice-like appearance of a white, greyish
or greenish white colour, whilst sulphuretted hydrogen is
evolved. When slowly cooled, they have an opaque stony
kernel, usually with a splintering fracture. At a regular pro-
cess in excess, i.c., at a very high temperature of the furnace
hearth, the slags formed are very viscid, and may be drawn
into light threads; after cooling they are of a light colour,
inclosing grains of iron and graphite, or graphite coats the
slag; sometimes they are stony with a porous texture, and
are easily cooled by the influence of the blast. This porosity
always indicates too great a refractoriness of the ore
mixture, easily causing the scorification of iron, and then
forming a viscid green or dark slag.
Owing to a lower state of silication, coke furnace slags are
more thinly liquid than charcoal furnace slags, they chill
more quickly, and after cooling they are opaque with a less
conchoidal fracture, are thoroughly stony or crystalline,
and sometimes stony in the middle and vitreous on the out-
side; they are seldom vitreous throughout, even when most
rapidly cooled.
Their colours are light, frequently pale.
yellow, leek or olive green (probably owing to the presence.
of manganese) and blue. Protoxide of manganese gives a
light green colour, peroxide of manganese a violet hue.
b. Slags resulting from the regular process for the pro-
duction of white iron. Owing to the presence of a larger
amount of manganese they are very liquid and chill quickly,
showing a light yellowish or yellowish green colour,
THE NATURE OF THE SLAGS.
567
but they do not boil before the tuyere, and when solidified
they vary from compact to very porous, and are vitreous or
stony according to the rapidity of cooling. Upon causing
the easy fusibility of the slags by a greater or smaller amount
of protoxide of iron, they assume a colour varying from light
to darkish green, and appear to boil up, owing to an oxida-
tion of the carbon in the carbide of iron by protoxide of iron.
Some iron will always be scorified, even at the most regular
process.
According to Klein* a normal slag from Vordernberg (very
easily fusible, lightish green, from stony to vitreous, and
forming but little thread) was composed as follows:-
3
SiO3
Al₂O₂
CaO
MgO
FeO.
45'7
5'6
MnO
S..
21.7
7.8
7'3
7'5
0° 113
It stands between mono- and bisilicate, its proportion
of oxygen standing as 24: 15.
A slag yielded at the production of spiegeleisen in Lohe,
analysed by Tiemann, contained-
3
SiO,
ALO
CaO
MgO
FeO
MnO
KO.
S
41'45
2.18
23'57
5'61
I'09
21'07
I'75
I'99
c. Slags resulting at an irregular process. Owing to the
presence of iron they are more fusible and chill more quickly;
charcoal furnace slags after cooling are green or black in colour
and glass-like in appearance; they are more or less blistered
owing to the formation of gas by the reaction of the ferri-
ferous slag upon the carbon of the iron, which reaction also
causes the slag to rise in the hearth. When the slag loses
*
LEOB., Jahrb., 1862, xi., 301.
568
IRON.
its vitreous nature and becomes dull, thickly liquid, and
earthy, showing a darkish green or black colour and inclosing
many grains of iron and particles of the ore mixture, &c.,
the irregular process in charcoal furnaces has reached the
highest point of irregularity.
The commencement of an irregular process in coke blast
furnaces is indicated when the green or blue colour of the
slag changes to brown; upon an increase of the irregular
process the colour becomes darker, duller, and lead-like, and
the slags are similar in appearance to the rich (gaar) slags
resulting at the conversion of pig-iron into wrought-iron in
open hearths.
Slags produced from a mixture containing silicates of iron
are always richer in iron than if produced from a mixture
without such silicates.
3. Behaviour of the Smelting Mass before the Tuyeres.
-The appearance of the tuyeres, such as the intensity of
their glowing, the regularity of the smelting, and the forma-
tion of deposits are the chief means of judging the process.
At a regular process the mouth of the tuyere is free from
deposits, and so brilliant as to make it impossible at the first
glance to discern anything in the hearth; the coal can only
be clearly perceived after looking into the tuyere for some
time, when we also see the regular, steady, quick falling of
heavy, round drops of grey iron (spiegeleisen shows white
hot pieces which quickly disperse; pig-iron poor in carbon
shows darker and more compact pieces) and slowly flowing
slag. The formation of a nose shows that the mixture has
been too refractory, the blast too moist, the quantity of
blast too large, &c., and the remedies mentioned on page
565 are to be applied. Unreduced pieces of ore entering
the hearth are of a darker colour, and sometimes fly into
pieces with an explosion. Too high a temperature of the
hearth increases the brightness of the tuyeres and also the
inclination to form deposits. A deficiency of slag-forming
components (page 554) facilitates the formation of malleable
iron.
At an irregular process the brightness of the tuyeres
decreases, and coal, pig-iron, and slag may at once be
BEHAVIOUR OF THE FURNACE FLAME.
569
clearly observed through the tuyeres. Owing to an oxidation
of the pig-iron the slag before the tuyere boils up, and by
its easy chilling facilitates the formation of a nose and the
formation of malleable iron; the disadvantages thus caused
may be remedied as stated on page 565. In coke furnaces
the brightness of the tuyeres is chiefly changeable in its
intensity.
4. Behaviour of the Flame from the Furnace Mouth
and from Below the Tymp.-The colour of the flame is
partly formed by the combustion of certain gases (carbonic
oxide, hydrogen, carburetted hydrogen) and partly by solid
particles (oxides of lead, zinc, and antimony, silica, carbonate
of potash, coal-dust, dust of ores and fluxes, ash of fuel, &c.)
which are suspended in the flame. Oxide of zinc causes a
white colouring of the flame, oxides of lead and antimony a
yellowish white colouring, carbonate of potash, coal and ore
dust, and fluxes reddish yellow and yellowish red colours.
According to Müller* the carbonate of potash plays an im-
portant part in the colouration of the flame of iron blast
furnaces. At a regular process the potash contained in the
ash of the fuel is reduced in the furnace hearth, the potas-
sium ascends in the furnace, is then converted into carbonate
of potash, and changes the colour of the furnace flame,
originally blue from the carbonic oxide gas, to violet. If the
process becomes irregular through a decrease of temperature,
the reduction of potash will not take place, and the furnace
flame appears more or less yellow coloured, causing an in-
crease of temperature in the furnace mouth.
The flue dust or smoket which accompanies the flame
when the furnace is carried on at a regular process, and
which deposits on the tymp and furnace lid with a yellowish
white colour, consists chiefly of earthy and metallic oxides;
it contains free silica (if volatilised sulphide, fluoride, or
chloride of silicon, or silicide of nitrogen comes into contact
with moist air) and alkaline salts (contained in the ash of
the fuel, in limestone, and in iron ore). Sublimed sulphur
is also sometimes contained in this smoke.
* Bgwkfd., v., 283.
EBERMEYER, über Gichtenrauch, in B. u. h. Ztg., 1858, p. 394.
570
IRON.
The smoke exerts an essential influence on the colouration
of the furnace flame, and sometimes contains substances.
which cannot be ascertained in the ore.
The following are some analyses of deposited smoke :---
I.
II.
III.
IV.
V.
Sio
Al₂O¸ .
21°53
2.90
5'36
26.6
57°27
2:46
17'00
1'70
I'74
Fe₂03
IOI.
7'50
II.60
48.7
12°54
and earths.
Mn₂O,
3
8°23
3°40
trace
SnO₂.
0'02
РЬО
1°33
O'12
0*85
10'5
I'27
ZnO .
22.93
13.16
11.86
14.2
I'00
CuO
0'36
SbO.
I'12
3
AgO
trace
CaO
6.14
0'31
I'96
17'73
MgO
0*40
0'34
0°30
0°79
NaO.
2.20
KO
10.90
}30
30*10 27.60
KC1
8.22
Loss by heating 2.50
24°32
Nos. 1 to 3 are furnace smoke from Heinrichshütte near
Lobenstein, analysed by Ebermayer. No. 4, the same, from
Altenau on the Hartz, by Bodemann. No. 5, the same,
from Steinrennerhütte (Hartz), collected in the gas conducting
tubes, by Bodemann.
At Durham, a deposit in the tubes for conducting the gas
from a coke blast furnace contained silica, alumina, lime,
oxide of zinc and protoxide of iron, sulphuric acid, small
amounts of oxides of manganese, potassium, iron, magnesium,
and chloride of sodium; it fused at above 3000° F., emitting.
vapours of zinc and forming a yellow slag.
The intensity, colour, and temperature of the furnace
flame, together with the smoke, indicate the regularity of the
process.
At a regular process the escaping flame is only slightly
heated, and passes from the furnace with a certain regularity
and vivacity, and of whitish yellow, reddish blue, or violet
colour, containing a bluish smoke, which settles as a dirty
white deposit; a still higher temperature of the furnace
hearth renders the flame hotter and lighter, containing more
PRODUCTS OF IRON BLAST FURNACES.
57I
smoke. An irregular process also renders the flame hotter
at first but afterwards less brisk, and the quantity of gases
is then insufficient for heating the blast (page 529). The
strong smoke which the flame gives off at the commencement
of an irregular process decreases, and the flame shows a
yellow or reddish yellow colour, derived from burned iron.
The flame from the tymp behaves similarly. However, the
indications of the flame differ in the different smelting
works.
THE PRODUCTS OF IRON BLAST FURNACEs.
These products are partly used and partly thrown away.
1. Pig-iron of the most different qualities (page 267) and
variable composition (page 267); it is produced in different
forms, according to whether it is intended for foundry or
forge purposes, or both.
The expenses attending the production of pig-iron are
influenced chiefly by the prices of the fuel and ore and of
labour; the cost for labour increases with the poorness and
the refractory character of the ores, current expenses, &c.
The size of the production depends chiefly-
a. On the richness, the state of aggregation, and the
fusibility of the ores. Therefore, in Upper Silesia,* the
production from pulverulent ores containing 29 per cent
of iron, in furnaces of equal dimensions, is much less than in
the iron works of Westphalia, Styria, Belgium, England,
and Scotland.
b. On the size of the furnaces, chiefly on the cross section
of the hearth and the boshes (page 471).
c. On the temperature in the smelting zone, which is
influenced chiefly by the blast (page 480), the dimensions of
the furnace, and the nature of the fuel (page 411). Smaller
furnaces with 300 or 400 cubic feet of blast produce from
160 to 500 cwts. of pig-iron weekly; furnaces with from 700
to 1600 cubic feet, from 600 to 1900 cwts.; and larger
furnaces with above 2000 cubic feet produce above 3000
cwts. of pig-iron weekly (page 415).
*B. u. h. Ztg., 1862, p. 164. Berggeist, 1861, No. 39.
572
IRON.
d. On the nature of the limestone serving as flux, chiefly
whether it is pure, or in a raw or burned state (page 406).
e. On the season (page 496).
2. Slags.-Slags from coke furnaces are usually thrown
aside; sometimes they are converted into a basaltic form
and used for paving roads, as in Upper Silesia,* &c.; they
are also mixed with small coal (Königin Marienhütte near
Zwickau), or with sand (Königshütte in Upper Silesia), and
transformed into a basaltic aggregation and used as building
stones. The more viscid acid slags from charcoal furnaces
are frequently used as building stones when cast into
moulds, and sometimes for the preparation of sulphur
vapour baths; but usually they are crushed and washed to
recover the mechanically enclosed iron. The slag which
runs off by itself, and the slag tapped off from Blauöfen con-
tains less iron than the slag which is mechanically removed
from the hearth.
The slag is broken up either by pounding stamps with
inclined soles (Hartz), more effectively by tilt hammers §
of from 3 to 5 cwts. (at Vordernberg is a hammer 380 lbs.
in weight, making 180 blows per minute), or by rollers which
have a still greater power, and are more cheaply worked than
the other machines, but yield less iron from the slags and
wear out in a shorter time. It is advisable to conduct the
muddy water through a grate into several gutters lying one
below another; the deposit is then frequently stirred up, thus
liberating the iron grains from the slags. At Vordernberg
3 or 4 per cent of wash iron is obtained, as the slag is never
tapped off separately, but always removed mechanically pre-
vious to tapping-off the iron. The wash iron resulting at
Altenau in the Hartz amounts to about 3 per cent of the
whole production of iron.
The wash iron is either converted into wrought-iron or
* B. u. h. Ztg., 1861, p. 357.
Preuss. Ztschr., xi., 192. B. u. h. Ztg., 1863, p. 388.
Allgem., B. u. h. Ztg., 1860, p. 31.
|| B. u. h. Ztg., 1862, pp. 159, 366.
§ WENIGER, pract. Schmelzmeister, p. 126.
PRODUCTS OF IRON BLAST FURNACES.
573
charged again in the iron blast furnace by adding from
5 to 20 lbs to each ore charge; it is also used in the lead
works at the precipitation process. The sand resulting at
the slag pounding is used for building purposes, and some-
times as foundry sand* (Mariazell).
3. Deposits in the Furnace and Aggregations in the
Lower Part of the Furnace.-They are either useless (al-
though frequently of geological and mineralogical interest), or
they are used for different purposes-for example, zinc deposits,
used for the production of zinc. The zinciferous deposits
sometimes occur in regular six-sided crystals, similar to
crystallized pyromorphite, consisting of oxide of zinc; but
usually these deposits consist of an admixture of oxide of
zinc with the oxides of iron and earthy components, form-
ing a compact, scaly, yellowish or greyish green or black
mass, sometimes with a laminar radiated structure.
Haldat found amorphous globules or laminæ coated with
honey-yellow rhomboidal crystals. The crystals of oxide of
zinc seem most apt to deposit on the boshes and in the
lower parts of the furnace, but they are sometimes found in
the upper parts. The ore as well as the fuel and fluxes may
contain zinc blende, thus contributing to the formation of
zinciferous deposits; these deposits sometimes contain small
Some of the
quantities of chlorides of lead and copper.
deposits are found to be composed as follows:-
I.
II.
III.
IV.
V.
VI.
ZnO
78.78
749 78 78 80'10 700 9300 97'77
Fe,0,
3
13.9
12.68
РЬО
0.8
2.0
90'0
2'10
I'21
CaO
I'7
A1203
I'25
SiO3
2.5 4'51
0'45
0*64
C
4'20
2'45
S
trace
ZnS 2.00
Earths
7.82 19.0
•
Particles of ore
* B. u. h. Ztg., 1863, p. 304.
KOCH, kryst. Hüttenpr., p. 24.
784. ERDM., J. f. pr. Chem., vii.,
HAUSMANN in KARST, Arch., 2. R., xvii.,
341. Bgwkfd., xi., 615. HAUSMANN,
Beiträge zur met., Krystallkde., p. 14. LEON., Hüttenerz., p. 377.
Gurlt,
Pyrog. Min., p. 42. Oesterr. Ztschr., 1857, p. IOI.
574
IRON.
No. 1. Compact deposits, analysed by Anthon.
No. 2.
The same, from Hasselö, by Johnson. No. 3. The same,
from Lauchhammer, by Lampadius.
No. 4. The same,
from Séraing, by Coste. No. 4. The same, from a furnace
in Shropshire, by Calvert. No. 6. The same, from a blast
furnace in Benton County in North America, by Mallet.
Ferriferous Bears are used for the production of wrought-
iron,* or are re-melted in blast furnaces when producing
forge pig-iron.
Cyanide of Potassium was first observed by Clarkt in a
blast furnace at Clyde, near Aberdeen (Scotland), and has
since been frequently found. It is formed at a high tempera-
ture by the contact of carbon and nitrogen (air) with strong
bases (alkalies, lime, baryta, &c.), conditions which the
hearths of iron blast furnaces supply. The potassium salt
is volatilised and evolved from the furnace mouth, or from
below the tymp, and deposits on the front wall of the fur-
nace, forming a white or grey powder mixed with more or
less coal and ore dust. Owing to a partial decomposition
in moist air, it usually forms a mixture of KCy, KOCyO,
and KOCO₂; when carbonate of potash is present it usually
contains a corresponding ammonia salt and an admixture of
carbide of iron.|| Bromeis states that crystals of cyanide
of potassium from the the blast furnace at Mägdesprung
contained:
Fe
K
Cy
но
12:4
37°4
37°4
12.8
Zinken and Bromeis§ found this salt in the hearth of the
above-named blast furnace in admixture with carbon and
lead.
* B. u. h. Ztg., 1862, p. 264.
† POGG., Ann., xl., 375.
+
+
BERZELIUS, Jahrsber, 1846, p. 213. ERDM., J. f. pr. Chem., xli., 161; xlix.,
191, 315; liv., 133.
61.
DINGL., xcv., 93.
|| Bgwkfd., v., 285.
§ Bgwkfd., iv., 289.
LIEBIG, Jahrsber, 1850, p. 350.
Berggeist, 1861, No. 18.
LEONH., Hüttenerz., p. 385.
POLYT. Centr., 1852, p.
PRODUCTS OF IRON BLAST FURNACES.
575
Bunsen and Playfair* found this salt in English blast
furnaces; Eck,t in the blast furnaces at Königshütte (Upper
Silesia).
From the blast furnace at Mariazell in Styria is obtained
commercial cyanide of potassium.
Bunsen and Playfair state that the cyanide of potassium is
contained in the space between the tuyere and belly, Müller||
found it above the belly, and Löwe and Redtenbacher found
it in the tubes conducting the gas from the furnace mouth;
Peters§ discovered it in the deposits on the furnace mouth
and the tymp.
Cyano-Nitride of Titanium is usually found in the
ferriferous bears, or in cracks of the hearth lining, forming
either a compact red mass, or copper red cubes, rarely oc-
curring in the shape of octahedrons.
Grignon found this product in the blast furnaces at Bayard,
and called attention to it in 1757; in 1822 Wollaston found it
in English blast furnaces, and it was afterwards observed
in many German blast furnaces.**
Laugier and Wollaston considered that the cubes consist
of metallic titanium, and this opinion was generally accepted
until Wöhler proved it to be a compound of TiCy+3Ti,N,
of the following composition :-
Experiment. Theory.
Ti.
N.
C
Graphite.
77'26
78'00
18.30
IS.II
3.64
3.89
0*92
According to Wöhler, the formation of these crystals stands
in a certain relation to the formation of cyanide of potas-
sium, and microscopical crystals were obtained by heating
*
ERDM., J. f. pr. Ch., xi., 124; xxv., 246; xxxiii., 197; xlii., 392.
KARST., Arch., xxiv., 286.
Ann. d. Chem. u. Pharm., xlviii., 150.
|| Bgwkfd., v., 285.
§ B. u. h. Ztg., 1858, p. 243.
¶ Oesterr. Ztschr., 1858, p. 367.
**
I
KARST., Arch., iv., 351; 1 R., ix.,
LEON. u. BRONN., Jahrb., 1837,
GURLT, pyrog. Min., p. 57. LIEBIG
SCHWEIGG., Journ., xli., 80; xliv., p. 47.
524. POGG., Ann., iii., 175; lxxxiii., 596.
p. 582. ERDM., J. f. pr. Ch., xvi., 212.
Jahresber., i., 401; iv., 343, 751.
576
IRON.
together of rutile and ferro-cyanide potassium to the fusion
point of iron. According to later experiments by Wöhler
and Deville,* nitrogen and titanium combine direct at a
high temperature and in the nascent state. Scheerert
suggests that the difficult fusibility of the titaniferous iron
ore is caused by the titanium absorbing cyanogen, which is
otherwise very effective for the carbonisation of the iron in
the lower part of the furnace shaft, forming the above com-
bination.
As Zinkent has demonstrated the volatility of these cubes
they must be considered as a product of sublimation, and
their occurrence in the lower part of the hearth may be ex-
plained by the pressure which is exerted in the hearth,
forcing the vapours to descend. From the frequent occur-
rence of the crystallised cyano-nitride of titanium along
with silica we may suppose that silica and titanium combine.
Wöhler|| produced a compound of aluminium, titanium, and
silicon; the frequent separation of the titanium compound
indicates that titanium has no great affinity for iron.
4. Waste Gases. Their composition, collection, and
use have already been described (pages 416 and 528).
THE EXAMINATION OF A BLAST FURNACE
ESTABLISHMENT.
When inspecting blast furnaces the following points have
to be taken into consideration :-The situation, extent,
and motive power of the establishment, and the nature
of the ores (kind of ore, injurious or advantageous asso-
ciates, analyses, behaviour in smelting, percentage of iron,
cost); the assaying of the ores (the manner of selecting
the samples, fluxes, furnaces, length of smelting, consump-
tion of fuel); the weathering of the ores (time, effect,
artificial watering); the roasting of the ores (roasted or
not roasted, in kilns or otherwise, kind of fuel and consump-
tion of fuel, quantity roasted in a certain time, number of
* LIEBIG'S Ann. d. Chem. u. Pharm., 1857, Bd. 103, p. 230.
+ SCHEERER, Metallurgie, ii., 186.
+
POGG., Ann., Bd. 28, p. 160.
|| ERDM., J. f. pr. Ch., Bd. 80, p. 255.
THE EXAMINATION OF BLAST FURNACES.
577
workmen, wages, cost of roasting); the breaking-up of
the ores (the method employed and its effect, size of the
broken ores, quantity, number of workmen, wages, cost of
breaking); the fluxes (quality, composition, application in
raw or burned state, mode of dividing the fluxes, size of the
broken pieces, weight per cubic foot, percentage of the quan-
tity added, charging separately or in admixture with the
ore, cost); the fuel (kinds of fuel, quality, amount of ash
and moisture, injurious components of the ash, size of the
pieces, weight of one cubic foot, volume of the delivered
fuel, &c., vessels for measuring, if used, method of carbon-
ising or coking, price); the ore mixture, including the
fluxes (rules adopted, varieties of iron ores, fluxes, and the
manner in which they are mixed with the ore, average per-
centage of iron, lifts, size of the charging vessels, the size
of the ore heaps when mixed with the fluxes [the ores for
charcoal furnaces are mixed with the fluxes by placing ores
and fluxes in alternate layers one above the other, the ore
thus mixed forms heaps of the shape of a truncated pyramid
and is termed in German" Möller"], the weight of one cubic
foot dry and wet, working tools, number of workmen,
working cost); the iron blast furnaces (construction of the
furnace, material used for the different parts of the furnace,
dimensions of the furnace, capacity of the hearth, construc-
tion of the apparatus for collecting the waste gases, age of
furnace since the blowing-in); the tuyeres (number and
dimensions of the tuyeres, material of which they are made,
open or closed, dry or cooled with water, the position of the
tuyeres with regard to their deviation from the horizontal
line, height of the tuyeres above the bottom stone, bright or
dark tuyeres, with or without nose); the blast (blast engine
and its motive power, regulators, blast heating stoves and
the mode of firing, blast mains, construction position and
diameter of the blowpipe, temperature, pressure and volume of
blast per minute, the volume calculated from the blast engine
and from the blowpipe, loss of blast, apparatus for measuring
the temperature and pressure of the blast); the charging
(size of one charge of fuel and ore with regard to volume.
and to weight, charging vessels and other apparatus required,
VOL. II.
2 r
578
IRON.
mode of charging, number of charges descending the furnace
in 24 hours, with consideration of the tappings when the
blast is shut off, the depth to which the charges descend in
a given time, time in which the charges enter before the
tuyeres, and how many charges the furnace contains above
the tuyeres, the descent of charges required for the regular pro-
cess, mode of keeping account of the charges); the manipu-
lations in and before the hearth (removal of the slag and
the pig-iron, cleaning the hearth and the tuyeres, forma-
tion of the slag drift and the gutters or feeders and moulds
for receiving the liquid iron, length of the operation, mode
of blowing-out, number of workmen, wages, tools, drawings
of blown-out furnaces); the products, such as pig-iron
(varieties, quality, application, the production in 24 hours or
per week, the yield in comparison to the assay, consumption
of fuel, ore and fluxes per cwt. of pig-iron, cost per cwt.);
the slags (nature of the normal and abnormal slags, trans-
port of the slags from the smelting works, further application
of the slags for recovering wash iron, building purposes, &c.);
lead (tapping off and further application); waste gases
(quantity, temperature, colour, rapidity of the ascent, fuming,
change at the different processes, methods of collecting and
applying); the deposits and other products (cyanide of
potassium, furnace cadmia or calamine) (Ofenbruch, Gicht-
schwamm).
The opposite tables, A and B, give details of some of the
above-mentioned points in different blast furnaces.
FOUNDING AND MOULDING.
Castings are produced by running liquid iron into hollow
moulds made for the purpose. The liquid iron is either
taken direct from the blast furnaces or obtained by re-melting
pig-iron.
Pig-iron is generally well adapted for castings owing to its
cheapness, its difficult fusibility, its strength, and its peculiar
behaviour on solidifying (page 291); and the slightly mottled
varieties (page 286) are most frequently used for castings, as
they excel in compactness, strength, and perfect filling of
the moulds, whilst they are sufficiently soft to allow a
IRON WORKS.
IRON ORES.
Fluxes.
I. Charcoal Furnaces.
I. Furnaces with Closed Front Walls
A. Spiegeleisen, Flowery Iron, &c.
Carinthia, Lölling
Styria, Fridau
Sessler
15
Steirer
""
Eisenerz..
17
Thuringia, Katzhütte
Siegen, Sieghütte
Müsen
B. Grey Pig-Iron.
Styria, Mariazell
St. Stephan
:
:
:
:
2. Furnaces with Open Front Walls
A. White or Specular Iron.
Gittelde (Hartz)..
Mägdesprung (Hartz)
Concordiahütte (Rhine)
Sparry, Brown, and Red
Iron Ore
Sparry and Brown Iron
Ore, Sphærosiderite
Red, Brown, and Sparry
Iron Ore
II. Coke Furnaces.
Belgian Iron Works..
Königshütte (Silesia)
Henrichshütte near Hattingen
Hasslinghausen.
Brown and Red Iron Ore
Pulverulent Brown and
Clay Iron Ore..
Blackband and Sparry Iron
Ore
Blackband
•
50 lime
per cent.
Percentage of Iron
in Ore or Mixture.
One Charge of
CHARGES.
Fuel
Mixture
consists of
consists of
Sparry and Clay Iron Ore
15 slate
40
1)
"
15 lime
35-36
15'75 cubic feet
250-300 lbs.
310-320
80
1'68
"}
87
74
"}
0'95-I,,
Number of Charges
in 24 hours.
Consumption of Fuel
per cwt. of Iron.
Production in
24 hours.
TABLE A.
Height of the
Hearth.
Height of thé
Boshes.
Total Height.
Height of the
Tuyeres.
DIMENSIONS of the FURNACES.
Width of the
Upper Hearth.
Width of the
Lower Hearth.
cwts.
feet.
feet.
feet.
feet.
feet.
feet.
feet.
feet.
degrees.
feet.
Diameter ofthe
Belly.
•
Sparry Iron Ore, Hæmatite
52'8
0'60-0'65
325-340
Sparry Iron Ore..
5-8 siliceous clay
42
31 cubic feet
10-15
17
43-45
31
11
570 lbs. and 20
lbs.wash iron.
650 lbs. and 20
lbs.wash iron.
40'
It'
33
3'
14'
3
2}"
0'9-1'1
160-175
1250
73
11'6 cub. ft.
500
42'
I′ 10″
6'
8'
4'
12'
5
"
11
:
2-3
45
13'5
"}
320 lbs.
100-120
78
100-118,
"}
I'16 cub. it.
300-360
140-145
No
I”—2″ 3″″
2″-2″ 3″
37'
I' 8"
8'
2′ 6″
9′ 3″
26'
I′ 3″
6′ 3″
2' I'
42
2" 8"
2″ 3″”—2″′ 4″
12"
5'5 slate..
11
})
50.8
19'5
}}
383 lbs. and 8
lbs.wash iron.
0'7 cwt.
280
Sparry and Clay Iron Ore
fluor spar
45
300 lbs.
36'
I' 8"
650 lbs. and 30
51
8'
21"
IO
викл
3
lbs.wash iron.
45
обод
4' 9"
3′ 44″
Little lime
35'-10"
I' 91″
2′ 5″
環
​I″′ 4″”—2″ 6***
I" 2"-2"
140-150
160
120
14-1600
1100
I" 6”—1″ 10"""
200
850-900
I' 8"
7' 4"
4' 5"
57
9′11″
N
71
"
13"
(2lbs.)
160-200
200
""
34
0.88
120
in
in
28'
2'-2'4'
S'
2′ 9″
30
1′ 6″—1′ 9″
9'
3′ 6″
45
60
3
I}"
I" 6"
150
I
I" S"
225
35-40
20-40 (30 average)
38
40-43
3' 10"
36-37 (mixture)
210 lbs.
6'47-6'74 cb.ft.
32-34
27-30 (mixture)
133-136 cwts.
1*18
55-58
4'
3'
28
I' 3"
2′ 9″
I' 8"
7
62
4'7"
IO"
30′
I'17
2°17′
I'
S
50
0'94
"
103
35'
9' 5"
Diameter of the
Furnace Mouth
Angle of the
Boshes.
Height of the
Belly.
Number of Tuyeres.
B. Mottled or Grey Iron.
Rothehütte (Hartz)
Königshütte
Altenau
Red and Brown Iron Ore,
Finery Slags
2 lime
""
}}
35-36 (mixture)
30-33 (mixture)
250 lbs.
250
>>
660 lbs.
650-700 lbs.
52
28-34
I'092 cwts.
1*167
114-121
66
tin
3'6"
34'
I' 3"
3'
2'
5'
2′ 9″
35'
I′ 3″
2′ 6″
I' S"
م
7' 8"
S'
to to
6'
56
~
45
2
I'′ 6″”—I'' 10"**
I" II""”—2″ I'''
175-225
238-255
850
950—1030
"}
Red, Brown, and Magnetic
Iron Ore
Lerbach
Red and Brown Iron Ore
II.2
1'4
11
31°5 (mixture)
230
""
21
"1
30'5 (mixture)
230
})
8-8 cub. ft.
7-78
28-30
I'202
}}
28-30
I'177
56-57
55
28′ 3″
I' 3'
I' 6"
7'3'
"}
31'
I′ 3″
I' 9"
7′ 3″
int
6' 6"
7'
I
24"
I”—I"7"
I"
240
150
520
600
Ilsenberg
19
Swedish Iron Works
:
Red, Brown, and Magnetic
Iron Ore, Finery Slags..
Magnetic and Specular Iron
Ore
..
Norwegian
Russian
37
11
"}
"1
}}
34-37 (mixture)
37-58 (mixture)
25-42
25 cubic feet
7'I
42-44
I'2
85
2′ 10″
4'
"
38'
12-18 cub. ft.
1'3-2'08 cwts.
60-160
33-74
0-4-7′
0-6'
28-42′
I′ 4″
'8"-2'
I' 8"
3'5'-4'5'
I' 4"
2-2.8'
3′ 9″
6′ 9″
21"
3-6
30-31′
7
4-5'
11-14'
12'
I-3
21-21"
I!"
0'6"-1"
0*87—1″ 15″”
250-300
800
15-250
200-300
Magnetic and Brown Iron
Ore
20-36 Pud.
30-59 Pud.
20-60
1'0-1'9",
165-400
7-8'
33-56'
3'5-4
2'3'-2'5'
II-14'
7-S'
2
2*6-4'37"
I'5-3'5"
Diameter of the
Tuyeres,
Pressure of the
Blast.
[To face page 578.
Temperature of the
Blast.
Quantity of Blast
inches.
inches water.
C.
cubic feet.
36"
'6"
ΙΟΥ
40"
2'
3'
45
IOʻ
3′ 6″
9'6"
21″
2' 8"
ΙΟ
285
I" II""'
I"
200
1
:
I
21"
31
20
N
1*625"
108
12"
15
200-250
455-555
4' 5"
2
2"
I'5"
140
600
582
1'5-2'4",
215-320
2-4
1*5—2'5″
5'5-8.5"
75-100
8.39 cub. ft.
2880-3640
150
7'
12'
48.
2'
3'
15'
7
1*396 cwts.
485
19'
3
7'9"
IO'
2*5-3″
1600 lbs.
3200 lbs.
1'67
54'
2′ 5″
17
450
51
2' 10"
45'
15' 3"
9′ 6″
3- 43"
7'
16'
9'
(n10
(24-24 lbs.)
(1.8 lbs.)
50-60
૩૦૦
3600
3710
(3-4 lbs.)
per Minute.
IRON WORKS.
To face page 579.]
Capacity of the
Furnaces.
cubic
tons.
metres.
Production in
24 hours.
Number of Tuyeres.
Diameter of the
Blowpipes.
!
Pressure of the
Blast.
Temperature of the
Blast.
Blast per ton of
Pig-Iron
calculated from
the Fuel consumed.
centimetres.
metres.
degrees.
cubic metres.
cubic
metres.
cubic
metres.
TABLE B.
Blast per Minute.
Blast per Minute
and per cubic metre
of the Furnace Shaft.
Roasted Ore.
Limestone.
The usual Furnaces in Scotland
Old Furnaces in Scotland (1833)
150-160
21
4-6
7'5-8'5
O'12-0 20
400
85-90
9
2-3
7'5
0'12-0'13
Blast Furnaces in Scotland
200
26
8-9
7'5
0'15
322
370
Blast Furnace in Bilston, No. 3 (Staffordshire)
175
24
5-7
(?)
New Furnaces in Staffordshire ..
Old Furnaces in Staffordshire (1840-1850)
New Anthracite Furnaces in Yniscedwin and
Ystalifera..
Blast Furnaces in Blaenavon (Wales)
120-130
70-75
16-20
5-7
ΤΟ
3
7'5
5-7
0'13—0'16 300-330
0'13-0*16
300-330
300 or 12
5*084
4*S07-5'798
5'575
6.690
6'244-6·690
6·690-8'028
74
30-36
0'50
1*750
0*400
2'10
0'50
0°40
2'120
0*330
2-24
0'95
100
0'50
I'940
0'400
2'30
0'50
I'14
1*08-13
125.
Darkish Grey.
1-1*225
I
III
70-93
! 0·65-0'73
46-55
0'63
0·58-0·71 2'00-2.25
2*150
2.150
0'45—0'50
0*500
0'45-0'60 2'75-3
3-3'50
3
0'35
0'75-I
0'7-I
1'50
1'4-1'5
1'5-18
o'$$3
}"
Grey Forge Iron.
o'927-0.95 Grey Foundry and Forge Iron.
0'85-0'75
་་
흐흐
​(a)
(6)
O'10-'12
80-85
10-12
6-7
3'2-4
IIO
Blast Furnaces in Tees-Side in Middlesbro'..
195-200
14-15
30
3
(?)
0'17-0'20 350-400
0'15
12
6*690
7·805
46-56
0'57-0.66
78
3
در
ΙΟ
300-330
5'620
117
0'71
0'58-0·60
2:380
2'700
2'500
o'Soo
1'75
I
I'5
1.82
(c)
(d)
0°970
0'65-0'70
3'05
0'7
I'75
I'2
2.60
0'75
1.26
1'23
Grey Forge Iron.
(e)
0'13-0'14
The usual Furnaces in Middlesbro', Erected
in 1851-1855
Large Blast Furnace in Dowlais
Blast Furnaces in Ebbw Vale
Usual Furnaces in Dowlais
Blast Furnaces in Victoria
Blast Furnaces in Tredegare
Blast Furnaces in Sir-Howy
135-140
200-235
160-170
140-150
115-120
120-125 20-22
105
20-25
45-50
25-30
25
24
3-5
ΤΟ
7
8.7
0'12—0'14 300-330
0'15—0'16
6°244
85-100
0·63-0'71
2'400
0'700
در
3
0*75
I'4
1'03
Grey Foundry and Forge Iron.
ان
6
5-7
5-6
7'5-8.1
7'5
7'5-8'1
0'13-0*16
0'13-0*16
0'13-0'16
315-330
300-330
300-330
300-330
5'352
5.789
ISO
100-120
077
0'63-0'70
2'400
0.850
2
0'35
I'2
1.62
White, Laminated or Grained.
(g)
2.630
1'000
2.40
0*25
I'3
I'SI
White, Lamellar.
(})
4'964
$6
0'57-0.61
2'400
0.850
I'90
0'35
I'113
1'7
White, Grained.
(i)
5'129
85
073
2.630
0'500
2'20
O'25
I'15
1'42
"
()
as in Victoria
5*709
SI
0'66
2'750
0'62 lime
2:35
0'25
1.28
1'43
21
*>
(K)
20-21
3 of 8.1)
I of 6.2 f
0'17
330
4'683
67
0*64
2'650
0'500
2
0.25
105
1'57
White.
(?)
tons.
tons.
tons.
(a) The higher statements of coal and limestone concern the foundry iron.
(b) The lower figures indicate cold blast.
(c) The blast furnaces in Ystalifera have up to ten tuyeres, but they are usually
not used at the same time.
(d) Coke and cold blast is used. The consumption of coke is 195 tons.
(e) The consumption of coke for forge iron is 140 tons, and 170 tons for
foundry iron.
The consumption of coke for foundry and forge iron is 160 tons.
(g) Raw coal is used almost exclusively.
(1) One-third of coal and two-thirds of coke.
(i) Raw coal. Some furnaces consume only from 150 to 175 tons.
() A great deal of puddling, &c., slags and siliceous ores are used.
(k) A mixture of coal and coke is employed.
(1) Half of coal and half of coke.
น
CONSUMPTION PER TON OF IRON.
For the Blast
Furnace.
Coal.
For Heating
the Blast, for
Engie, and
the Blast
for Roasting.
tons.
Burnt Carbon
tons.
in the
Blast Furnace
without Ash.
Ore and Limestone
per Ton of Coal.

Quality of the Pig-Iron.
Remarks.
FOUNDING AND MOULDING.
579
mechanical treatment of the castings. For some purposes,
hard rollers, &c., the grained white pig-iron poorer in carbon
is well fitted.
Whenever the quality of the pig-iron, &c., will allow it,
the castings are made direct from the blast furnace, as they
are then much cheaper. This is usually done in the case of
charcoal furnaces where special precautions are employed for
the formation of a suitable mottled iron. These are an addi-
tion of raw ore (page 553), or of iron finery cinders (page 554)
charging the furnace through the tuyeres (in the case of
Blauöfen), or the fore-hearth, with roasted ores or iron finery
cinders (page 546), charging heavy ore burdens (page
538), &c.
The following circumstances may, however, make it ad-
visable to re-melt pig-iron :-
1. When re-melted it may be done at any time, and an
appropriate selection may be made of pig-iron suitable to the
different castings; for instance, a thinly liquid iron rich in
phosphorus is best for thin and fine castings; a very com-
pact slightly mottled iron for steam and blowing cylinders;
an easily chilling iron for rollers, &c.
2. Some sorts of pig-iron are improved by a re-melting,
which modifies their chemical composition; for instance, the
dark sorts of pig-iron rich in graphite and silicon (Scotch
pig-iron) are improved by being re-melted in reverberatory
furnaces. In other cases the iron increases in strength by
a different arrangement of its molecules without chemical
change. This may be observed in metal of more finely grained
texture (re-melting in crucibles and cupola furnaces). Fair-
bairn's investigations* have shown that pig-iron increases
in strength up to the twelfth re-melting, and that a further
re-melting decreases its strength.
3. Large and heavy castings require more pig-iron than can
be contained by the hearth of one or even two blast furnaces.
4. The founding is not impeded by the blowing out of the
blast furnaces.
5. Iron is re-melted where there is a favourable market
Report of the British Association for the year 1853.
2 P 2
580
IRON.
for castings, and where blast furnaces cannot be profitably
carried on, owing to a want of ores, &c.
I. RE-MELTING PIG-IRON.
The re-melting of pig-iron is effected either in crucibles or
in cupola or reverberatory furnaces, according to the quality
of the iron, the size and the quality of the casting, and the
It is always
manner in which the foundry is carried on.
necessary to melt as quickly as possible and with the smallest
consumption of fuel.
As we have before mentioned, a slightly mottled, finely
grained pig-iron is suitable for most castings. The different
kinds of grey iron have a tendency, when re-melted at a suit-
able temperature, to become lighter, owing to a more inti-
mate combination of the carbon, and therefore the slightly
mottled iron is produced by re-melting grey pig-iron of the
regular process, not too rich in graphite, in one of these appa-
ratus, chiefly in crucibles or cupola furnaces. The varieties
of dark grey iron rich in graphite (Scotch iron, for instance)
when intended for the production of a sufficiently strong iron,
require to be re-melted in reverberatory furnaces, in which
a more or less strong chemical action of the atmospheric air
may take place. A re-melting in cupola furnaces or crucibles
causes less chemical modification of the iron, and the varieties
of graphitic iron when re-melted in these apparatus require
to be mixed with less grey pig-iron, or with scraps of
wrought-iron. Mottled, chiefly strongly mottled, iron easily
becomes white when re-melted, particularly in reverberatory
furnaces, when it is no longer fit for most foundry purposes.
However, the same kind of pig-iron may vary in its be-
haviour, according to the different construction of the fur-
naces and to the manner in which the smelting operations
are carried on. The liquid iron must be hotter, and conse-
quently more thinly liquid, the smaller and thinner the cast-
ings are intended to be.
The subject of founding and moulding is most elaborately
treated in the following books:-
KARSTEN'S Eisenhüttenkunde, 3 Aufl., Bd. iii.
RE-MELTING PIG-IRON IN CRUCIBLES.
581
VALERIUS, Handbuch der Roheisenfabrikation, Deutch v.
Hartmann, 1851.
KARMARSCH, mechanische Technologie, 3 Aufl., Bd. 1, p. 74.
WIEBE, die Maschinenbaumaterialien. Stuttgart, 1858.
GUETTIER, de la fondrie, telle qu'elle existe aujourd'hui
en France, &c.: Paris, 1858.
GUETTIER, de l'emploi pratique et raisonné' de la fonte et
de fer, &c.: Paris, 1861.
HARTMANN, Handbuch der Eisengiesserei nebst Atlas:
Weimar, 1863.
A. Re-Melting Pig-Iron in Crucibles.
This mode of re-melting is very simple, and requires but
little apparatus; but, on the other hand, it is very expensive
as regards consumption of fuel, labour, crucibles, and loss of
iron, and is only adapted for the production of small castings
which may be very expensive on account of the difficulty of
their moulding; also the production can only be small, per-
haps 2 cwts. at a time, which may be cast several times
daily. As the pig-iron is fused at a moderate temperature,
and protected from the influence of the atmosphere, it is but
very slightly modified. Grey iron of the regular process
which has been several times re-melted, and which is too
rich either in carbon or silicon, is best adapted for the treat-
ment in crucibles; different sorts of iron, either more or less.
grey, are sometimes employed in admixture.
The melting is effected in clay or in plumbago crucibles
containing charges of at most 30 lbs. of pig-iron in order to
facilitate their manipulation. The furnaces used are either
common air furnaces with a chimney 30 or 40 feet high, or
Sefström's furnaces, provided with apertures on the bottom
for the introduction of blast, and which are more advan-
tageously fed with coke than with charcoal. The consump-
tion of fuel is smallest when using larger crucibles, coke
and blast. Each furnace contains one or two crucibles,
and is gradually heated at the commencement, and at
last so strongly that the melted iron (after three or four
hours) is sufficiently hot to remain liquid for some time after
582
IRON.
the crucibles are removed from the furnace. The crucibles
are provided with a lid or with a cover of coal dust or slag;
the loss of iron by scorification may thus be reduced to from
2 to 4 per cent, but still the total loss caused by spilling and
the excess in the ingots may rise to 25 per cent and more.
The re-melting of 100 lbs. of pig-iron requires from 50 to 100
cubic feet of charcoal, or from 10 to 75 cubic feet of coke.
once.
}
B. Re-Melting Pig-Iron in Cupola Furnaces.
This mode of re-melting is that most in use; it allows the
most perfect use of the fuel, at any time provides liquid
iron of the quality required for foundry purposes, and
excels in the facility and quickness of working; these advan-
tages are not possessed by the process in reverberatory
furnaces, as these furnaces melt larger quantities of iron at
a time, and the charges must be cast into moulds all at
Also the process in reverberatory furnaces is more
expensive, chiefly on account of the larger consumption of
fuel; reverberatory furnaces also yield a less homogeneous
iron than cupola furnaces, in which the iron is melted under
a greater pressure; therefore, in many foundries (Berlin,*
Gleiwitz, &c.,) reverberatory furnaces are replaced by
enlarged cupola furnaces, and are used only as an assistance,
when making very large castings which do not require the
iron to be very homogeneous, or when re-melting large and
faulty castings unfit for cupola furnaces. Hoppet has con-
structed a reverberatory furnace which is said to unite the
advantages of the cupola and the reverberatory furnaces;
admits of continual charging and tapping, with a flame-fire
without blast.
it
Owing to the rapidity with which the iron passes through
the cupola furnace, the varieties of iron richer in carbon do
not receive any new carbon in combination; on the contrary,
the iron is always somewhat decarbonised by the blast-by
this means, and the more intimate combination of the carbon
with it, the iron has a tendency to become white.
* B. u. h. Ztg., 1862, p. 23.
↑
HARTM., Fortschr., i., 238.
-1-
B. u. h. Ztg., 1862, p. 33.
RE-MELTING PIG-IRON IN CUPOLA FURnaces. 583
The smelting materials are :—
1. Pig-Iron. That best adapted is grey iron of the regular
process, produced from a moderately fusible ore mixture; the
resulting iron is thus slightly mottled. Easily fusible grey
iron containing phosphorus is rendered white upon re-melting
with coke, and can only be re-melted without being modified
in high cupola furnaces with charcoal. The dark grey
graphitic pig-iron, containing a larger amount of silicon
(page 287), yields, when re-melted, a less strong grey iron,
and must be improved by an admixture of lighter iron. For
example, the dark grey Scotch iron is melted in the
Berlin foundries, together with Silesian coke pig-iron, and
with Silesian and Swedish charcoal iron. Old iron is usually
mixed with the new iron, thus producing different sorts of
metal according to the proportion of each kind present. It was
stated on page 295 that the grey foundry iron comprises three
numbers, No. I designating the darkest kind; the letters
CB and HB indicate whether the iron is produced with
cold blast or hot blast. The iron is broken into pieces of
from 8 to 16 cubic inches in size-with large masses this is
effected by means of a heavy iron ball, which falls upon them
from a considerable height.
2. Scraps of Cast and Wrought Iron, mostly in the
form of chips and borings. The scraps of wrought-iron absorb
some carbon, lessen the loss of iron, and produce a very
compact tough cast-iron; this mode of increasing the
strength of cast-iron was formerly patented by Stirling.†
He states that grey Scotch iron, No. 1, may be mixed with
from 24 to 40 per cent. of scraps, No. 2 with from 20 to 30,
and No. 3 with from 15 to 20 per cent, and that hot blast iron
admits of a larger addition than cold blast iron. Too much
addition of wrought-iron renders the cast-iron hard; this
circumstance is made use of when casting rollers (Berlin),†
and Pattinson's pans. Burnt iron also takes graphite into
combination, rendering the cast-iron more compact and
harder. To prevent the iron chips from blocking up the
* E. F. DURRE, die Eisengiessereien in Berlin in B. u. h. Ztg., 1862, p. 3.
+DINGL., Bd. 117, p. 307; Bd. 122, p. 212. HARTM., Fortschr., i., 236.
B. u. h. Ztg., 1862, p. 22.
584
IRON.
furnace, Stenson* says that the chips should be rammed
into a cast-iron vessel and all melted together, or else
charged in the centre of the furnace throat, surrounding.
them with limestone, which then is surrounded with the
fuel; also the flux should be charged upon the chips.t At
Mariazell, Ruttner‡ rams the chips into moulds, moistens
them with salt water and exposes the lumps for about eight
days to the atmospheric air, when the chips will oxidise and
harden. Keck employs hot water containing a little
muriatic acid for oxidation instead of salt water, but, at the
highest, he only gives 30 per cent of wrought-iron chips in
admixture, as otherwise white iron is produced. As the
quick smelting prevents the reduction of the scale formed by
oxidation, an addition of pounded blast furnace slag and
limestone is given for the formation of a slag. Chips which
have been too strongly oxidised are more fit for the blast
furnace than for smelting in cupola furnaces. Mayrhofer§
recommends the placing of wrought-iron chips in the moulds
for receiving the iron from blast furnaces, as they cause the
separation of silicon.
3. Fluxes, namely :-
a. Limestone, oyster shells, or fluor-spar, which are added
to scorify the ash of the fuel (chiefly of the coke), the sand
sticking to the iron pigs, portions of the furnace lining
which may come off, silica which may separate from the
pig-iron before the tuyere by oxidation, &c. According to
Wernecke,¶ an addition of from 40 to 60 lbs. of fluor-spar
per 100 lbs of iron produces a slag more thinly liquid and
poorer in iron than limestone, and also a purer cast-iron.
When using coke as fuel the usual addition of limestone
does not exceed 3 or 4 per cent.
b. Sand, or iron blast furnace slags, chiefly when employ-
ing oxidised iron chips.
* Polyt. Centr., 1857, p. 828.
HARTM., Fortschr., i. 242.
+
++
‡ Ibid., v., 172.
Oesterr. Ztschr., 1861, No. 25.
|| Oesterr. Ztschr., 1861, Nos. 36, 38. HARTM., Fortschr., v., 173.
§ LEOB., Jahrb., x., 328.
¶ B. u. h. Ztg., 1863, p. 96.
SMELTING MATERIALS FOR CUPOLA FURNACES.
585
4. Fuel.-Coke and charcoal are mostly used; when
employing raw fuel, for instance brown coal,* the tempera-
ture of the furnace is too much lowered,† as the formed
products of carbonisation absorb part of the heat; however,
at Ransko, kiln-dried turf has been successfully employed.
Owing to its sulphur, pit-coal is not so suitable as anthracite,
At St.
which has been used in experiments with success.
Stephan, in Styria, attempts were formerly made to collect
the waste gases at some distance below the furnace mouth
through pipes which were introduced for this purpose, and
to conduct them into the tuyere together with air by means
of a double nozzle, as in a oxy-hydrogen blowpipe.
Differing from the blast furnace process (page 412), the re-
melting of a certain quantity of pig-iron requires less coke
than charcoal, owing to the fact that in cupola furnaces a
smelting heat only is to be produced without a reducing
action. Now, as the carbonic acid formed upon combustion
is less easily reduced to carbonic oxide by coke than by char-
coal, coke produces a greater heat and fuses the iron more
quickly than charcoal; consequently, also, the waste gases
are richer in carbonic oxide when employing charcoal
than when using coke.|| Hard charcoal, chiefly good
beech coal, is therefore more effective than soft coal, and
more so the more uniform its quality and size.
The use of hot blast leads to a considerable saving in fuel
(from one-third to one-half), an increased production, a
diminished loss of iron (from 4 to 5 per cent), and a quicker
operation, without deterioration of the product if the tem-
perature of the blast, particularly when employing coke, is not
raised much above 250° C.
According to Karsten, 100 lbs. of pig-iron require for re-
melting with cold blast 2.75 cubic feet of charcoal, or o'66
cubic feet of coke, or 28 and 24 lbs. respectively. When
using hot blast in coke cupola furnaces 5 feet high, and in
*
LEOB., Jahrb., 1857, vi., 186.
KARST., Arch., 1 R., i., 1; ii., 165. B. u. h. Ztg., 1843, PP. 714, 964.
LEOB., Jahrb., 1857, vi., 131.
+
|| Bgwkfd., viii., 467.
586
IRON.
charcoal cupola furnaces 15 feet high, 100 cubic feet of coke
will replace 450 cubic feet of charcoal. The coke consumed
in melting 100 lbs. of pig-iron may amount to from 10 to 50 lbs.,
according to the construction of the furnace, the quality of
the pig-iron, &c., but on an average it amounts to from 20 to
24 lbs., and with charcoal to from 30 to 50 lbs. Von Mayr-
hofer* states the average smelting temperature of cupola
furnaces carried on with cold blast to be about 1600° C., and
he has calculated by formulæ the quantity of fuel (charcoal)
required at different temperatures of the hot blast:—
Temperature of the blast
Consumption of fuel per 100 lbs. of
iron (lbs.)
•
do. blast (cubic feet)
Quantity of iron to be smelted with
100 lbs. of fuel (lbs.). .
14° ´ 100°
200° 300°
24*7 230 21.8 20*2
1668 1578 1472 1367
4046 427·8 458*3 493*6
The employment of charcoal or coke renders necessary
essential modifications in the construction of the furnaces
and in the process.
The furnaces differ-
a. In their height, which is greater in charcoal furnaces
than in coke furnaces, in order better to utilise the lower
temperature produced before the tuyeres. If the coke furnaces
are constructed of the same height, the temperature would
be too much increased, wasting the furnace hearth still more
quickly than is already the case. The furnace may be
higher the more refractory its building material, and the
higher the furnace the greater the effect to be obtained from
the fuel.
b. In their interior shape. Charcoal cupola furnaces in
their interior form approach that of blast furnaces, in order
better to utilise the temperature. They are narrower at the
top and bottom than in the middle, and have usually more
or less formed boshes, thus decreasing the number of charges,
but, on the other hand, increasing the burden. Coke cupola.
furnaces for a higher temperature usually have a cylindrical
hearth, and taper somewhat towards the top; sometimes
* LEOB., Jahrb., 1861, x., 422.
THE CONSTRUCTION OF CUPOLA FURNACES. 587
they are enlarged again above the contraction to facilitate the
smelting and the uniform descent of the charges. In rare
cases only are coke furnaces provided with boshes and with a
contracted hearth in order to raise the temperature; this
facilitates the smelting and gives a large production.
c. In the construction of the furnace front. The charcoal
furnaces usually have an open front wall in connection with
a fore-hearth for ladling the iron. These fore-hearths are
very convenient in many cases, as the smallest quantity of
iron may be ladled out at any time without loss, and allowing,
without disadvantage, the return to the fore-hearth of the
liquid iron which may remain in the ladle. A fore-hearth
also enables the furnace to contain larger quantities of iron,
though it is less hot than iron in furnaces with a closed front.
wall; it also facilitates the removal of obstructions in the
furnace, thus allowing longer operations. Coke furnaces
are usually tapped off only, and if provided with a fore-hearth
for casting smaller pieces, it is so constructed that it may be
closed with easily removable iron plates coated with loam.
Charcoal furnaces with a fore-hearth are sometimes emptied
by means of a tapping hole if no ladling is required.
Concerning the process of re-melting, it is effected in coke
furnaces more quickly with an increased admission of blast
(500 or 600 cubic feet) and a larger production; but the
operations are shorter, as the furnace is more strongly at-
tacked by the higher temperature, and the formation of slag
from the ash of the coke and the added limestone. The
lower parts of the furnace must therefore be more frequently
cleaned through the tapping-hole by means of a scraper.
The pressure of the blast* is generally lower in cupola fur-
naces than in blast furnaces, as the escaping gases offer less.
resistance, and as a too rapid ascent of the blast would cause
combustion in the upper part of the furnace and an oxidation
of the iron. In coke cupola furnaces, the pressure (four to
eight lines mercury) is lower than in charcoal furnaces
(nine lines to two inches mercury), because, in order to pro-
duce the required temperature, a much greater quantity of
i
Bgwkfd., iv., 414.
588
IRON.
charcoal must be burnt in a given time, whilst the same tem-
perature is produced at a lower pressure from a smaller
quantity of coke (page 412). For this reason, and as large
quantities of blast must be introduced in order to obtain a
large production, wider tuyeres (up to 7 and 8 inches in
diameter) are used in coke furnaces than in charcoal fur-
naces, and fans are sufficient, whilst charcoal furnaces may
require stronger blast engines.
In order to save fuel, hot blast apparatus above the furnace
mouth are more frequently used in charcoal furnaces than in
coke furnaces, as the apparatus of coke furnaces, chiefly at
the commencement of the operations, are only slightly effec-
tive, and also as the operations are short, and the tempera-
ture in the lower part of the furnace is easily raised too high.
Generally the hot blast apparatus suffers more and begins to
leak sooner by these short operations than in blast furnaces.
The apparatus and plant required are chiefly the smelting
furnaces, the hot blast apparatus, blast engines, and also the
necessary tools.
A. Smelting Furnaces.-Since abandoning the portable.
furnaces, 2 feet high, and from 6 to 9 inches wide, and the
furnaces which could be tilted over and which were until
lately used in Sweden, cupola furnaces have been used for
re-melting iron, and the waste iron produced in iron works
is only occasionally added to the blast furnace mixture
(page 347).
The construction of the cupola furnaces depends chiefly on
the nature of the fuel employed, and on the intended size of
the production. As we have before mentioned, charcoal fur-
naces are similar in shape to iron blast furnaces, whilst coke
furnaces are more cylindrical, or are in shape a combination
of a cylinder and a truncated cone. However, some of the
coke furnaces are also constructed with boshes when adapted
for a large production. Coke furnaces usually have a closed
front wall, and charcoal furnaces an open front wall and at
fore-hearth (page 587). As the ladling of the iron from the
fore-hearth has the same disadvantages as the ladling from.
iron blast furnaces, the iron is frequently tapped off into ladles
(Königshütte in the Hartz), or in special pans. The tapping
THE CONSTRUCTION OF CUPOLA FURNACES.
589
hole is sometimes placed on the side of the furnace opposite
the fore hearth. The furnace lining is either walled up with
fire-bricks, or formed partially or wholly of an admixture of
refractory sand and clay, which is preferable, as bricks or
stones are liable to crack and burst, owing to the frequent in-
terruption of the process. The furnace lining is surrounded by
an iron mantle, and the space between them is sometimes
filled up. This light construction lessens the loss of tem-
perature at the frequent blowing-in of the furnaces. The
cast-iron mantle, from 3-4 to 5-4ths of an inch thick, consists
either of one or more cylinders, or is formed of iron plates
screwed together. It rests upon a bottom plate which covers
the foundation; the foundation is provided with channels for
the escape of moisture. The top of the mantle is covered
with an iron plate, leaving an opening for the furnace mouth.
The sole of a furnace with a closed front wall is from 6 to
8 inches thick, and formed of refractory sand and clay, with
an inclination towards the tapping-hole, to which is fixed an
iron gutter coated with sand. The tapping-hole is placed in
an opening of the furnace 12 inches wide and 15 inches high,
which is walled-up during the process with loam and coke,
or closed with an iron plate coated with loam, which may be
removed after the melting to clean the furnace hearth.
Some of the furnaces with closed front walls are also pro-
vided with openings for running off the slag.
The furnaces stand below a chimney, and some distance
above their mouth the hot blast apparatus is fixed, if one is
used. When making very large castings, the furnaces are
made movable by placing them upon cars running on rails
(Gleiwitz).*
The height of coke furnaces is from 6 to S, sometimes 10
or 12 feet, that of charcoal furnaces from 10 to 20 feet. The
width at the level of the tuyere in smaller furnaces varies.
between 12 and 24 inches, the tuyere being 12 or 15 inches
above the bottom, and the capacity of the furnace from 6 to
20 cwts. Larger furnaces are up to 3 feet wide, whilst the
tuyere is 20 inches high, and the capacity of the furnace
NK
HARTM., Fortschr., i., 239.
590
IRON.
from 50 to 55 cwts. The cupola furnaces of the foundries
of Berlin,* for example, vary in capacity; the furnace of
Egells contain from 40 to 100 cwts.; those of Borsig, from
30 to 40 cwts.; of Freund, 40 cwts.; of Eckert, from 1 to
40 cwts.; of Beermann, 20 or 30 cwts. ; of Wedding, from
6 to 40 cwts.; at Gleiwitz the furnaces contain from 70 to
100 cwts.; the charcoal furnaces at the Hartz, 12, 15, or 20
cwts.; and those at Dirchau,† 100 cwts. A cupola furnace
9 feet high and 21 inches wide, and charged with burdens of
1 cubic feet of coke and 200 lbs. of pig-iron, produces
25 cwts. of iron per hour, and may contain at the highest
90 cwts.
The number of the tuyeres and their height above the fur-
nace sole depend on the amount of the production and the
width of the furnaces.
In small furnaces, with a capacity of about 10 cwts., one
tuyere is sufficient; furnaces with a capacity of 20 or 25 cwts.
require two tuyeres; and furnaces containing 30 cwts. and
more are provided with one tuyere on the back wall and two
tuyeres on the two sides respectively; or the tuyeres are
divided according to Sefström's principle over the periphery
of the furnace. In some cases they are placed round the
furnace in a spiral form (Woolwich Arsenal, Gleiwitz), so
that the blast enters the furnace at different heights. The
charges are thus quickly heated at a rapidly increasing tem-
perature, inducing a quick descent; and a greater quantity of
very hot iron is produced in a given time, whilst the con-
sumption of coke and the loss of iron are decreased. In
order to accelerate the melting, and consequently to econo-
mise the fuel, Hinton|| employs two rows of tuyeres after
Sefström's principle, one above the other, whilst the furnace
is enlarged both above and below the tuyeres. In the Las-
witz establishment in Breslau, a furnace 9 feet high and of
a capacity of 80 cwts. contains four tuyeres, two of which
* B. u. h. Ztg., 1862, p. 6. Zeichnungen des Ver. Hütte, Jahrg., 1854,
No. 4; 1855, No. 15.
+ Zeichnungen des Ver. Hütte, 1855, No. 21; 1856, No. 8 k.
‡
Polyt. Centr., 1861, p. 444. HARTM., Fortschr., v. 171.
|| B. u. h. Ztg., 1855, pp. 170, 187. Preuss. Ztschr., iii., A. 166.
THE CONSTRUCTION OF CUPOLA FURNACES.
591
lie one above the other and opposite to the other two.
Bocard's furnace (Fig. 171) has no tuyeres, and the
blast enters the furnace through an opening round the
movable hearth. In order to increase the capacity of the
furnace two or more tuyeres are laid on above another
(Figs. 163, 165, 166, 167). The lowest tuyere is then used
first, whilst the upper ones are closed with clay. This tuyere
is then closed as soon as the liquid iron reaches it, and the
tuyere next above is opened, and so on. Supposing the
lowest tuyere or row of tuyeres to be 15 inches above the
furnace sole, the distance to the next is always less by or
I inch, thus forming spaces for collecting the iron up to 30
and 60 inches high, capable of holding as much as 200 cwts.
of iron.
3
Another mode of collecting more liquid iron is to tap the
contents of the furnace into a large pan or a special receptacle,
and after the furnace is re-filled, to use both lots of iron.
together. But this mode can only be employed when the
temperature is very high.
The space between the tuyeres and the sole of the furnace
must be sufficient to allow room for collecting iron in the
hearth in such quantity that it will not become thickly
liquid or come in contact with the current of blast. In
charcoal furnaces the height is usually 10 or 15 inches, and
in coke furnaces 20 or 24 inches and more.
The tuyeres
usually lie horizontally, and never have a downward inclina-
tion, or the iron would be decarbonised; a slight upward
inclination sometimes improves the quality of the iron.
The width of the copper or cast-iron tuyeres and of the
nozzles depends on the quantity of blast to be introduced,
and on the pressure of the blast. The tuyeres of coke fur-
naces requiring a great quantity of blast of low pressure (page
587), are from 3 to 8 inches in diameter, whilst charcoal
furnaces, requiring less blast of a higher pressure, have nar-
rower tuyeres (from 2 to 4 inches diameter) in which the
movable blowpipe nozzles closely fit.
In order to save fuel, the furnaces at Blansko in Moravia
are provided with a separate tuyere close to the furnace sole.
On blowing-in, charcoal and then coke are placed on the
592
IRON.
furnace bottom, and the lower tuyere only is put in action
for three or five minutes, thus quickly heating the furnace
sole; the lower tuyere is then closed, blast is introduced
through the higher tuyeres, and the full burden of iron is
charged at once. As the furnace sole is already very hot,
the melted iron will be grey from the commencement. A
similar method has also been applied at the blowing-in of
blast furnaces.*
The furnaces only allow a certain number of smeltings (up
to 30), according to whether the building materials are fire
proof; and the faulty parts, chiefly the hearth and boshes,
require repairing from time to time. In order to preserve
the furnace shaft during these repairs, the upper part of the
furnace is sometimes constructed so as to be independent of
the lower part (Maillard's cupola furnaces*), and it may be
raised on screws, thus allowing the lower part to be repaired;
or the upper part is fixed, and the lower part is movable
on rails (Boccard's furnace).†
The following may be taken as illustrations of cupola
furnaces.
A. Charcoal Cupola Furnaces.-Figs. 157, 158, and 159
show a furnace used in Lerbach in the Hartz. a is the
foundation of brickwork provided with a cross channel; b,
the bottom plate; c, the cast-iron mantle; d is an iron plate
covering the top of the furnace; e is the fore hearth resting
upon a projection of the bottom plate and screwed to the
mantle, c; f is the tapping hole; g, the opening for charging
the furnace; h, the Wasseralfinger hot blast apparatus;
i, the chimney. The tuyeres and blowpipes are 1 inches in
diameter; the blowpipes are movable and may be pushed
forward so as to reach into the mouth of the tuyeres, thus
closing them. The two tuyeres in one furnace are placed
opposite each other, 1 inches apart. k is sand beaten down
round a hollow wooden mould, which is afterwards burnt in
the furnace. The tymp of these furnaces is sometimes formed
of sandstone.
* B. u. h. Ztg., 1859, p. 167.
† Polyt. Centr., 1858, p. 1462. HARTM., Fortschr., ii., 260.
THE CONSTRUCTION OF CUPOLA FURNACES.
593
Fig. 160 represents the charcoal cupola furnace used at
Königshütte (Hartz). a is sand; b, sandstone; c, bricks ;
FIG. 157.
FIG. 158.
57
B
f
C
K
b
D
FIG. 159.

9
d, filling; c, rough walling; f, an iron mantle. The tuyere,
2 inches in diameter, is inclined upwards one-fourth of an
inch. The diameter of the blast pipe is 23 inches.
B. Coke Cupola Furnaces.-Figs. 161 and 162 show a
coke cupola furnace in use at the Royal foundry at Berlin.
a is the blast main; b, a cylindrical ring; c, the tuyeres; d is
the sole plate; g, an iron plate inside which coke is placed;
h is the tapping hole; i is an opening closed with a glass lid,
giving a view of the furnace through the nozzles, and also.
for cleaning the tuyeres; k are fire bricks; shows the
filling-in material, and m, an iron mantle.
VOL. II.
2 Q
594
IRON.
Fig. 163 represents a coke cupola furnace used at Wasser-
alfingen. a is the furnace shaft; b, the tuyeres ; c is an iron
FIG. 160.
FIG. 161.


+
V
a
4
:a.
نام
m
r
m
C
plate on which the charges glide down; d, the opening for
charging; e a Scotch hot blast apparatus; f, the chimney.
FIG. 162.

で
​The furnace used at Gleiwitz (Silesia) is shown in Fig. 164.
The furnace is provided with six blast pipes 1 inches in
diameter, which are placed in the blast channel, a, in a
spiral form, so that the sixth blast pipe lies 2 inches higher
than the first one. The outside of the channel, a, contains
openings closed with glass lids and corresponding with the
THE CONSTRUCTION OF CUPOLA FURNACES.
595
blowpipes; b is the fore hearth; c is an opening for cleaning
the hearth, kept closed during the process; the tapping
FIG. 163.
}
d
2-
α
b
3
4
S
J FEET
FIG. 164.


نا
α
2
4.
6
8
FI
hole is also on this part of the furnace; dis an opening for
charging the furnace. The iron used for casting cooking pots,
&c., runs continually through the tapping hole into a ladle.
Figs. 165, 166, and 167,* represent the plan, section, and
elevation of a coke cupola furnace capable of founding 5 tons.
of cast-iron at a time. It is 3 feet wide inside and 13 feet
high. m, m is a solid body of masonry forming the basis to
the furnace; b, b is an octagonal platform of cast-iron, with
Dr. URE'S Dictionary of Arts, &c., vol. ii., p. 394.
292
596
IRON.
a ledge on which the plates a, a, a, a rest; a, a are eight
plates of cast-iron, I inch thick, exactly alike; only one of
them is notched at its lower part, at c, to allow the melted
FIG. 165.
FIG. 166.
880

ad
Th
d
O
a d
d
(IN
d
이
​Dd
d
ad
(I)
D
a
Id
d
ad
Q
T
d
та
Ed
I
d
Ad
Dd
I
万
​N
C
ሪ
Ъ
metal to run out, and two of the others have six apertures,
g, g, g, to admit the tuyeres; c is an orifice for allowing the
metal to flow out. A kind of cast-iron gutter, e, lined with
FIG. 167.

Λ
CA
loam, is fitted into the orifice; d are hoops of hammered iron,
44 inches broad; the lower ones are half an inch thick, and
the upper ones a quarter of an inch. The intermediate
hoops decrease regularly in thickness between these limits.
e is a cast-iron gutter or spout, lined with loam, for running
off the metal; f, f, are cylindrical pieces of cast-iron for in-
creasing the height and draught of the furnace; g, g, are
side openings for receiving the tuyeres, of which there are
six upon each side of the furnace. Either of them may be
THE CONSTRUCTION OF CUPOLA FURNACES.
597
shut at pleasure by means of a small cast-iron plate, h,
made to slide horizontally in grooves sunk in the main plate,
pierced with the holes, g, g; k, k is the interior lining, made
of somewhat argillaceous sand, in the following way after
laying at the bottom of the furnace a bed of sand a few
inches thick, slightly sloped towards the discharging orifice,
there is set upright, in the axis of the cupola, a wooden
FIG. 168.
FIG. 169.


}
h
C
d
C
C
b
1
ん
​O O O O O O
O
h
d-
Ъ
a
a
m
2
3
4
b
ઊં
5
SFI
cylinder of the full height and of a diameter a little less than
that of the vacant space at the top of the furnace. Sand is
then rammed in so as to fill the whole of the furnace; after
this the wooden cylinder is withdrawn, and the lining of the
sand is cut or pared away until it has received the proper
form.
This lining generally lasts five or six weeks, when there are
six meltings weekly. i, i is a cast-iron circular plate through
598
IRON.
which the mouth of the furnace passes to protect the lining
in k, during the introduction of the charges; N, N is the level
of the floor of the foundry. The portion of it below the
running-out orifice consists of sand, so that it may be
readily sunk when the melted metal is to be received in
ladles or pots of large dimensions.
Maillard's cupola furnace,* with a movable upper part, is
shown in Figs. 168 and 169. a is the hearth of fire-bricks or
sand, surrounded up to the height of 1'1 metre with an iron
FIG. 170.
FIG. 171.


о
O----
O
711
e
I
a
f
K
2
3
LET
mantle, b, which is provided with three apertures for the
tuyeres, c, and two for discharging the slag, d; e is the upper
B. u. h. Ztg., 1859, p. 167.
HOT BLAST APPARATUS.
599
part of the hearth, and ƒ, the boshes, likewise formed of fire-
bricks or sand; this part of the furnace is not surrounded
with an iron mantle, in order to facilitate repairs which may
be required; g is the upper part of the furnace, the shaft,
surrounded with the mantle, h, formed of iron plates, and
the border of which rests upon the four pillars, i. The
pillars rest on the projections, h, of the mantle, b, to which
they are screwed by means of sockets. By loosening the
sockets, the upper part of the furnace may be lifted some
centimetres above the boshes, which then can easily be re-
paired when cool. m is the tapping hole; ", a gutter in
which the iron flows off. The tuyeres are o°145 metre in
diameter, and the blast pipes o'140. A furnace of this kind
contains from 900 to 1000 kilos. of iron.
Maillard's movable furnace, represented by Fig. 170, has
two tuyeres o'07 metre in diameter, and an aperture through
which the slag runs off, o'06 metre wide, and resting upon the
car, b; otherwise its construction is similar to that of the
preceding furnace. This furnace only contains from 60 to
100 kilos. of iron.
Fig. 171 shows Boccard's furnacet with a movable hearth
or crucible. a is the cast-iron box for the furnace hearth
lined with fire-bricks, resting upon the car, b, which runs on
the rails, c. The hearth, a, is surrounded by the box, d, and
this again by the second box, e. The blast enters into the
space, ƒ, and thence. at g, into the furnace all round the cir-
cumference of the hearth instead of by tuyeres, thus causing
a very uniform smelting; k is the tapping hole; and h, the
aperture for running off the slag.
B. Hot Blast Apparatus.-The hot blast has also proved
advantageous in the cupola furnace process with regard to the
consumption of fuel, the production,, and the loss of iron in the
smelting process, but these advantages are less obvious owing
to the short operations of the furnaces. The saving in char-
coal amounts to one-third, and that of coke even to one-half;
the former loss of iron (9 or 10 per cent) has been reduced to
* B. u. h. Ztg., 1859, p. 168.
†DINGL., Bd. 150, p. 186. HARTM., Fortschr., ii., 260.
600
IRON.
5 per cent, and a hotter and denser iron is produced. Hot
blast enables the re-melting of wash and scrap iron without
danger, and more easy manipulations in the hearth; the
resulting slag is hotter, the waste flame more lively, and the
tuyeres brighter. The blast is seldom heated above 250° C.,
otherwise the furnace is too strongly attacked, and the iron
will be too hot for foundry purposes. For these reasons, and
for that stated on page 588, cold blast is frequently employed
(Hartz) when re-melting grey foundry iron with coke, as in
this case it produces a higher temperature.
The usual apparatus for heating the blast are either the
Wasseralfinger (vide Fig. 158 on page 593), or the Scotch (as
shown in Fig. 163 on page 595). Sometimes, also, the
apparatus represented by Figs. 172, 173, 174, are advan-
tageously used. They are fixed above the furnace mouth
FIG. 172.
FIG. 174.
FIG. 173.

E
f
B
Lu
פוז
B
22:010
2-
D
E
3 FT
J FI
A
B
g
F
and consist of two annular tubes or boxes, A and B, which lie
parallel and horizontally one above the other. These boxes
are divided by twelve partition plates into as many divisions,
a, which communicate with each other by means of vertical
tubes. The cold blast enters from the tube, d, through the
adjoined tube, c, into the upper annular box, passing up and
down in the vertical tubes until it issues fully heated from
the tube, c, as the boxes and the vertical tubes are heated by
BLOWING ENGINES.
601
the waste gases of the cupola furnace; ƒ is the top plate of
the furnace, and g, g, are the supports of the apparatus.
C. Blowing Engines.-As the smelting process in cupola
furnaces only requires a large quantity of blast of a low
pressure (page 587), fans are usually employed in preference
to the more expensive cylinder blast engines, which for equal
quantities of blast produce a high pressure, and can only
produce the large quantity required by making many revolu-
tions and having cylinders of large diameter. The cylinder
engines also require expensive regulators.
Figs. 175 and 176 represent a fan much used in Germany;
it is constructed by Schwarzkopf* of Berlin, upon Schiele's
FIG. 175.
FIG. 176.

9
C
d
g
d
α
ur
Ꮕ
12.
d
TR
principle, and makes less of that disagreeable noise peculiar
to most of the other ventilators. g is a fixed iron casing;
f is the fan-wheel carrying the air, entering by a, towards
the circumference, k, whence it enters the compartment, d,
which is provided with curves for conducting the air into the
main tube, c. The wings, m, are cast with the wheel, f, in
one piece; is the frame supporting the fan; r, the driving
pulley; w, the driving shaft; p is a grease box from which
grease is conducted to the axle by the pipe, i.
*
The fant represented by Fig. 177, is of the best excentric
WEISBACH, Ingen. u. Maschinen-Mechanik, iii., 1154. ROMBERG, Ztschr.
f. Baukunst, 1855. WIEBE'S Maschinenbaumaterialien.
+ URE's Dictionary of Arts, &c., vol. ii., p. 396.
602
IRON.
་
form, as constructed by Messrs. Braithwaite and Ericsson;
D is the circular orifice round the axis (by this the air is
FIG. 177.
E
B
??

admitted), and c, c, в is the excentric channel through
which the air is wafted towards the main discharge pipe, E.
The fan distributes the blast from the main pipe to three
BLOWING ENGINES.
603 ·
principal points by three branch distributing tubes. A register,
consisting of a cast-iron plate sliding stiffly in a frame, serves
to intercept the blast at any moment when it is not desirable
to stop the moving power. A large pipe of zinc or sheet
iron is fitted into the orifice of the slide valve. It is square
at the beginning, or only rounded at the angles, but at a
little distance it becomes cylindrical, and conducts the blast
to the point of divergence. There, each of the branches
turns up vertically, and terminates at b, b, where there is a
circular orifice 7 inches in diameter. Upon each of the
upright pipes, b, one end of an elbow tube of zinc, c, c, is
adjusted rather loosely, and the other end receives a tuyere
of wrought-iron, d, d, through the intervention of a shifting
hose or collar of leather, c, c, d, hooped with iron wire to
both the tube and the tuyere. The part, c, c, may be raised
or lowered by sliding upon the pipe, b, in order to bring
the nozzle of the tuyere, d, d, to the requisite point in the
furnace. The part, c, c, c, c, may also be made of wrought-
iron. A 4-horse power is required for driving this fan to
supply blast to three furnaces.
Most of the fans used in the foundries at Berlin are 3 feet
in diameter, and 15 inches broad; they are set in motion by
a steam engine of 6-horse power; they make Soo revolutions
per minute, and provide three large cupola furnaces with the
required blast (about 2000 cubic feet) of or 1 lbs. pres-
sure.
4
The quantity of blast required depends chiefly on the size
of the production, which again is influenced by the dimen-
sions and construction of the furnace, and also depends on
the nature of the fuel and the iron to be melted.
Coke cupola furnaces require from 400 to 700 cubic feet of
blast per minute, according to the compactness of the coke,
&c.; charcoal furnaces from 250 to 400 cubic feet. The
pressure in coke furnaces amounts to from 4 to 8 lines, and
with charcoal to from 9 lines to 2 inches, and seldom less
than 6 lines of mercury.
As each 100 lbs. of iron to be smelted per hour in a coke
cupola furnace requires about 5000 cubic feet of blast, the
604
IRON.
following important data concerning the blast to be intro-
duced, the diameter of the blast pipes, &c., will be useful:-
Pressure of the
Blast.
Quantity of Blast
per square inch of
the section of the
Blast Pipe.
Section of the
Blast Pipe per
100 lbs. of iron
per hour.
Square Inches.
Quantity of iron
per hour and per
I square inch of
the section of
the Blast Pipe.
lbs.
Water
Inches.
Cubic Feet.
2.91
0'70
4°19
0.84
5.67
0.98
7'36
I'12
9°28
I'26
2'00
1.66
1'43
1*25
I'II
50
60
70
80
90
The following are some illustrations for the determination
of the amount of blast :-
a. A cupola furnace has three tuyeres 6 inches in diameter;
each is fed with blast of 7.36 inches of water pressure; how
much iron can it fuse per hour? At the already stated pressure
of the blast 80 lbs. of iron will result per hour per square
inch of the blast pipe; three blast pipes, therefore, with
85 square inch section, will produce 80 × 85=68 cwts. of iron
per hour.
b. How many square inches must be contained in the
section of 4 blast pipes; 40 cwts. of iron are to be produced.
per hour with a pressure of 7.36 inches?
As 100 lbs. of iron at the given pressure require 1*25 square
inches in the section of the nozzles, 40 cwts. will require
40 × 1*25=50 square inches; therefore, each blast pipe must
have a section of 125 square inches or about 4 inches
diameter.
The older coke cupola furnace at Lerbach (Hartz) having
a blast pipe 5 inches in diameter, was worked with 600 cubic
feet of cold blast per minute with a pressure of 2 or 3 ounces;
the charcoal furnace (Figs. 157, 158, 159) with a blast pipe of
2 inches 9 lines diameter was fed with from 395 to 425 cubic
feet of blast of from 170° to 200° C.
The Cupola Furnace Process.
When blowing in a new charcoal cupola furnace, glowing
coal is kept in the furnace hearth for 8, 12, and even 24 hours;
the hearth is then cleaned from deposits and the furnace is
THE CUPOLA FURNACE PROCESS,
605
I
gradually filled with charcoal, when the preceding layer will
ignite the following one; blast of low pressure (4 to 6 lines
mercury) is now introduced, and the first charge of 25 lbs.
of pig-iron is made (at the later startings of the furnace
higher charges of 50, 60, or 70 lbs. are commenced). As soon
as the seventh or eighth charge enters the hearth, after 1 or
13 hours, the tapping hole will be sufficiently warm and is
closed with a plug consisting of two-thirds loam and one-
third horse-dung or sand; the opening below the tymp is
closed with sand, pounded slag, or a mixture of small coal
and loam, to prevent the blast from blowing through; the
full pressure of from 9 to 36 lines of mercury is then given;
the iron charges, according to the state of the process,
are gradually increased by from 10 to 15 lbs. up to from 90 to
105, or 110 lbs. at the highest, at the same time charging
constant burdens of charcoal (about 2 cubic feet of beech coal
= from 19 to 22 lbs.). The quicker mode of blowing in which
is used at Ransko has been explained on page 591.
The smelting materials are raised to the furnace top by a
lift, or on an inclined plane. The iron to be charged (by
weight) is broken into pieces of about 8 cubic inches, and
in coke furnaces into larger pieces; the fuel must be dry, as
uniform as possible, and in moderate sized pieces, and it is
charged by volume of a known weight; these measures of fuel
being from time to time controlled by direct weighing. The
size of the single charges depends on the quality of the fuel
and the pig-iron, on the dimensions of the furnace, on the
process, &c.; for example, coke is charged, according to the
dimensions of the furnace, in burdens of from to I cubic foot
(24 to 36 lbs.), or from 2 to 3 cubic feet (72 to 108 lbs.), and
iron in burdens of from 100 to 500 lbs. and more; the charcoal
burdens are from 1 to 2 cubic feet. Usually from 200 to
400 lbs. of iron are charged in pieces of 8 cubic inches, with
from 25 to 50 lbs. of coke.
At a regular process the iron in the hearth is hot and grey
(gaar), the tuyeres are not too light, and the coal may be
seen when looking through the tuyeres; the descending pieces
of iron rapidly melt, the whitish yellow flame of the furnace
mouth is lively and hotter than that of iron blast furnaces,
606
IRON.
and the smoke issuing from the furnace mouth and from
below the tymp is mostly blue.
If there is too high a temperature in the furnace it pro-
duces very hot and thinly liquid iron, which strongly attacks
the hearth lining; the charges of iron must then be increased,
and either the furnace is fed with burnt iron through
the tuyeres, or the liquid iron is allowed somewhat to cool
in the ladles, sometimes even a small quantity of pig-iron
is added to it; otherwise the castings sink, and the moulds.
are injured. If an increase of the iron burden causes the
thinly liquid iron to be easily converted into thickly liquid
iron it is a sign that the furnace is too wide.
In the reverse case, if the temperature is too low the
iron will be thickly liquid and inclined to chill, the slags
stick to the tuyeres, forming a nose, and the flame
of the furnace mouth is sluggish. The burdens of iron
must then be decreased, or the temperature of the blast
raised; the tuyeres and hearth are cleared of deposits,
pounded flame is repeatedly forced below the tymp with
a view to conducting the flame more to the front of the
furnace, thus causing the iron in the fore hearth to be
hotter; the iron must be always kept covered with glowing
coal.
If the iron is thickly liquid without chilling, either the
blast is too weak, or the slag too pasty, and the production
of thickly liquid iron at the commencement of the fusion
indicates that the tuyeres lie too high. If the charges become
suspended in the furnace, the blast pipes must be moved back-
wards and the blast turned off. The slag is either removed
from the fore hearth, or runs off by itself through apertures
into boxes; in some cases it is tapped off from time to time
or it runs off together with the iron, when it is removed from
the founding ladles.
After some hours (three or four), when a certain number of
charges have been made, the hearth is filled with iron, the
blast is put off, and the iron is either ladled out from the
fore hearth or it is tapped off in larger pans; when casting
large castings it is made to run direct into the moulds.
If more iron is to be collected in the hearth, the lower
THE CUPOLA FURNACE PROCESS.
607
tuyeres are closed with clay, and blast is introduced through
the higher tuyeres.
In order to ascertain the quantity of iron which a furnace
hearth will contain, and which may be required in certain
cases, the iron is tapped off when the hearth is filled, and
its weight taken, observing, at the same time, the number
and the weight of the charges which were required to fill the
hearth. In order to ascertain the quantity of iron contained
in the hearth at any moment, it will afterwards only be
necessary to calculate that quantity from the charges which
have passed through the furnace.
After emptying the hearth, cleaning it from deposits, and
closing the tapping hole, it is re-filled with coal, the space
below the tymp is closed, and the smelting operation re-
commenced. The tapping hole must be carefully preserved
from injury, and brushed over with black lead.
Smelting in charcoal furnaces is usually interrupted every
evening after an operation of 14 or 15 hours. An hour pre-
vious to the last tapping, and at the last charge, double the
usual quantity of iron is charged, and the pieces of iron are
larger, the better to utilise the coal; also from 8 to 10 lbs.
of limestone are frequently added to the last charges on
account of the scorified particles of the furnace lining. When
all the melted iron is contained in the hearth, it is ladled out
or completely tapped off, the fore hearth and tymp are
cleansed of deposits and coal, the tapping hole is repaired
and closed with a mixture of small coal and loam, or with
an iron rod; the tuyeres are closed with loam, the hearth filled
with fresh coal, and the fore hearth as well as the furnace
mouth closed with a luted iron plate; the heat is thus re-
tained in the furnace with a consumption of only about
2 cubic feet of charcoal. In the morning the fore hearth,
furnace mouth, tuyeres, and tapping hole are reopened, the
furnace filled with coal, and light burdens of iron of from
25 to 35 lbs. are charged at the commencement, whilst blowing
with a pressure of 4 or 5 lines; the burdens and pressure
are afterwards rapidly increased.
According to the nature of the refractory material it is
built with, the furnace admits of longer or shorter operations,
608
IRON.
usually from four to six weeks; slight repairs may be done
during the operations. The loss of iron upon re-melting
varies according to the quality of the pig-iron, fuel, and the
process, and usually amounts to from 5 to 7 per cent, in the
most unfavourable case to 12 per cent, and upwards.
The process in coke cupola furnaces is similar to that in
charcoal cupola furnaces, with the modifications stated on
page 585. At the commencement the coke is ignited with
wood-shavings, wood, and glowing coal, and as much as
5 per cent of limestone may be added to each charge for the
better scorification of the coke ash; the slag is either tapped
off from time to time through an opening with which the
furnace is provided for this purpose (Fig. 168), or it is allowed
to flow off continually. Most of the furnaces have no fore
hearth, and their operations last only four or five hours, when
they require cleaning, as the formation of slag gives rise to
many deposits in the furnace.
According to Knop,* a saving of fuel is obtained if at the
commencement the furnace is filled with coal to half its height
only, thus warming the furnace, and charging the first burden
of iron 1 or 1 hour previous to the introduction of the blast.
The products of the cupola furnace process are:-
1. Cast-Iron, more compact and of finer grain than before
re-melting, and from mottled to grey according to the process.
2. Slags formed of the slag of the fuel, the sand and slag
sticking to the pig-iron, particles of the furnace lining, the
limestone added, &c. ; they are blue, grey, brown, or yellow,
vitreous, porcelain-like, stony or earthy, and frequently
evolve sulphuretted hydrogen when water is thrown upon
them. The slag of a cupola furnace in which cast-iron was
melted with an addition of lime showed the following com-
position:-
Sio,
A₁₂O₁
3
FeO
45'59
11.88
I'II
MnO.
O'91
CaO.
38.20
CaS
1*76
*DINGL., Bd. 168, p. 157.
THE CUPOLA FURNACE PROCESS.
609
Grey coke pig-iron from Rans* containing-
Fe.
Si.
C.
94'50
I'95
3°09
yielded, when smelted with coke containing 12 per cent ash,
in a cupola furnace at Casamène, a slag of the following
composition :-
SiO 3
Al₂O₂
FeO
CaO
MgO
3
31.8
24°2
22.4
21'0
0.6
The slags of charcoal cupola furnaces are pounded and
washed to recover the contained iron.
3. Waste Gases.t-These are hotter and richer in carbonic
acid than the waste gases of blast furnaces, and the waste
gases of coke cupola furnaces more so than those of charcoal
furnaces; often no use is made of them, or they are used to
heat the hot blast apparatus. Sometimes they are used for
a preparatory heating of the water for steam boilers (machine
manufactory of Hoppé in Berlin), or for the direct produc-
tion of steam (Bernburg).||
The following are examples of the process in charcoal
cupola furnaces :-
At Königshütte in the Hartz (Fig. 160 on page 594).-The
pig-iron used for re-melting is cast so as to form cubes of
from 2 to 4 cubic inches; this iron is mixed with deficient
castings, waste of the foundry, and chips and borings of cast
and wrought-iron in the proportion of two-thirds to one-third,
and in charges of So lbs. The burden of fuel (beech-wood
charcoal) consists of 2 cubic feet, or about 20 lbs. The dia-
meter of the tuyere is 2 inches, and that of the blast pipe
23 inches; the pressure of the blast is 18 lines (15—24), and its
temperature 156° to 312° C.; the quantity of blast introduced
per minute is 350 cubic feet, and from 70 to So charges are
B. u. h. Ztg., 1862, p. 26.
+ Analyses: Bgwkfd., viii., 467.
125. Polyt. Centr., 1847, p. 917.
B. u. h. Ztg., 1862, p. 33.
Employment of the Gases: Bgwkfd., xi,,
B. u. h. Ztg., 1852, p. 261; 1855, p. 177.
|| Ibid., 1855, p. 177 (with drawing).
VOL. II.
2 R
610
IRON.
made in 12 hours. The loss on melting 100 lbs. of pig-iron
amounts to 2.66 per cent, and to 5'01 per cent per 100 lbs.
of castings.
At Lerbacherhütte in the Hartz (Figs. 157, 158, and 159 on
page 593). The iron to be melted is of the same kind
as that used at Königshütte. One charge consists of from
= 22 lbs. of char-
90 to 100 lbs. of iron and 2 cubic feet
coal; about five charges are melted in an hour. The tuyere
and blast pipe are 14 cubic inches in diameter, the pressure
of the blast is 30 lines, and its temperature 245° C. The
loss on smelting is 3'14 lbs. per 100 lbs. fused iron, and
5*26 per 100 lbs. of resulting castings. The consumption of
coal per 100 lbs. of castings is 45'4 lbs. The furnace hearth
may contain from 12 to 15 cwts. of iron, or even 20 cwts.
if a second tuyere is employed above the lower one.
At Mariazell, one charge consists of 33 cubic feet of char-
coal and from 100 to 110 lbs. of iron (at the commencement
from 60 to 80 lbs.) The furnaces have two tuyeres 1 inch
wide and 3 of an inch high; the blast pipe is 1 inch wide,
and the pressure of blast from 11 to 13 lines mercury. From
60 to 100 cwts. of iron are melted in 12 hours at a loss of
5 per cent.
4
At Blansko, from 60 to 90 cwts. are melted in 12 hours at
a loss of from 6 to 8 per cent, and 3 cubic feet of hard coal
are consumed per 100 lbs. of iron. The diameter of the
blast pipe is from 14 to 24 lines, and the pressure of the
blast from 18 to 36 inches of water.
At Lauchhammer, one charge consists of I cubic foot of
charcoal and from 40 to 50 lbs. of iron (at the commencement
20 lbs). The blast pipes are 1 inches wide and 3 of an inch
high, and the pressure of the blast amounts to 16 inches of
water. From 40 to 50 cwts. of iron are melted in 12 hours
at a loss of 4 or 5 per cent, and a consumption of from 2 to
3 cubic feet of coal per 1 cwt. of iron.
The following are examples of smelting in coke cupola
furnaces:
Katzhütte in Thuringia.-The furnace is provided with a
fore hearth, and is 8 feet high, and I foot 6 inches and
I foot 9 inches wide at the top and bottom respectively. It
THE CUPOLA FURNACE PROCESS.
611
is provided with five tuyeres, each 5 inches wide; the pres-
sure of the blast is from 14 to 16 inches of water. One
charge consists of 1 cubic feet of Westphalian coke and up
to 4 cwts. of pig-iron (25 per cent Scotch pig-iron and
75 per cent pig-iron of own production). Some fluor spar is
added to every second charge; the hearth may contain
50 cwts. of iron.
An older furnace at Lerbach on the Hartz without a fore-
hearth is 7 feet high, 2 feet wide on the bottom, and 2 feet
2 inches in the belly. The diameter of the blowpipe nozzles
is 5 inches, the pressure of the blast 2 or 3 ounces, the tem-
perature of it 19° C., and the quantity of blast amounts to
600 cubic feet per minute.
At Saynerhütte, in 1861, 33°12 lbs. of coke were consumed
per 100 lbs. of pig-iron, and the loss of iron in melting
amounted to 8'91 per cent.
I
At Gleiwitz, from 110 to 120 cwts. of iron are melted in
five hours, the pressure of the blast being 1 lbs. The fur-
nace used is represented in Fig. 164. The foundry is ar-
ranged so as to allow the production of castings of more than
70 cwts.
In 1861, 46°5 lbs. of coke and 114°2 lbs. of pig-iron
were consumed per cwt. of castings.
Maillard's cupola furnace is shown in Figs. 168 and 169.
When blowing in this furnace, wood shavings and glowing
coal are placed upon the sole of the furnace, then from 150
to 250 kilos. of coke up to the shaft; an hour later 600 kilos.
of pig-iron are charged, and then alternate burdens of 41 kilos.
of coke and from 250 to 350 kilos. of iron, with an addition
of 2 or 3 kilos. of limestone per 15 kilos. of pig-iron, which
is charged between the fuel and the iron. The blast is next
introduced by the two lower tuyeres, and the tapping-hole
closed as soon as the liquid iron enters the hearth. When
blowing through the higher row of tuyeres the lower row is
closed with coke and foundry sand. The slag is tapped off
from time to time through apertures left for the purpose.
From 1200 to 1500 kilos. of very hot liquid iron are pro-
duced per hour, the pressure of the blast being 15 or 18 centi-
metres of water, and the consumption of coke 10:29 kilos.
2 R 2
612
IRON.
per 100 kilos. of pig-iron; the loss of iron in melting is less
than 4 per cent.
In the locomotive furnace, represented by Fig. 170, burdens
of 6 kilos. of coke, 50 kilos. of pig-iron, and 15 kilos. of
flux are charged.
In Boccard's cupola furnace (Fig. 171) from 1500 to 2000
kilos. of pig-iron are melted per hour. The slags continually
run off through the slag-hole. The iron is tapped off when
the hearth, a, is filled; the hearth in its box, a, is then re-
moved in order to be cleaned, the smelting column being
suspended with iron rods; a new hearth is sufficiently warmed
and put in the place of the old hearth.
The management of the furnace represented in Figs. 165,
166, and 167 is described in Dr. Ure's "Dictionary of Arts,
Mines, &c.," ii., 394, as follows:-
The furnace is kindled by laying a few chips of wood
upon its bottom, leaving the orifice, c, open, and it is then
filled up to the throat with coke. The fire is lit at c, and in
a quarter or half an hour, when the body of the fuel is suffi-
ciently kindled, the tuyere blast is set in action; the flame
then issues by the mouth as well as by the orifice, c, which
has been left open on purpose to consolidate it by the heat.
Without this precaution, the sides, which are made up of
argillaceous sand after each day's work, would not present
the necessary resistance. A quarter of an hour afterwards
the orifice, c, is closed with a lump of moist clay, and some-
times when the furnace is to contain a great body of melted
metal, the clay is supported by means of a small plate of
cast-iron fixed against the furnace. Before the blowing
machine is set going, the openings g, g, g, have been kept
shut; those wanted for the tuyeres are opened in succession,
beginning at the lowest, the tuyeres being raised according
as the level of the fused iron stands higher in the furnace.
The tuyere is either cylindrical or slightly conical, and its
diameter varies between 3 and 5 inches. A few minutes
after the tuyeres have begun to blow, when the coke sinks
in the furnace, alternate charges of coke and pig-iron must
be thrown in. The metal begins to melt in about twenty
minutes after its introduction, and successive charges are
THE CUPOLA FURNACE PROCESS.
613
then made nearly every ten minutes, each charge containing
from 3 to 5 cwts. of iron and a quantity of coke in proportion
to the estimate given below. The amount of the charges
varies, of course, with the size of the furnace and the speed
of the operation. The pigs must be previously broken into
pieces weighing at most 14 or 16 lbs. The vanes of the
blowing fan make from 625 to 650 revolutions per minute.
In the course of a year, a foundry of some extent will con-
sume about 300 tons of coke in melting 1240 tons of cast-
iron, consisting of 940 tons of pigs of different qualities, and
300 tons of broken castings, runners, &c. Thus it appears
that 24 lbs. of coke are consumed in melting I cwt. of metal.
Somewhat less coke is consumed when the fusion is pushed
more rapidly, to collect a great body of melted metal for
casting heavy articles; and more is consumed when, as in
making many small castings, the process of the founding has
to be slackened from time to time; otherwise, the metal
would remain too long in a state of fusion and probably
become too cold to receive sharp impressions from the
moulds.
It sometimes happens that in the same day, with the same
furnace, pieces have to be cast containing several proportions
of different kinds of iron; in which case, to prevent an
intermixture with the preceding or following charges, a con-
siderable bed of coke is interposed. Though there is thus a
little waste of fuel, it is compensated by the improved adapta-
tion of the castings to their specific objects. The founding
generally begins at 3 o'clock p.m., and goes on till 6 or 8
o'clock. One founder, aided by four labourers, for charging,
&c., can manage two furnaces.
The following is the work of a well-managed foundry in
Derby:-200 lbs. of coke are required to melt or bring down
(in the language of the founder) 1 ton of cast-iron, after the
cupola has been brought to its proper heat, by the combustion
in it of 9 baskets of coke each weighing 40 lbs. = 360 lbs. in all.
The chief talent of the founder consists in discovering the
most economical mixtures and so compounding them as to
produce the desired properties in the castings. One piece,
for example, may be required to have great strength and
614
IRON.
tenacity to bear heavy weights or strains; another must
yield readily to the file or the chisel; a third must resist
sudden alternations of temperature; a fourth must be pretty
hard, &c.
The filling in of the melted metal is managed in two ways.
For strong pieces whose moulds can be buried in the ground
at 7 or 8 yards distance from the furnace, the metal may be
run in gutters, formed in the sand of the floor, sustained by
plates or stones. When all is ready the clay plug is pierced
with an iron rod.
When from the smaller size or greater distance of the
moulds the melted metal cannot be run along the floor from
the furnace, it is received in cast-iron pots or ladles, lined
with a coat of loam. These are either carried by the hands
Between
of two or more men, or transported by the crane.
the successive castings the discharge hole of the furnace is
closed with a lump of clay, applied by means of an iron
or wooden rod having a small disc of iron fixed at its end.
After the metal is somewhat cooled the moulds are taken
asunder and the excrescences upon the edges of the castings
are broken off with a hammer; they are afterwards more
carefully trimmed or chipped by a chisel when quite cold.
The loss of weight in founding is about 6 per cent upon
the pig-iron employed. Each casting always requires the
melting of considerably more than its own weight of iron;
this excess forms the runners, false seams, &c., the whole of
which being deducted shows that I cwt. of coke is consumed
for every 3 cwts. of iron put into the furnace; for every
138 cwts. of crude metal there will be 100 cwts. of castings,
32 of refuse pieces, and 6 of waste.
C. Re-melting of the Pig-Iron in Reverberatory
Furnaces.
The advantages of cupola furnaces over reverberatory fur-
naces have been mentioned on page 580, and the latter
furnaces are now more and more often replaced by the
cupolas, the dimensions being enlarged. Reverberatory fur-
naces are principally used for casting very large pieces or
for re-melting large faulty castings; these furnaces are also
RE-MELTING IN REVERBERATORY FURNACES.
615
employed to remove certain components of the iron (silicon,
phosphorus, earthy metals, &c.) by the influence of the
atmospheric air, thus improving the quality of the iron and
producing a stronger iron for the manufacture of cannons,
rolls, &c. Sometimes the employment of reverberatory fur-
naces is necessitated by the fuel at disposal, brown coal for
example, or in cases when a foundry is to be carried on at a
certain place for a short period only, &c. Unlike the
cupola furnaces reverberatory. furnaces allow the iron to keep
liquid for any length of time, and also to be modified from
the darkest grey iron to all different tints of lighter grey up
to white iron. On the other hand, reverberatory furnaces
permit the advantageous re-melting of certain sorts of pig-
iron only, they do not allow the formation of castings in
uniform succession, and give rise to a greater loss of iron by
oxidation; owing to this circumstance, and to the larger con-
sumption of fuel, these furnaces are only admissible in large
foundries where several large castings are to be made in
succession.
Smelting Materials.
The iron used for re-melting in reverberatory furnaces is
mostly chosen from the grey sorts produced from refractory
ore mixtures in furnaces with high and narrow hearth; as
grey pig-iron from easily fusible ore mixtures in furnaces.
with low hearths, as well as mottled iron, are inclined to
become too light, when they are no longer fit for all
castings. Common grey iron of the regular process requires
to be only once re-melted to render it sufficiently strong,
whilst darkish grey or very impure iron requires a repeated
re-melting for that purpose; very impure or darkish grey
iron is liable to much scorification in the re-melting process,
but less so than white pig-iron, which, upon fusion, becomes
thickly liquid and more easily oxidises. The strongest iron
is obtained by re-melting white pig-iron (chiefly grained iron
poor in carbon) at a very high temperature, and slowly
Behaviour of the English, Scotch, and Belgian sorts of pig-iron in the pro-
cess of re-melting. LEOB., Jahrb., 1852, ii., 116, 160.
616
IRON.
cooling it; the same result is obtained on quickly cool-
ing a fused grey iron. Reverberatory furnaces enable the
production of good, pure, grained white iron from grey
pig. Different sorts of iron are frequently melted in ad-
mixture.
Fluxes are not usually employed at this melting,
unless an addition of limestone, which is given when re-
melting in blast reverberatory furnaces. At Mariazell,*
attempts have been made to extract some of the sulphur
from the iron by an addition of litharge, but without success;
the litharge, on the other hand, caused the formation of a
hotter iron, as it entered the froth-like slag and reduced it to
a thin layer, through which the temperature reacted more
effectively upon the iron.
The best fuel for re-melting in reverberatory furnaces is a
bituminous long flaming sinter coal; when this coal is not
to be had, kiln-dried wood and turf,‡ and also good brown
coal, can be advantageously used. Wood and turf, without
being kiln-dried, do not give the required temperature, but
they are applicable to puddling and re-heating furnaces. In
some cases fuel in the form of gas has been applied.
Karsten states that, according to the fusibility of the iron
and the amount of ash contained in the fuel, 100 lbs. of iron
require for re-melting from 40 to 80 lbs. of coal, or 6.5 cubic
feet of 20 lbs. each 130 lbs. of pine wood, or 43 cubic feet
= 185 lbs. of kiln-dried black turf.
According to Tunner the materials used at Mariazell for
re-melting 1 cwt. of pig-iron are from 6 to 8 cubic feet of
kiln-dried wood, from 50 to 80 cubic feet of good coal, from
15 to 18 cubic feet of kiln-dried turf, or from 70 to 100 lbs.
of brown coal.
Smelting Furnaces.||
These furnaces must be constructed in such a manner as
to produce iron of the required degree of liquidity in the
LEOB., Jahrb., 1862, xi., 300.
† Oesterr. Ztschr., 1857, p. 116.
B. u. h. Ztg., 1843, p. 765.
|| Ibid., 1860, p. 15. Oesterr. Ztschr., 1857, pp. 115, 188.
REVERBERATORY FURNACES.
617
shortest possible time, with the smallest consumption of fuel
and loss of iron. The production of the required very
high temperature is facilitated by a high chimney (fre-
quently of 100 feet), or by the application of blast below
the grate, by a fire place large in proportion to the
furnace hearth, by a suitably constructed roof of the fur-
nace, by forming the hearth of bad conductors of heat, by
good fuel, by frequently charging the fuel in small portions,
&c. Sometimes the furnace is placed in the open air in
order to facilitate its draught. The above-named conditions.
may be better obtained when it is intended to melt larger
quantities of iron quickly without considerably modifying it,
than when aiming also at a purification of the iron. Both
these purposes have also led to different constructions of the
furnaces, and we may clasify the furnaces into those having
a hearth inclined towards the fluc, and into furnaces with a
hearth inclined towards the fire-bridge. The former offer a
larger surface to the flame which is partly oxidising, and the
oxidation may be further facilitated by stirring the iron or
introducing the blast. But this operation always causes
great loss and the formation of much impure iron partly
decarbonised and partly burnt, which sticks to the hearth
and is termed slag by the founder; it is therefore usually
preferable to melt at once, and as quickly as possible, the suit-
able sorts of pig-iron of sufficient purity, in furnaces of the
second class. The formation of slag is lessened by a
higher fire-bridge, but, on the other hand, the temperature.
on the hearth of the furnace decreases.
In furnaces with the hearth inclined towards the flue, the
hearth is formed either of refractory bricks or of quartz
sand, and sometimes also of a mixture of quartz sand with
fromto of pulverised coke (Mariazell); this composi-
tion is a less powerful conductor of heat, but it is
also less durable, and requires to be renewed after each
melting, whilst the former constructions stand several opera-
tions. Although a horizontal position of the hearth permits
the production of a more uniform temperature in the furnace,
it is less often employed (Eck's gas furnace), as it causes a
greater consumption of fuel; too great an inclination of the
618
IRON,
hearth gives rise to a more easy scorification of the iron, as
it runs in thinner jets into the sump, and the iron charged
on the fire-bridge does not come in contact with the iron
which has become liquid before, thus its melting is impeded
and the formation of slag facilitated. The hearth has either
a uniform inclination, or a sump on its lower part, in which
the iron is more protected from oxidation. The tapping
hole is placed accordingly.
The dimensions of the furnaces chiefly depend on the size
of the charges of iron (20 to 200 cwts., mostly from 50 to
90 cwts.) on the quality of the fuel, &c. They are of the
right proportion if as uniform a temperature as possible is
produced in all parts of the furnace. When the iron melts
more quickly on the flue than on the fire-bridge, the flue is
too large, and if it is too small the quicker melting will take
place on the fire-bridge. The following proportions are
applicable to many cases :-Section of the hearth to the total
fire-place, from 3 to 4: I; to the open grate, from 6 to 7: 1;
the grate to the flue, from 6 to 8 : 1; the depth of the hearth
to its length, 1 : 2 or 3, when using bituminous coal, and
2: 3 when using non-caking coal; height of the fire-bridge,
from 4 to 12 inches (the higher if a slow melting and slight
oxidation of the iron is intended); the height of the chimney
from 30 to 100 feet, its width from 12 to 24 inches. The
opening between fire-bridge and roof is about equal to the
section of the mouth of the chimney at the top, and o'4 or
0'5 of the grate. The interstices between the grate are from
to 1 inches. According to Weniger the section of the
flue is 1-5th, 1-7th, and 1-9th, of the grate respectively,
according to whether coal, brown coal, or turf is used.
4
A furnace for smelting 100 lbs of pig-iron per hour with
coal must be of about the following dimensions:-The hearth
sole, 432; total fire-place, 1'08; open grate, o'65; open-
ing for the flame above the fire-bridge, from o‘26 to 0*325;
the chimney mouth, from o°26 to 0°325; and the flue o‘I
square foot.
When employing a lower chimney the blast is sometimes in-
troduced below the grate by means of a fan, or an exhauster is
applied to the chimney. In the first case the waste flame may
REVERBERATORY FURNACES.
619
be used for other purposes. The chimney is frequently pro-
vided with a small hole for observing the temperature.
Corbin-Desboissières* has constructed a parabolic furnace
with an inclination towards the flue, blast being con-
ducted below the grate. This furnace produced a most.
uniform temperature. Its construction is shown in Figs.
178, 179. a is the grate; b are racks; c, the ash-pit, with
FIG. 178.

~
72
m
P
TL
FIG. 179.
?
四
​3
U
ང་ མཀམ་ གོ་པ་
16
t
S
9
....
.....

て
​S
FI
the door, d; e are coal-rakes; f, a pipe conducting the blast
into the ash-pit; g, sliding doors through which coal is
thrown on the grate (the doors can be drawn up by means
of the forked blocks, h); i, space between the doors and the
grate; k and l are iron plates; m are draught holes connected
with n, for cooling the furnace walling; o is a roller before
the working door to support the working tools; p is the
chief smelting furnace; q is the tapping-hole; r, door
* B. u. h. Ztg., 1860, p. 15.
620
IRON.
provided with a loop-hole before the furnace sump; s are
doors leading to the space t, in which deficient castings
of small size are placed; the castings then melt in such a
time, running through the channels, u, that they unite with
the chief mass of iron in the sump; v are flues.
The furnaces with an inclination to the fire-bridge, which
in Germany are called Staffordshire furnaces, are in frequent
use in England as well as in Germany and France. The
sump of these furnaces is made of a refractory mixture of
clay and coal dust, and the hearth, or place on which the
iron is charged, is constructed of a still more refractory mass.
The door closing the charging opening of the furnace is
provided with loop-holes 1 inches wide which are closed
with glass, and the lower part of this door is also provided
with an aperture allowing the introduction of a rake. The
lowest point of the sump is provided with a tapping-hole,
but the working opening of the furnace also allows a ladling
out of the iron.
I
1
A furnace of this construction,* as used in the Royal
Foundry at Berlin, is shown in Figs. 180, 181, constructed
by Wöhlert. a is the sump; b, the aperture closed with a
FIG. 180.

b
FIG. 181.

α
*
Zeichnungen des Ver. Hütte, Jahrg., 1855, No. 15; 1854, No. 4; 1859,
No. 5; 1861, No. 15, r.
THE PROCESS IN REVERBERATORY FURNACES.
621
clay plate, containing the tapping-hole on the lower part, and
loop-hole on the upper part; c, the opening for charging
the furnace; d, flue conducting into a chimney.
The Process in Reverberatory Furnaces.
When possible, the pig-iron is previously heated and placed
in the red-hot furnace, especially when the furnace is new, thus
saving fuel and lessening the loss of iron by the formation
of slag. In order to effect a uniform melting, the smaller
pieces of iron as well as iron of easy fusibility are placed
lowest in the furnace, and the larger and more refractory
pieces of iron on the top of them. In furnaces with hearths
inclined towards the flue the pieces of iron must be placed
in such a manner as to enable the flame to pass freely through.
Larger pieces, which are less inclined to form slag than
smaller pieces, are placed near the fire-bridge upon a founda-
tion of fire bricks.
When charging the iron into the cold furnace but little air is
admitted at the commencement, in order to gradually warm
the hearth, and by producing a smoky flame to avoid the
oxidation of iron; the full heat is given after the contents
of the furnace have become red hot. During the melting
the working door is kept closed, and coals are thrown fre-
quently upon the grate, a few at a time, whilst the sliding
door before the grate is quickly closed again in order to
limit the admission of air. An iron rod is now introduced
into the furnace through an opening in the working door, by
means of which the unfused pieces are moved so as to form
a heap, or to bring them in contact with the fused iron so as
to cause them to melt. When fusing rapidly it is possible
also to fuse the iron scale which may have been formed in the
operation. In order to enable the temperature thoroughly
to permeate the fused mass, the slag formed must be thinly
liquid and not frothy; an addition of litharge (page 616)
will prevent frothiness.
When all the iron is melted, and is sufficiently liquid
and hot, it is tapped off, either into a sump and thence into
the mould, or into a vessel coated with refractory slag,
which is provided with either a spout or an opening at the
622
IRON.
bottom which may be closed by means of a lever. In rare
cases only is the iron ladled out whilst the chimney is closed.
Sometimes the iron is kept liquid for some hours in order to
purify it or to make it adapted for certain applications.
(rollers for instance). In order to ascertain the quality of
the iron, samples are ladled out and the quality ascertained
from their colour and texture, usually after cooling; stirring
renders the iron more uniform. When calculating the
quantity of iron to be melted, the loss of iron, the formation
of slag, the runners, &c., have to be taken into consideration.
One operation usually lasts from 2 to 6 hours, and the loss
of iron is from 6 to 9 per cent, and sometimes more, according
to the size of the castings, the length of the operation, the
quality of the iron and fuel, &c.
Two furnaces are frequently built side by side, so as to
form double furnaces, which are then put in operation at the
same time.
After having tapped the furnace, the iron sticking to
the hearth (schaalen) is removed, and the furnace is re-
charged whilst still hot, unless a renewal of the hearth be
required, which is the case at Mariazell, for instance, after
each operation.
The following are some examples of smelting in reverbe-
ratory furnaces with a sump on the flue :-—
At Mariazell cannons are cast from the reverberatory
furnace, and the furnace is charged with, on an average,
75 cwts. of suitable pig-iron in pieces weighing from 6 to
12 cwts. The melting is effected with fir-wood. The iron
which is placed upon fire bricks before the fire-bridge, is
melted in about 5 hours, and after taking samples it is
tapped off and conducted, by means of iron gutters coated
with loam, into cast-iron boxes, each with an aperture on
the bottom; from these boxes the iron is made to run
quickly into the vertical moulds. All the openings of the
furnace are then used for cooling, the debris in the furnace
is removed and the hearth renewed. The loss of iron is
* Oesterr. Ztschr., 1857, pp. 116, 188; 1858, p. 245.
GAS REVERBERATORY FURNACES.
623
8 per cent, and from 6 to 8 cubic feet of wood are consumed
per cwt. of resulting castings.
At Belgium* 490 kilos. of coal were consumed in melting
1000 kilos. of pig-iron when making several operations in
succession; 3.8 per cent of scale was formed, and 6'93 per
cent of slag. In 1861, the consumption in the foundry of
Gleiwitz per 1 cwt. of castings amounted to 116.6 lbs. of
pig-iron and 66'5 lbs. of coal; at Saynerhütte 88'97 lbs. of
coal were consumed in melting 100 lbs. of iron, and the loss
of iron amounted to 8:14 per cent.
At St. Gertraud charges of from 60 to 85 cwts. are melted
with brown coal; upon passing over the fire-bridge the gases.
are burnt by means of blast, which is introduced through
13 tuyeres at a pressure of 2 lines. 75 lbs. of coal are con-
sumed in melting 100 lbs. of iron.
In furnaces with the hearth inclined towards the fire-bridge
the charges of iron are uniformly distributed on the hearth
and melted as quickly as possible at a high temperature ;
the iron last remaining on the hearth in a solidified state
will be partly decarbonised, and is transferred to the sump
when all the iron is tapped off.
This quick melting causes a loss of iron of from 5 to
7 per cent only, and a saving of 25 per cent of fuel com-
pared with furnaces with hearths inclined towards the flue.
At Creuzott 6000 kilos. of iron (of grey iron and of
mottled iron) are melted in 5 hours with a consumption of
1050 kilos. of coal.
At Wöhlert's foundry in Berlin, 50 cwts. of pig-iron are
melted in 4½ hours, consuming from 27 to 36 lbs. of coal
per 100 lbs. of iron. Sometimes 150 cwts. of iron are charged
at once.‡
Gas Reverberatory Furnaces.
These furnaces|| are employed when intending to produce
castings of a certain degree of hardness and strength from
* B. u. h. Ztg., 1860, p. 17.
† Oesterr. Ztschr., 1857. p. 116.
B. u. h. Ztg., 1862, p. 32.
On the applicability to foundry purposes of iron refined in gas reverbera-
tory furnaces:-KARST., Arch., 2 R., xvii., 195; xx., 475 ; xxi., 512. Bgwkfd.,
xii., No. 1. B. u. h. Ztg., 1843, p. 611; 1846, No. 39; 1847, PP. 359, 814.
624
IRON.
grey coke pig-iron. The same kind of furnace provided
with a firing of gas is also employed for refining grey iron
for the puddling process, and one of these furnaces is shown
in Figs. 214 to 224 of this volume.
I
The hearth of the furnace is formed of a layer of sand
6 inches thick placed upon an iron plate, and it has a slight
inclination towards the tapping hole. A charge consists of
from 30 to 40 cwts. of pig-iron and is melted in about 3 hours
with a consumption of 33 cubic feet of coal per hour. The
liquid mass is frequently stirred up with a hook, thus freeing
the hearth from deposits; 5 lbs. of pounded limestone are
then uniformly spread out on the metal bath, and blast is
introduced by tuyeres at the long sides of the furnace. The
blast repels the thinly liquid slag and has an oxidising re-
action upon the iron. More limestone is added from time to
time (about 1 per cent of the pig-iron), the iron being frequently
stirred, and its quality ascertained by samples which are ladled
out. As soon as the samples show a lightish grey or slightly
mottled compact fracture, the blast pipe on the side of the
furnace containing the tapping hole is removed, and the iron
tapped off, when part of the slag runs out along with the iron.
The iron is employed either direct for castings, or it is tapped
off into sand moulds and re-melted in admixture with
other kinds of iron of inferior quality (Price and Nicholson's
method).* After the removal of the iron the tapping hole
is closed, first with some small coal and then with sand,
the hearth is repaired, the tuyeres are cleaned, and the
furnace is re-charged. All the slags are thoroughly re-
moved from the hearth when the furnace has been used a
fortnight.
When employing this furnace for the production of iron for
the puddling process, the process is carried on until the
samples show a pure white radiated fracture; this occurs in
from 3 to 5 hours, and causes a loss of from 5 to 9 per cent
of iron, and sometimes even more. The consumption of coal
amount to 1 cwts. of coal per cwt. of refined iron.
In 1861, the consumption at the foundry in Gleiwitz
*DINGL., Bd. 141, p. 368.
MOULDING AND CASTING.
625
amounted to 115'7 lbs. of pig-iron and 106 cubic feet
= 33°4 lbs. of coal per 100 lbs. of refined iron, the loss of
iron amounted to 13.6 per cent. At Königshütte the loss
was 10'37 per cent, the consumption of coal 1'46 cubic feet,
and that of limestone, 1'06 lbs.
2. MOULDING AND CASTING.
The moulder's art consists in forming in suitable materials,
which will be described later on, a hollow space (the
mould) having the shape of the desired casting, and in
then filling the mould with liquid iron. The latter opera-
tion is effected either by running the liquid iron direct from
the furnace into the moulds by means of slightly inclined
gutters formed of sand or of iron coated with sand, or the
iron is tapped into ladles of different construction, which are
lifted by hand when their weight is moderate, but which are
moved about by cranes if they are heavy.
tance.
Materials used in Moulding.-The materials applied to
the formation of different castings must possess certain pro-
perties, and a judicious selection is therefore of great impor-
Good materials for mouldings must be infusible at
the temperature of the liquid pig-iron, and must not evolve
gaseous substances (carbonic acid, sulphur compounds, &c.) ;
they must also be sufficiently fine and plastic, and sufficiently
adhesive when rammed together, that the mould may retain
its form in spite of the pressure of the melted iron.
These materials are chiefly the following:-
1. Green Sand.-This is a quartz sand containing a small
amount of clay, and in an undried state, capable of pre-
serving the forms of the objects impressed upon it against
the pressure of the melted iron. The sand must also be so
porous as to allow the escape of the air and the gases formed.
This sand is either found in native beds (decayed sandstone,
for instance,) or it is artificially prepared by mixing sand
rich in clay with sand very poor in clay.
The excellent Berlin sands for moulding are found in the
brown coal formation, or in the diluvial formation of the
VOL. II.
* B. u. h. Ztg., 1862, p. 4.
2 S
626
IRON.
Mark, in the coal measures of Wettin, &c., and they are
partly imported from England. A small quantity of oxide of
iron is not injurious to the quality of the sand, like a large
amount of lime, which causes caking, and an evolution of
gas.
Karmarsch, Bennigsen-Förder, and Reinhardt† have
lately ascertained that the quality of the sand for moulding
depends less on its chemical composition than on its
physical properties, namely, whether the grains are round,
angular, scaly, &c., and whether they are of uniform size.
Therefore the examination of sand for moulding requires a
microscopic investigation, and an elutriation in order to
ascertain the proportion of the coarse, moderately fine, and
fine particles it contains, besides the determination of its
chemical composition. According to Karmarsch's investi-
gations, the amount of clay in sand alone does not constitute
its adhesiveness; this being also increased by acute angled sand
grains and by a mixture of smaller and larger grains never
exceeding a pin's head in size. According to Reinhardt, a
good sand for moulding must consist to the naked eye of
uniform grains; when rubbed between the fingers it must
feel sharp to the touch, yet without the character of the sand
used by builders. When strewn upon dark paper it must
be free from dust, and when moistened with from 10 to
20 per cent of water, it must be capable of being formed into
balls without becoming pulpy, or being too easily crushed.
At Mariazell, iron blast furnace slags of the regular process
in a pulverised state, and mixed with clay, have been used
as sand for moulding, and according to Gjers,|| puddling and
finery slags also ; but these materials are too fusible.
+
The preparation of the sand for moulding consists in
crushing the grains to the size required; this is effected by
hand, by grinding,§ by pounding, or by rolling; the sand is
then dried and sifted.
* Mittheil. d. Hannov. Gew. Ver., 1862, Hft. 4.
B. u. h. Ztg., 1862, p. 4.
Oesterr. Ztschr., 1858, p. 236.
Polyt. Centr., 1863, p. 619.
§ Ibid., 1858, p. 244.
MOULDING AND CASTING.
627
The following is the composition of some varieties of this
sand:
Rothehütte
(Hartz), by
Streng.*
Königshütte
(Hartz), by
Ilsenburg
(Hartz), by
Bierworth.+
Spiller. +
Sheffield
by
Sauerwein.[]
SiO3
I0'00
II'99
79'02
86.68
A120
2.II
3
2.78
13'72
9'23
Fe₂O₂
2'53
3'77
2'40
3'42
CaO
trace
0*73
o'96
MgO.
0'71
KO,NaO
trace
4°58
но
2.68
1'73
Quartz
81.61
76'01
Birmingham, Lüneburg,
Manchester,
Naples,
by
Sauerwein.
by
by
by
Sauerwein.
Kumpmann. §
Jwanow.
SiO3
87.6
90*25
92'913
77-83
Al₂0₁.
7.7
4'10
5.850
7.7—8.2
Fe₂O₁
3
3.6
5'51
I'249
2.8-4.0
CaO.
o'96
0°23
MgO.
0*38—0*73
KO,NaO
3*17—5°78
ΚΟ
2*40—3'40
CaOCO₂
0*80—10*2
The following statements concerning the sand of this
country are extracted from Mackenzie's "Imperial Journal
of the Arts and Science," Division i., p. 63:-
"The sand of the London basin is the finest in the
country. It is universally employed in the manufacture of
fine goods, such as grates, fenders, and the like. The sand
in the neighbourhood of Falkirk is coarser and opener in the
pores, which unfits it for such work. It is employed for
casting hollow ware, pots and kettles for example, as the
enclosed air escapes freely through the inside body of sand
in the moulding of these articles. It affords a beautiful
smooth skin to the castings from Scotch iron, so remarkable
* B. u. h. Ztg., 1857, p. 198.
† Ibid., 1859, p. 60.
PERCY'S Metallurgy, i., 239.
|| B. u. h. Ztg., 1863, p. 138.
§ Ann. d. Min., 4 ser., viii., p. 689.
¶Oesterr. Ztschr., 1852, p. 403.
2 S 2
628
IRON.
in the hollow goods of the Carron Iron Works in Stirling-
shire, and of the Phoenix Iron Works at Glasgow.
The Belfast sand is finer than that from Falkirk, and is
principally used for fine machinery castings; it is sometimes
used for facing the moulds of ornamental work to give a finer
surface. It is also excellent for hollow moulding when mixed
with the Falkirk sand, but it is too expensive for general
adoption in that way. It is a mixture of a very fine adhesive
sand and an opener kind. Rocks and, the débris of abraded
rock, and free sand from the sea-shore, are employed for
making cores. Rock sand by itself does very well for short
cores, which open into the sand of the mouldings at both
ends, as it contains a proportion of clay in its composition,
which gives it cohesion; but it requires to be moderated with
free sand to make it opener for the better escape of the air
in its pores when used for cores of considerable length,
which, of course, are surrounded on all sides by the iron,
except at the small portions at the extremities through
which alone the air can find exit. Free sand is also used
alone for such cores, but as it is wanting in adhesiveness, it
requires to be tempered with clay, water, barm, or the refuse
of pease-meal. In the use of the last, accuracy is required
in proportioning it; clay water is used in ordinary cases,
and barm only in very particular cases.
2. Dry Sand (fetter sand, masse) is cither found in nature
or it is composed artificially of clay or loam and fresh rock
sand. It is firmer and better adapted than green sand for
moulding pipes, columns, shafts, and other long bodies of a
cylindrical form. It is named dry sand in contradistinction
to green sand, because, after being moulded, it must be dried
by heat to fit it for use, whereas green sand is employed as
it comes from its native bed, new and damp. Sometimes
even moulds formed of the dry sand are sharply burnt.*
3. Clay, when mixed with sand, is also much employed
for loam moulding. The loam is dried, pounded, sifted,
and moistened with water, and kneaded into a uniform
paste. The loam is then mixed with cow-hair, horse-dung,
DINGL., Bd. 131, p. 432.
MOULDING AND FOUNDING.
629
saw-dust, or straw, to render the loam moulds porous, and
also partly to give tenacity.
4. Metallic Moulds are employed for superficially hard-
ening the castings by a quicker cooling. The castings
will be harder the thicker the iron sides of the moulds (from
3 to 8 inches). The moulds are warmed previous to use,
they are brushed inside with graphite or coal tar, and they
must be protected against cracking by iron hoops. Bental*
makes cold air or water circulate in the hollow side of the
moulds. Holmest employs superheated steam for warming
the moulds; and Passet‡ employs for chilling the castings
superheated steam instead of the expensive iron moulds.
5. Blacking and Coal Dust|| are employed to resist the
penetrating action of the iron on the sand. Blacking is
simply charred oak wood ground to powder. For this pur-
pose oak charcoal is superior to all other ordinary wood
charcoals, as it is the heaviest.
According to Mushet's experiments, oak produces 22.6
per cent, that is fully one-fifth of its weight of charcoal. Were
the iron allowed to come into direct contact with the sand
of the mould, it would enter its minute interstices, and thus
yield but a rough surface. To avoid this, blacking is dusted
over the surface of the mould, pressed down on it, and
smoothed, in the case of green sand castings, but it is mixed
with clay water for covering loam mouldings. Its inflam-
mability is its essential property as a protector of the sand.
All combustible solid substances peculiarly resist liquid iron;
this may be exemplified by pouring it over a smooth surface
of wood. It rolls about in as lively a manner as mercury on
account of the continued effusion of gaseous matter by the
combustion of the wood throwing off the iron from the sur-
face. In cases of heavy castings in green sand, when the action
becomes too powerful for the blacking, this is assisted by
coal dust, which is mixed uniformly in the sand. It never
forms more than one-tenth of the sand in bulk, and the best
* DINGL., Bd. 131, p. 434.
B. u. h. Ztg., 1859, p. 410.
+
Ibid., 1857, p. 50.
|| MACKENZIE's Imperial Journal of Arts, &c., Division i., p. 63.
630
IRON.
kind of coal for the purpose is the rich hard splint coal.
Rouy recommends potato starch instead of blacking for
coating the mouldings.
For this purpose the coal is comminuted either by grinding
or pounding, or by rotation in an iron cask movable round
its axis.t
Patterns and Cores.
When iron moulds are not employed, the hollow spaces or
moulds are formed either by pressing the pattern into the
moulding material and removing it afterwards, or the workman
forms the mould without a pattern. Patterns are employed
when many castings of the same kind are to be produced.
When moulding without patterns, mechanical contrivances.
are sometimes employed (for instance, De Louvrié's and
Jackson's machine‡ for moulding large cog-wheels); machines
are also employed for lifting larger patterns out of the sand
(De Bergue's machine,|| Jobson's machine.§
Many castings have recesses or holes passing quite through
them; these holes must be provided for in the moulds by
suitable fillings or cores.
The patterns must be so constructed as to allow their
removal without damaging the mould; they must be dry and
of a smooth surface, and of the proper dimensions so as to
yield castings of the desired dimensions and weight. When
constructing the patterns, the contraction of cast-iron (page
291) must be taken into consideration.
According to Karmarsch,¶ the probable weight of an iron
casting can be calculated from the weight of the pattern by
multiplying its weight by 14 if it is made of pine-wood; patterns
of oak must be multiplied by 9, patterns of beech by 9'7,
patterns of birch by 13°4, patterns of alder-wood by 13.8,
of brass by o'84, of zinc by 1, of tin by o'89, of lead by o'64,
and of cast-iron by o'97. These ratios of multiplication
* Oesterr. Ztschr., 1855, p. 119.
† B. u. h. Ztg., 1847, p. 230.
LEOB., Jahrb., 1857, vi., 302.
Ibid., 1861, vi. 303.
B. u. h. Ztg., 1857, p. 127.
Mittheil. d. Hannov. Gew.-Ver., 1854, Hft. i., 38.
PATTERNS AND CORES.
631
p. m
n
have also been determined at Veckerhagen and found to be
somewhat larger than at Karmarsch. The weight of a casting
may also be calculated by the formula ", in which m ex-
presses the weight of the pattern, n the specific gravity of
the substance from which the pattern is made, and p is the
specific gravity of the substance of the casting.
The patterns are usually made of wood; alder, pine,
chestnut, plum tree, and maple being the kinds principally
used, as they contract but slightly in their cross-section.
Mahogany does excellently for small patterns, but its
expense limits its application. To prevent their bending, the
patterns are frequently formed of a number of pieces of wood
in such a manner that the pieces are joined alternately, one
piece with the fibre running lengthwise and the next piece
crosswise. These patterns are either painted with a shellac
and lamp-black varnish or soaked in oil to preserve them.
These wooden patterns must be kept in dry and mode-
rately warm places, and though cheap they do not admit of
the formation of exact and thin castings.
Of the metals, cast-iron is most used for patterns; brass,
copper, zinc, and alloys of lead and tin are also used. These
metallic patterns yield castings of accurate dimensions, and
are employed when thin and sharply cut castings are required,
or many from the same pattern.
Patterns of gypsum are occasionally employed, but chiefly
for reproducing wax patterns; when a direct use is made of
the gypsum patterns they are previously soaked in a varnish
of linseed oil.
Wax patterns, covered with fine clay, are used in moulding
complicated castings; they are formed either of one piece or
of separate pieces, and are afterwards carefully melted out of
the mould.
Cores for forming the holes or recesses in castings are
made either of sand (more or less argillaceous) or of loam,
by hand or in different kinds of boxes.* Some of the smaller
cores are made of loam over wooden bars wrapped round
with straw; these bars are taken out when the cores are
Polyt. Centr., 1859, No. 9.
632
IRON.
blackened and dried. Larger cores are usually formed over
hollow iron bars with vent holes distributed over their sur-
face, and provided at the ends with pivots. These core-bars
are next wrapped up in straw ropes; the inside of the cores
is sometimes formed of bricks. A wire is placed in the loam.
alongside the core bar, and afterwards removed, thus forming
a channel for the escape of air. In order to hold the cores
securely in their positions in the mould their ends project
into corresponding holes in the sand, and are there fixed.
These holes are formed by projections made in the pattern,
named core-prints.
The hollows in castings are sometimes formed by a suit-
able construction of the patterns, without employing cores,
in the casting of railway chairs for instance.
Apparatus, Contrivances, Tools, and Implements
used in Moulding.
1. Pits. A foundry should contain several cylindrical
pits, usually from 5 to 7 yards deep, placed near the fur-
naces. These are usually lined with brickwork and left full
of moulding sand. They are emptied in order to receive
large moulds, care being taken that the top is always below
the orifice from which the melted metal is tapped. These
pits are sometimes lined with iron plates instead of with
brickwork; when brickwork is adopted the pit walls are
sometimes provided with small fire-places. Very large pits
are sometimes placed outside the foundry, and if the ground
is marshy the brickwork lining is placed on an inverted
cupola of brickwork.
After having placed the moulds in the pit they are gene-
rally rammed in on all sides with sand, and heavy weights
are placed on the top to prevent their being lifted by the liquid
iron. When the mould is thus arranged nothing of it is to be
seen but the apertures for introducing the iron and the pipes
for egress of the air and gases. The iron must be conducted
through the channels in a continuous current with a certain
*
HARTM., Fortschr., iv., 177.
+ HARTM., Eisengiesserei, 1863, p. 180.
PITS.
633
velocity, so that it shall neither chill nor injure the mould;
care must be taken to prevent slag from entering the mould
together with the iron. Upon casting, lighted straw is
placed over the mouths of the air channels or wind pipes,
in order to ignite the escaping gases, otherwise they accumu-
late in the moulds. Larger castings are provided with
several holes in order to fill all parts of the mould as quickly
as possible.
If the runners are too small, the iron is liable to chill too
quickly, and if they are too large an excess of air is carried
into the mould along with the iron; in the latter case the
runners can only with difficulty be broken from the castings,
and they sometimes require to be separated by turning.
The iron is either introduced into the mould in such a
manner that it fills the mould from above, or it is conducted
into the mould from below by means of a tube on the side
of the mould; the tube and mould communicate below by
means of a channel. This mode of filling the moulds is
employed principally when castings of great density are re-
quired, and when it is feared that the iron running into the
mould from above will break off parts of the mould and
carry them away with it.
with it. Also the runner must always be
placed at a higher level than the mould; the liquid iron con-
tained in the runner will then fill the hollow spaces which
are formed in the mould by the solidification of the iron. The
hydrostatic pressure produced by a high runner contributes.
much to the soundness and solidity of the casting. Those
parts of the casting requiring the greatest density (the
chambers of cannons for instance) must be placed lowest
when casting them, and if this should not be feasible, with
large pans for instance, the moulds must be provided with
a high runner or a feeding head.
Cannons are occasionally cast direct from blast furnaces.
(Sweden), but more frequently from reverberatory furnaces
(Mariazell,+ Liège, &c.); their moulds are formed of argil-
TUNNER, das Eisenhüttenwesen in Schweden, 1858, p. 36.
† Oesterr. Ztschr., 1857. p. 269.
+
‡ DINGL., Bd., 129, p. 313. KARST., Arch., 2 R., xxv., 682.
634
IRON.
laceous sand (Mariazell) or of loam (Sweden). The iron
employed must be particularly strong.* At Malapane† and
Mariazellt rollers are cast in iron moulds, and in Belgium, ||
France, and Banat|| in loam moulds. The casting is
effected direct from the blast furnaces at Malapane and
Gmünd, the blast furnace process being so modified as
to produce a suitable iron; or the casting is made from
reverberatory and cupola furnaces.
2. Drying Chambers.-These are used for drying the
moulds and cores, and are usually low vaulted chambers
provided with sliding doors. Larger moulds are usually
moved into the chambers by means of cars running on rails.
These drying chambers are heated either by means of little
stoves of iron or clay placed outside them, or by tubes cir-
culating in them, and sometimes by open fires on grates
inside the chambers; the waste heat of other furnaces** is
occasionally employed for the purpose. These chambers must
always be provided with small chimneys for the escape of
the moisture; the chimney openings in the stoves are placed
a little below the floor.
Foundries for casting larger pieces must also be provided
with travelling or stationary cranes, or, better still, with
both. A stationary crane consists of an upright shaft,
embraced at the top by a collar, and revolving below upon
a pivot next a horizontal beam, stretching from nearly the
top of the shaft and having an oblique stay running to the
shaft. The horizontal beam supports a movable carriage
to which is suspended the tackle for raising the weights.
This carriage is made to slide backwards and forwards along
the beam by means of a simple rack and pinion movement,
the long handle of which descends within reach of the work-
man's hand.
*
Polyt. Centr., 1857, p. 664.
Berggeist, 1860, No. 99.
† LEOB., Jahrb., 1861, x., 490.
|| Ibid., 487.
§ Oesterr. Ztschr., 1861, No. 36.
¶ Berggeist, 1860, No. 99.
** B. u. h. Ztg., 1860, pp. 313, 371.
MOULDING IN GREEN SAND.
635
The chief requisites of the moulder are the cases or boxes
for containing the sand moulds. These are of different
sizes and shapes, and are made of wood or cast-iron, and
recently also of rolled iron.
Methods of Moulding.
The methods of moulding may be classified into moulding
in green sand, moulding in baked sand, and moulding in
loam; the moulding of such articles as statues, ornaments,
works of art, &c., forms a special branch, and, lastly, the
casting in iron moulds may be considered as a separate
department.
1. Moulding in Green Sand. This method is the easiest
and cheapest, and is employed when the quality of the cast-
iron and the appearance of the casting will allow it. As
the green sand must always be somewhat damp that it may
not be broken up by the liquid iron, the surface of the iron
chills and becomes hard, so that the castings are difficult to
treat with the drill or file. Grey pig-iron produced from a
refractory ore mixture in furnaces with high narrow hearths
is the least liable to chill.
This method is applicable when moulds are not too large
so that they resist the pressure of the liquid iron, and with
moulds not containing too many cores or projecting ornaments
liable to break off, and also to castings not requiring very
great softness.
Boxes are used or not according to the shape of the casting
to be moulded, and in the latter case the patterns are simply
impressed in the sand covering the floor of the foundry. The
sand for this purpose must be moistened and mixed with
powdered charcoal, coke, or coal, and the patterns must
be simple, chiefly flat, provided with ornaments on one side
only or perhaps with holes. The pattern must be placed
perfectly horizontal in the sand by the assistance of rule
and level; the runner for introducing the iron is then formed,
the borders of sand are wetted with water, holes are pierced
in the sand below the pattern, and the pattern is lifted out.
The mould is then dusted over with charcoal powder, and
636
IRON.
smoothed, and after placing the pattern once more in the
mould, iron is run into it.
To prevent the casting from warping, it is kept covered with
coal dust and sand until sufficiently cool; sometimes it is
weighted with iron plates to keep it straight.
The casting of pigs, &c., is analogous to this method of
moulding.
When plates provided with ornaments on both sides are
to be moulded the pattern is first moulded in the manner
already described, but in an inclined position, a box with
lattice-work sides filled with argillaceous sand, in which the
ornaments of the upper side of the plate are moulded, is
then placed on the mould on the floor, thus completing the
mould for casting the plate in question.
When moulding cog-wheels in this manner, baked cores
are employed to form the teeth, as cores of loose sand are
liable to be destroyed by the liquid iron when running in the
mould.
Some articles, such as anvils, hammers, &c., require to be
hardened on their surface, which is effected by introducing
into the desired portions of the mould pieces of iron blackened
with coal.
Moulding in boxes is employed for objects of very different
shapes, which do not allow the mould to be simply lifted
out as in the first method.
In Ure's "Dictionary of Arts," &c., vol. ii., 390, the
moulding in boxes is described as follows:-
"A couple of iron frames form a case or box which serves
as an envelope to the mould. Such boxes constitute an
essential and very expensive part of the furniture of a foundry.
It is a rectangular frame without bottom or lid, whose two
largest sides are united by a series of cross bars, parallel to
each other, and placed from 6 to 8 inches apart.
"The two halves of the box carry ears corresponding exactly
with one another, of which one set is pierced with holes,
but the other has points which enter truly into these holes,
and may be made fast in them by cross pins or wedges, so
that the pair becomes one solid body. Within this frame there
is abundance of room for containing the pattern of the piece
MOULDING IN GREEN SAND.
637
to be moulded with its encasing sand, which being rammed
into the frame is retained by friction against the lateral faces
and cross bars of the mould.
"When a mould is to be formed, a box of suitable dimensions.
is taken asunder, and each half, No. I and No. 2, is laid
upon the floor of the foundry. Green sand is thrown with
a shovel into No. 1, so as to fill it, when it is gently pressed
in with a rammer. The object of this operation is to form
a plane surface upon which to lay in the pattern with a
slight degree of pressure, varying with its shape. No. I
being covered with sand, the frame No. 2 is laid upon it so
as to form the box. No. 2 being now filled carefully with
the green sand the box is inverted, so as to place No. 1 upper-
most, which is then detached and lifted off in a truly vertical
position, carrying with it the body of sand formed at the
commencement of the operation. The pattern remains im-
bedded in the sand of No. 2 which has been exactly moulded
upon a great portion of its surface. The moulder condenses
the sand in the parts nearest to the pattern by sprinkling a
little water upon it and trimming the ill-shaped parts with
small iron trowels of different kinds. He then dusts a little
well-dried finely-sifted sand over all the visible surface of
the pattern and of the sand surrounding it; this is done to
prevent adhesion when he replaces the frame No. 1.
'He next destroys the preparatory smooth bed or area
formed in the frame, covers the pattern with green sand, re-
places the frame I upon 2 to reproduce the box, and proceeds.
to fill and ram No. 1 as he has previously done No. 2. The
object of this operation is to obtain very exactly a concavity
in the frame No. 1 having the shape of the part of the model
impressed coarsely upon the surface formed at the beginning,
and which was meant merely to support the pattern and the
sand sprinkled over it till it got imbedded in No. 2.
"The two frames in their last position, along with their
sand, may be compared to a box, of which No. I is the lid,
and whose interior is adjusted exactly upon the enclosed
pattern.
"If we open this box, and after taking out the pattern close
its two halves again, then pour upon it melted metal till it
638
IRON.
fills every void space and becomes solid, we shall obviously
attain the wished for end, and produce a piece of cast-iron
similar to the pattern. But many precautions must still be
taken before we can hit this point. We must first lead
through the mass of sand in frame No. 1 one or more channels
for the introduction of the melted metal, and though one may
suffice for this purpose, another must be made for letting the
air escape. The metal is run in by several orifices at once
when the piece has a considerable surface but little thick-
ness, so that it may reach the remotest points sufficiently
hot and liquid.
"The parts of the mould near the pattern must likewise be
pierced with small holes by means of wires traversing the
whole body of the sand, in order to render the mould more
porous, and to facilitate the escape of the air and the gases;
for this reason, also, the sand close to the pattern must
not be rammed too hard. Then, before lifting off the frame
No. 1, we must tap the pattern slightly, otherwise the sand
enclosing it would stick to it in several points and the operation
would not succeed. These gentle jolts are given by means
of one or more pieces of iron wire which have been screwed
vertically into the pattern before finally ramming the sand
into the frame No. 1, or which enter merely into holes in
the pattern. These pieces are sufficiently long to pass out
through the sand when the box is filled, and the horizontal
blows of the hammer are given upon their upper ends, their
force being regulated by the weight and magnitude of the
pattern. These rods are then removed by drawing them
straight out, after which the frame No. I may be lifted off
smoothly from the pattern.
"The pattern itself is taken out by lifting it in all its parts
at once, by means of screw pins adjusted at the moment.
For large pieces this manoeuvre is almost always executed
by several men, who, while they lift the pattern with one
hand, strike it with the other with small repeated blows to
detach the sand entirely, in which it is generally more
engaged than it was in that of the frame, No. 1. But in
spite of all these precautions there are always some degra-
dations in one or other of the two parts of the mould, which
MOULDING IN BAKED SAND.
639
are immediately repaired by the workman with damp sand,
which he applies and presses gently with his trowel, so as to
restore the injured forms.
"Hitherto it has been supposed that all the sand rammed
into the box is of one kind, but, from economy, the green
sand is used only to form the portion of the mould next the
pattern, in a stratum of about an inch thick; the rest of the
surrounding space is filled with the sand of the floor, which
has been used in former castings. In this case the interior
layer round the pattern is called new sand.
The pattern is sometimes too complex to be taken out
without damaging the mould by two frames alone; then
three or more are mutually adjusted to form the box.
"When the mould, divided into two or more parts, has been
properly repaired, its interior surface must be dusted over
with finely powdered charcoal tied up in a small linen bag,
which is shaken by hand. The charcoal is thus sifted at
the moment of application, and sticks to the whole sur-
face, which has been previously damped a little. It is after-
wards smoothed with a fine trowel. Sometimes, in order to
avoid using too much charcoal, the surfaces are finally dusted
over with finely pulverised sand, from a bag like the charcoal
was. The two frames are now replaced with great exactness,
made fast together by the ears with wedged bolts laid truly
level or at the requisite slope, and loaded with considerable
weights. When the casting is large the charcoal and the
fine sand dusting are suppressed. Everything is now ready
for the introduction of the fused metal."
2. Moulding in Baked or Used Sand.-This method of
moulding is employed for larger castings when green sand
would not resist the pressure of the liquid iron, and also for
small castings with ornaments, where a sharper copying of
their outline is required. The mechanical part of this pro-
cess is the same as that of the last process, with the difference
that the sand is rammed perfectly hard into the iron frames.
When the mould is finished it is transported to the drying
stove, where it may remain from 12 to at most 24 hours,
until it is perfectly dry. The sand is then said to be baked
or annealed. The experienced moulder knows how to mix.
640
IRON.
the different sands placed at his disposal, so that the mass
of the mould as it comes out of the stove may preserve its form
and be sufficiently porous. Such moulds allow the gases to
pass through them much more readily than those made of
green sand, and the castings they turn out are usually less vesi-
cular and are smoother upon the surface. The dried moulds
are brushed with a blacking consisting of pulverised coal,
wheaten flour, and glue water boiled together; barm is some-
times used instead of glue water. A mixture of clay water
and pulverised coal is also now and then employed.
This method of moulding is employed for large cylinders,
long tubes, cannons, medals, ornaments, &c.
Mr. Robert Mallet, F.R.S., has kindly allowed the following
extracts to be made from his highly elaborate treatise "On
the Physical Conditions involved in the Construction of
Artillery."*
His researches concerning the casting of guns are of
general interest and involve important considerations on the
constitution of cast-iron, and the influence of the shape and
size of the castings upon the physical properties of the iron.
Concerning the molecular constitution of crystalline bodies
(cast-iron included), Mr. Mallet states:
"It is a law of the molecular aggregation of crystalline solids
that when their particles consolidate under the influence of heat
in motion, their crystals arrange and group themselves with their
principal axes in lines perpendicular to the cooling or heating
surfaces of the solid; that is, in the lines of direction of the heat
wave in motion, which is the direction of least pressure within the
And this is true, whether in the case of heat passing
from a previously fused solid in the act of cooling and crystal-
lising on consolidation, or of a solid not having a crystalline
structure but capable of assuming one upon its temperature
being sufficiently raised, by heat applied to its external sur-
faces, and so passing into it.
mass.
For example, if an ingot of sulphur, antimony, bismuth,
zinc, hard white cast-iron, or other crystallisable metal or
atomic alloy; or even any binary or other compound salt or
* London: Longmans & Co., 1856.
MOLECULAR CONSTITUTION OF CAST-IRON.
6.41
haloid body, as sulphide of antimony, calomel, sal ammoniac,
various salts of baryta and lime, chloride of silver or of lead;
or even certain organic compounds, such as camphor and
spermaceti ;-provided only it be capable of aggregating in
a crystalline form under the influence of change of tempera-
ture, as from fusion or sublimation-if an ingot or mass of
any such body be broken when cold, the principal axes of the
crystals will always be found arranged in lines perpendicular to
the bounding planes of the mass; that is to say, in the lines of
direction in which the wave of heat has passed outwards from the
mass in the act of consolidation.
Now cast-iron is one of those crystallising bodies which
in consolidating also obeys this law more or less perfectly,
according to the conditions; so that generally it may be
enunciated as a fact that in castings the planes of crystallisation
group themselves perpendicularly to the surfaces of external
contour; that is to say, in the directions in which the heat of
the fluid cast-iron has passed outwards from the body in
cooling and solidifying.
Because the crystals of cast-iron are always small, and
are never well pronounced, these directions are seldom very
apparent to the eye, but they are not the less real.
Their development depends :--
Ist. Upon the character of the cast-iron itself, whether
or not it contains a large quantity of chemically uncombined.
carbon (suspended graphite), which Karsten has shown to
be the case with all cast-irons that present a coarse, large-
grained, sub-crystalline, dark, and graphitic, or shining
spangled fracture; such irons form in castings of equal size
the largest crystals.
2nd. Upon the size or mass of the casting, the largest
casting presenting, for any given variety of cast-iron, the
largest and coarsest aggregation of crystals, but by no means
the most regular arrangement of them, which depends chiefly
upon-
3rd. The rate at which the mass of casting has been
cooled, and the regularity with which heat has been carried
off by conduction from its surface to those of the mould
adjacent to them, and hence it is, that of all castings in iron
VOL. II.
2 T
642
IRON.
those called 'chilled,' that is to say, those in which the fluid
iron is cast into a nearly cold and very thick mould of cast-
iron whose high conducting power rapidly carries off the
heat, present the most complete and perfect development of
the crystalline structure perpendicular to the chilled surface
of the casting. In such the crystals are often found pene-
trating an inch and a half or more into the substance of the
metal, clear and well-defined.
These prevailing directions of crystalline arrangement may
be made more clear to the eye by Figs. 182 to 191.
Figs. 182 and 183 are sections of a round and a square bar
FIG. 182.

ዘረ .
of any of the crystalline solids, or of cast-iron when the
crystallisation is well developed (the circumstances affecting
which we shall consider further on). In the round bar the
crystals are all radiated from the centre; in the square bar
they are arranged perpendicularly to the four sides, and
FIG. 183.

hence have four lines (in the diagonals of the square) in
which the terminal planes of the crystals abut or interlock,
and about which the crystallisation is always confused and
irregular.
In Fig. 184 a flat plate is shown in section. The direc-
tions of the crystalline axes follow the law of Fig. 183, with
an extension in one direction.
MOLECULAR CONSTITUTION OF CAST-IRON.
643
In Fig. 185 a section is shown of a hollow cylinder, in
which, as in the case of Fig. 183, the arrangement of the
crystals is always towards the centre, or axis of the cylinder.
This figure applies to every cast-iron hollow cylinder,
whether water pipe, gun, mortar, &c.
FIG. 184.

V
Fig. 186 represents a portion of the lower or closed end of
the cylinder of the hydraulic press as first made for the pur-
pose of raising the tubes of the Britannia bridge, and which
broke in the attempt, the end of the cylinder having broken
out from the sides in the form of a flat frustrum of a cone,
FIG. 185.

as in Fig. 187, under the severe water pressure to which it
was exposed; that is to say, the fracture took place all round,
along the plane of junction of the conterminous crystals
formed perpendicular to the external and internal surfaces
of the bottom and of the sides of the cylinder, proving that
such planes of junction, where, as in Figs. 183 and 184, the
crystals join and interlace confusedly, are planes of weakness-
planes in which the cohesion of the metal is less and less,
for this reason, than in any other parts of the mass. These
lines of weakness extend from v to v throughout all the figures.
The form of the bottom of this cylinder was changed by Mr.
Stephenson from a distinct appreciation of the fact that the
!
2 T 2
644
IRON.
fracture of the part was in some way connected with the
sharp and sudden termination, square to the axis of the
cylinder, though without apparently any clear conception of
the crystalline laws upon which the fact depended, and a
FIG. 186.

V
new cylinder with a sort of semiovoidal end was made, a
section of a portion of which is represented in Fig. 188.
This stood the strain uninjured. Here the principal axes of
the crystals all are directed, as in Figs. 182 and 185, to the
centre. They, therefore, gradually change their direction,
and no planes of weakness are produced.
FIG. 187.

Concerning the fracture in the base of the cylinder of the
Britannia bridge hydraulic press, Mr. Mallet states in a note
that the direction of fracture indicated in the engraving is a
little erroneous. The fracture, striking outwards from the
neighbourhood of the internal angle, made by the base with
the sides of the cylinder, passed outwards (as in the figure)
at first nearly at 45° to the line of the sides, but gradually
curved upwards, and cut through the outer surface of the
MOLECULAR CONSTITUTION OF CAST-IRON.
645
cylinder in some places, round the circumference, rather
above than through the external salient angle formed by the
meeting of the exterior of the base with the sides, thus de-
parting towards the outside more or less from the plane of
FIG. 188.

W
weakness. At first sight this appears to militate against
the views of the former statement as to the existence here
of such a plane of weakness, as, wherever was the weakest
plane, the fracture should have followed it quite through;
but a more careful consideration of the question than was
before given will show that the facts, thus corrected, point
the opposite way, and perfectly sustain the views advanced.
The fracture was a diagonal one, tending generally from the
internal angle outwards; but if there were no plane of weak-
ness here at all-if the metal were of the same cohesion per
square inch throughout all its parts-the weakest place must
have been that of the least section of metal in the direction
exposed to pressure, and as this is in a plane at right angles
to the axis of the cylinder, the sides of the latter would in
such case have been torn directly across somewhere; the
cross section of fracture then being less in total area
than in case of a fracture from the internal to the external
angle at the cylinder's base in the ratio of 1:
2
√2
The diagonal must therefore have been the weakest place.
Why did it not break straight through it? The reason is
obvious, when we come to consider the nature of the fluid
forces to which it was exposed before fracture.
646
IRON.
The normal or radial pressures against the interior of the
curved sides of the cylinder and against the base at right
angles of itself and to the former, commenced a rent at the
interior angle, a certain amount of flexure, however small,
being produced in the metals at both sides of it. This flexure,
however slight and instantaneous, had necessarily the same
effect as if the fracture took place by rotation round con-
secutive points whose loci were in circles all round the outer
edge of the progressive fracture, and as the greater motion
was in the base which was projected off, so the fracture
curved upwards, just as the fracture in burst guns turns off
to one side, very near the outer surface.
The irregularity of broken surfaces, and of the line of rup-
ture, with reference to a plane parallel to the base, was, no
doubt, due to irregularities in the casting itself or other acci-
dental conditions.
It is to be hoped that these illustrations have served to
make clear the general law as applied to cast-iron artillery—
that every abrupt change in the form of the exterior, every salient,
and every re-entering angle, no matter how small, upon the
exterior of the gun or mortar is attended with an equally sudden
change in the arrangement of the crystals of the metal, and that
every such change is accompanied with one or more planes of
weakness in the mass.
Figs. 189 and 190 are sections of portions of a large cast-
iron gun. The former, part of the breech, through the 'vent-
field' square to the axis of the bore; the latter, a section near
a trunnion, also square to the axis. In Fig. 191, a section.
of a reinforce ring in the plane of the axis. In all of these
are shown, in an exaggerated form, the directions of crys-
talline aggregations and the planes of weakness resulting
from it.
It will be remarked that the square projection of the 'vent-
field' produces at each angle planes of weakness, which, in
the case of the re-entering angles, penetrate deep into the
thickness of the gun, and that these planes really do exist
is evidenced by the lines of fracture in burst guns, which
almost always follow along the angle at the sides of the
vent-field;' so also in the case of the trunnions, Fig. 190.
(
MOLECULAR CONSTITUTION OF CAST-IRON.
647
A gun, like every other body that fails under strain, must
fail in the weakest place, and the places of fracture and
positions of these planes of weakness most remarkably coin-
cide. The conclusion, therefore, seems inevitable, that,
FIG. 189.
FIG. 190.

V
V V
V
however incapable the unaided eye may be to discern any
difference in the crystalline arrangement of one part of the
gun more than of another, such planes of weakness do exist.
in the positions and from the causes here pointed out.
FIG. 191.

The external forms of cannon have been greatly modified
and simplified in modern times from the complex forms of
remoter periods; but even still, in the plainest forms of guns,
648
IRON.
such as Sir William Congreve's and Monk's patterns, &c.,
mouldings, astragals, reinforces, &c., are still adhered to,
and from the unwillingness to give up altogether antiquated
forms, originally adopted and continued in ignorance, we
have the folly still to cling to making numerous and useless
sharp angles and corners, and sudden changes of form and
of dimension on the exterior of all our ordnance, and so
prolong in the most needless way one cause of their weak-
ness. That gun, however plain externally, will look best to
the really educated eye that most fully conforms to the laws.
upon which its perfection as an instrument depends."
Mr. Mallet has given the following passage concerning the
physical conditions induced in moulding and casting:
"It is known to every practical iron-founder upon a large
scale that, generally, the larger the mass of the casting he
makes with any given quality of cast-iron the coarser is the
grain, that is, the larger are the crystals that develop them-
selves in the mass. The same metal that shall produce a
fracture, bright grey, mottled, and without a crystal visible
even to a single lens, in a bar, cast, say two inches diameter,
shall, if cast into a cylinder of two feet in diameter, produce
a dark confusedly crystalline surface of fracture, as coarse as
granite rock.
To meet this, the practice is to prescribe for material for
large castings a certain proportion or mixture of small close-
grained scrap metal with the pig-iron of whatever best
quality may be denoted. The remedy fails-as fail that
always must which is founded upon a misconception of the
laws of the phenomena; as well might small seeds be sown
to produce small trees. The scrap is no sooner re-cast into
the large mass than it resumes the large crystalline grain.
The experiments of Mr. Fairbairn (Trans. Brit. Ass., 1853)
on the repeated melting of the same cast-iron, by casting
into inch square bars, are concluded by him to prove that
the grain of the metal and the physical qualities of the
casting improve by some function of the number of meltings,
and he fixes on the thirteenth melting as that of the greatest
strength.
PHYSICAL CONDITIONS INDUCED IN CASTING.
649
Some most important conditions modifying, if not invali-
dating, such a conclusion, and more especially the effects of
the variable mass of the casting, seem, however, wholly to
have escaped him. Indeed, these experiments, rightly con-
sidered, only prove what was well known before.
Experiments of the same character as those of Mr. Fair-
bairn have been made in the United States upon the larger
scale of casting guns at various periods. These experiments
on the effects of re-melting, or of prolonged continuance of
fusion, are of the same inconclusive character, and the few
deductions made are sometimes anomalous and inconsistent.
It is known that white cast-iron (No. 4 pig) has a far
higher ultimate cohesion than any of the grey, mottled, or
dark grey varieties (Nos. 1, 2, and 3 pig). It is known, also,
that the latter may be converted more or less perfectly into
the former by fusion in direct contact with fuel and blast, or
in reverberatory furnaces.
Now it follows as a matter of course from these well-known
facts that, as perfectly white cast-iron has at once the highest
cohesion and the greatest brittleness, while properties the
reverse belong to the darkest grey graphitic cast-iron, some
mixture of the two qualities must give the best material for
gun-founding, or for any other mechanical purpose in which
the highest product of tenacity and toughness is demanded,
and in this consists the value of mottled iron for cannon.
It is also obvious that a more or less perfect approach to
such a mixture may be made by repeated melting and cooling
up to a certain point of any grey iron; but the number of
meltings and coolings necessary to effect this will differ, not
only with the original grey iron tried, but with the conditions.
of the cupola furnace in every consecutive melting, and with
the conditions of cooling at every casting, so that probably
no two series of experiments could possibly be made that
should give co-ordinate results, or that would be applicable
to any other make of iron, or to any other cupola, fuel, and
blast. Moreover, the quantity of graphite eliminated at each
cooling is greater in some proportion as the cooling is
more rapid. The trial, therefore, that shall give the number
of meltings producing the best result for castings of one
650
IRON.
dimension cannot be true or applicable to castings of any
greater or less scantling.
Thus, if Mr. Fairbairn concludes the thirteenth melting
gives the best metal for trial bars cast one inch square, with
the original 'make' of pig-iron and mode of melting and
casting he employed, it does not follow that for bars of two
inches square it would be so; or that, with the one-inch
bars, but a different original 'make' of pig-iron, or a different
cupola, it would be so; or even with the same pig-iron and
conditions of melting, but a different mode of moulding and
casting the same one-inch bars, the result should be alike.
If, with the very same pig-iron, cupola, and fuel, the meltings
be performed with a surcharge of metal and flux in propor-
tion to fuel, and an excess of blast, the one-inch square bars,
when cast, would have been found to have arrived at their
assumed best quality, perhaps at the fourth in place of the
thirteenth melting. If the bars themselves had been cast in
'chills' in place of sand-moulds, so as to have been cooled
as fast as possible, the point would have been still sooner
reached, and if cast in dry sand-moulds' or in 'loam,'
would have been reached later.
<
Upon sample bars so small as one inch square, even a little
more or less wetting of the sand of the green-sand mould on
the part of the moulder, would have made the most formid-
able difference as to the rate of progress towards white iron.
Finally, if, instead of bars of one inch square, the experi-
ments had been made upon a sufficient scale to admit of
casting bars of a foot square, these, when broken after the
thirteenth melting, in place of presenting the same assumed
improvement, would in the interior have presented very little
change in fracture from the original pig-iron (unless, indeed,
peculiar care had been taken so to work the cupola as to
burn out of it the graphite), and in place of the thirteenth
melting being the charmed one, it might not be reached at
the 13 × 13th melting.
Again, by some iron-founders, one 'make' or sort of pig-
iron is presumed to give a closer grain than another, and he
prefers it; and although this is to a certain extent true, i.e.,
that some cast-irons do under equal conditions produce
PHYSICAL CONDITIONS INDUCED IN CASTING. 651
rather smaller crystals than others, still this view equally
fails to attain the object of close-grained heavy castings.
But, furthermore, it is a fact familiar to iron-founders that of
several castings of the same form and mass, made at nearly
the same time from the same mixture of metal, and melted
in the same furnace, some will, when cold, have a much
more coarsely crystalline grain developed in them than
others.
Now, while the regularity of development of the crystals in
cast-iron depends, as we have already seen, upon the regu-
larity with which the melted mass cools, and the wave of
heat is transmitted from its interior to its surface, arranging
the crystals in the lines of least pressure in its transit, the
extent of development, or what is the same thing, the size of
each individual crystal, depends upon the length of time during
which the process of crystalline arrangement is going on,
that is to say, upon the length of time that the casting takes to
cool; hence, then, may be announced as a law that—
The size of crystals or coarseness of grain in castings of
iron depends for any given 'make' of iron and given mass
of casting upon-
Ist. The high temperature of the fluid iron above that
just necessary to its fusion, which influences
2nd. The time that the molten mass takes to cool down
and assume again the solid state.
These laws have recently received the most striking con-
firmation from some quite analogous researches "Upon the
Molecular Properties of Zinc," made by P. W. Bolley, and
published in the "Annalen der Chem. und Pharm.," xcv.,
p. 294.
The lower the temperature at which the fluid cast-iron is
poured into the mould, and the more rapidly the mass can
be cooled down to solidification, the, closer will be the grain
of the metal; the smaller its crystals, the fewer and least
injurious the planes of weakness, and the greater the specific
gravity of the casting, cæteris paribus.
The very lowest temperature at which the iron remains
liquid enough fully to fill every cavity of the mould without
652
IRON.
risk of defect, is that at which a large casting, such as a
heavy gun, ought to be 'poured.'
A certain amount of contraction on becoming solid from
the liquid state occurs in all castings. It is well known to
iron-founders that for cast-iron this is variable, and depends
on the the mass of casting, being greatest for small and
least for large castings of the same 'make' of iron, but it
is obvious, and it follows from M. Bolley's researches, that
the contraction will also be greater in proportion as the
metal is poured into the mould at a higher temperature,
although from the expansion in the act of crystallising, the
specific gravity of the solid mass may be less at the higher
than at the lower temperature of 'pouring.'
As, therefore, there are two conditions that principally
affect the degree of contraction-the total change of volume
between the liquid metal and its solid casting, namely, the
extent to which the fluid metal as entering the mould has
been expanded by elevation of temperature, and the state of
final aggregation of the crystalline particles, which we have
seen depends much upon the former-so there will be a deter-
minate amount of contraction due to a determinate thick-
ness or mass of casting, irrespective of, although related to,
the co-efficient of contraction for any particular 'make' of
iron, for there is no doubt that different makes, cæteris paribus,
contract somewhat differently. From whence it follows,
that different parts of the same casting if differing materially
in scantling or mass, will have different amounts of final
contraction, and hence-
Sudden changes of form or of dimensions in the parts of
cast-iron guns, besides the injury they do to the crystalline
structure of the mass, introduce violent strains, due to the
unequal contraction of the adjoining parts, whose final con-
traction has been different.
The amount of lineal contraction due to solidification of
cast-iron appears to vary with metal and circumstances of
casting, from 1-120th up to 1-90th of the dimensions of the
cold mass. Its contraction in volume, therefore (more than
three times this), and probably not equal in the directions of
three rectangular axes, owing to the crystalline structure, is
PHYSICAL CONDITIONS INDUCED IN CASTING. 653
so great, and the difference such between its measure for
large and small parts of the same casting, that the latter
should never be neglected.
The effects of this difference are well known to founders
by causing castings of certain forms to become distorted or
spontaneously broken after they have solidified. To multiply
instances would be tedious; but one instance requires.
remark, as proving that these internal strains occurring in
castings of variable bulk exist where little suspected, and
that it is with extreme slowness that the molecules after
consolidation appear gradually to assume minute changes of
arrangement, and to adjust themselves within certain limits
to a state of permanent equilibrium.
It is a fact well known to working mechanics engaged in
boring or turning or otherwise cutting into large castings of
iron that have cooled safely and without crack or flaw, that
when a part of the whole mass shall have been cut away,
as, for example, when a large and thick-flanged cylinder, or
a large toothed wheel, or other irregular discoid mass, is
'bored out,' the form of the exterior of the mass changes
during the operation. The portion cut away destroys the
temporary equilibrium that was established in the mass, and
it again changes its form and perhaps its symmetry, and
sometimes even its volume.
For some most valuable illustrations of the singular forms
or lines of direction which the curves of internal tension and
compression take in solids of various forms thus under elastic
constraint, Mr. Maxwell's paper in the "Transactions of the
Royal Society of Edinburgh," vol. xx., part 1, may be consulted.
Sometimes a casting which has cooled safely will fly to
pieces on receiving a sudden jar or blow of a trifling degree
of force, a fact which is in analogy with that observed by
Captain Parry in his earlier Arctic voyages, viz., that the
astronomical instruments exposed to extremely low tempe-
ratures for long periods and quite undisturbed did not con-
tract to their extreme point until after they had been
subjected to some slight jar or blow, when the metal of the
instrument suddenly became reduced in volume and its
dimensions again stationary.
654
IRON.
The extreme slowness, continuing sometimes for months,
with which these molecular changes take place, due to the
gradual adjustment of such internal strains, has been beauti-
fully shown in a memoir on the elastic properties of solids
(Ann. de Chem., vol. xli., p. 61) by M. Savart, who found
that plates of sulphur cast into flat discs continued to change
their state of molecular arrangement for long periods after
solidification.
It follows from this that old guns that have been long
bored and laid in store are likely to be more trustworthy
than those hastily cast, bored out, and brought into service;
and this seems to be supported in some degree by experience.
In general extension and support of the views advanced
as to castings in iron becoming endowed with variable power
of resistance, depending upon external form and mode of
casting, &c., the important memoir of M. Savart above
alluded to should be consulted. By refined and delicate
methods of investigation, founded upon sonorous vibrations
elicited, he has shown that numerous bodies, such as zinc,
lead, cast copper, glass, plaster of Paris, sealing wax, and
others, though possessing apparently a perfectly homoge-
neous structure, have it not; but, on the contrary, all possess
lines and planes usually crossing each other at right angles,
in which their resisting powers are enfeebled, and which he
has called axes of greatest and of least elasticity, and which he
attributes to the arrangement of their molecules assumed in
the process of cooling. The relations of these phenomena
to the conditions of cooling and external form of the body as
affecting these, however, do not appear to have been per-
ceived by M. Savart, but have been stated in a distinct form
for the first time by Mr. Mallet.
Besides the effects already referred to, due to the contrac-
tion of cast-iron in becoming solid, another class of abnormal
strains introduced by the consolidation of one portion of a
casting before another, must not be passed over, as often
producing results of the most important character in artillery.
This will be more readily understood by immediate reference
to example. When a large gun, or, still more, a large
mortar, is cast solid, and the metal cools in the ordinary way,
PHYSICAL CONDITIONS INDUCED IN CASTING.
655
the external portions solidify long before the interior has
ceased to be liquid, and the process of solidification is propa-
gated as it were in parallel'couches' from the outside to
the centre of the mass. The lineal contraction of any one
couche assumed of infinite thickness is in the direction of
its circumference directly proportionate to that circumference;
and so it would seem (at first) that the contraction of the
whole assemblage should be at every point proportionate to
its distance from the centre, and thus the solid, when all
cold, should be left in a state of molecular equilibrium. This
is not the case, however, for no sooner has the first couche
or thickness of solid crust formed on the exterior than it
forms a complete arch all round, so that the contraction
between fluidity and solidification of each subsequent couche
is accommodated (the continuity of the mass remaining un-
broken throughout) by portions of matter withdrawn radi-
ally from the interior towards the still cooling exterior:
that is to say, from a smaller towards a larger circumference.
The final effect of this, propagated to the centre of the mass,
is twofold:
Ist. To produce a violent state of internal tension in the
molecules of the metal, in radial lines from the axis of the
gun viewed as a cylinder, tending to tear away the external
portions of the mass from the internal nucleus; a force
which is zero at the axis and at the exterior, and a maximum
between and probably at a point of the radius somewhere
between R and from the exterior.
R
2
2nd. To produce about the centre or along the axis a line
of weakness, and one in which the texture of the metal is
soft, porous, of extremely low specific gravity, with coarse
and frequently distinctly separated crystals, and often, not-
withstanding the precautions of the founder in 'feeding'
the head of the casting, that is to say, in slowly adding
fresh quantities of hot and fluid metal while it is yet possible
to get it introduced into the centre of the solidifying mass,
leaving actual cavities in the centre of the casting.
In a casting 2 or 3 feet or more in diameter, it is not un-
usual (with the founder's best care) to find a central portion
of from 6 to 8 or more inches in diameter, consisting of a
636
IRON.
spongy mass of scarcely coherent crystals of cast-iron,
usually in arborescent masses made up of octahedral crystals,
the whole so loose that upon a newly cut section dark cavities
can be seen by the naked eye in all directions, out of which
single or grouped crystals can often be picked with the hand,
and so soft that a sharp pointed chisel of steel may be easily
driven several inches into the mass, as if into lead or soft
stone. It fortunately happens that in pieces of artillery a
large portion of this defective core of spongy metal is re-
FIG. 192.

&A Kh
V
V
V
V
V
V
V
moved in the process of boring out; but where the hollow
thus taken out does not extend very close back to the ex-
terior of the breech-in other words, where the thickness of
the breech in the line of the axis is considerable, a portion of
THE EFFECTS OF BULK AND FLUID PRESSURE.
657
the spongy uncompact metal is left remaining, and forms
the part of the gun at the bottom of the bore or chamber.
This is most remarkably the fact in large mortars.
Fig. 192 is a section in line of axis and plane of trunnions of
a 13-inch sea-service mortar with the head of metal remaining
attached, and the whole in the position in which the mortar
is usually cast, with the parts to be cut off and to be bored
out marked by a black line, exterior to which is the finished
mortar. The shaded central portions represent the weak
and porous parts of the metal about the axis, extending down,
it will be observed, below the bottom of the chamber, where
it leaves a soft spot, easily hammered and burnt away by the
shock and blaze of the powder. From the conditions of in-
ternal strain already explained, the exterior of the cylinder
is in a state of compression, and the interior in a state of
tension, a state precisely the reverse of that calculated to
give the metal its greatest power of resistance to internal
strains in the direction of the radius.
The effects of bulk and fluid pressure are stated thus :—
It is a remakable fact, though one not yet fully explained,
that a small bar cast on to or projecting from a casting in
iron of very large scantling, when afterwards broken off and
tested, will not sustain by a good deal as great a transverse
or longitudinal strain as the same sized bar or the same
metal cast alone (i.e., in an isolated mould), and under the
same "head" of metal. This circumstance, no doubt due
to the conjoint influence of several molecular conditions that
have been under discussion, appears to be in part due to the
extreme slowness with which the small bar cools in close
proximity to the large mass of which it forms an appendage.
Slow cooling developes a coarse uneven grain with large
but thoroughly irregular and confused crystallisation. Cast-
iron with such a grain is never strong or cohesive, though
perhaps soft and extensible. The more rapidly a casting
once consolidated can be cooled without introducing injurious
effects, the finer, closer, and more even will be its grain on
fracture, and with any given metal the greater will be its
strength. The rate of cooling cannot be accelerated beyond
a moderate limit. If this limit be exceeded, as by casting
VOL. II.
2 U
658
IRON.
in a cold, thick, highly conducting metallic mould, the iron is
'chilled.' It cannot be so fast as to endanger unequal con-
traction, nor must it be so fast in large castings, such as guns
requiring to be 'fed' from a ‘feeding head' with fresh portions
of fluid metal during consolidation to fill up the internal
cavities or porosity due to contraction and crystallisation, as
already explained, that this feeding cannot be accomplished.
The prevalent notion, however, that the soundest and
strongest castings are obtained by letting them cool slowly
in the moulds, is founded on a radical error.
The enormous time required by a large casting for cooling,
especially if left to cool in the mould, and hence jacketted
with its badly conducting material (clay and sand) is not
generally known. The hydraulic press cylinders for raising
the Britannia Bridge tubes, which were about 12 feet long
and about 3 feet in diameter, and weighed in the mould,
perhaps, 20 tons, were found red hot at the expiration of
72 hours after having been cast; they only became cold enough
to handle 10 days after having been stripped from the loam,
and required feeding for more than 6 hours after having been
poured. During the greater part of this time molecular
changes were going on, increasing the coarseness of the
crystalline grain of the metal and reducing its tenacity. It
would have been much better practice to have kept the
exterior of the 'loam' wet, and thus inducing cooling by
evaporation as soon as ever the setting' of the metal had
rendered it safe to do so.
The cooling must be uniform as far as uniformity is prac-
ticable. Uniformity, strictly, is impossible, in any casting;
the approach to it is most difficult in heavy solid castings, such
as guns and mortars, and hence the great advantage which
would result from a return to the ancient practice of casting
them hollow upon suitably made "cores," as admitting of
internal cooling by artificial means, such as a current of air
at the same time that the outside is cooling. It is understood
that the American Government requires its guns to be so
cast, and cools them by a current of water passed into the
interior-a practice of very doubtful advantage, as not under
sufficient control to insure avoidance of an evil greater than
THE EFFECTS OF FLUID PRESSURE.
659
that which it is proposed to remedy, namely, cooling the
interior of the gun much faster than the exterior.
Unequal cooling, especially if very rapid, involves all the
injury that violent internal wrenching and straining can do
to strength.
Mr. Mallet has made extended and careful experiments to
ascertain the relation between the head of fluid pressure and
the specific gravity of the casting, showing the great value
of increased head of metal in adding to the density and
strength of the castings.
These experiments were made upon cylindrical shafts of
cast-iron, cast vertically in dry sand moulds, and under heads
gradually increasing up to 14 feet in depth, and all poured
from "gates" at the bottom.
The increase of density in castings of large size, due to
their solidification under a head of metal varying from 2 to
14 feet in depth is shown below:-
No. of Depth of Casting
Experiments. in Inches.
First
Specific
Gravity.
Difference.
I.
О
6'9551
2.
24
6.9633
0.0082
3.
48
7°0145
0'0512
4.
72
7°0506
0.0361
5.
96
7'0642
0.0136
6.
120
7'0776
0°0134
7.
144
7'0907
0.0131
8.
168
7'1035
0*0128
I.
О
7'0479
2.
24
7*0576
0'0097
3.
48
7°0777
4.
72
7'0890
O'0201
0°0113
5.
96
7°1012
0˚0122
6.
120
7.1148
0.0136
7.
144
7*1288
0'0140
8.
168
7 1430
0'0142
I.
O
7'0328
2.
24
7'0417
3.
48
7'0558
0*0089
Ο'ΟΙ4Ι
4.
72
7*0669
5.
96
7'0789
Ο ΟΙΙΙ
O'0120
6.
120
7'0915
7.
144
71046
S.
168
71183
0.0126
O'0131
0*0137
Apedale, No. 2. Blaenavon, No.
Hot Blast.
Cold Blast.
No. 1. Hot Blast.
1. Calder Cast-iron,
2 U 2
660
IRON.
These experiments show an increase of density due to
14 feet head about equal to a pressure of 44.8 lbs. per square
inch on the casting, from 6·9951 to 7∙1035 for Scotch cast-
iron."
In the following table the decrease of specific gravity fol-
lowing increase of bulk is obtained; the iron castings are
made from the same sort of cast-iron, and under similar
circumstances :—
Mark of
Experiment.
First
Dimensions of
Casting.
Specific
Gravity.
Difference.
A.
5 × 5 × 0*25
7'0560
B.
5 × 5 × 0*50
7*0261
0'0299
C.
5X5 XI
7.0627
0'0366
D.
5 × 5 × 2
6.9856
0°0771
E.
5 × 5 × 4
6.9588
0.0268
A.
5 × 5 × 0*25
7°1449
B.
5 × 5 × 0.50
7.1464
0'0015
C.
5 X5 XI
7.1423
0*0041
D.
5 × 5 × 2
7'1153
0'0270
E.
5x5x4
7'0942
0'02II
A.
5×5X0*25
7.1876
B.
5 × 5 × '050
7°1735
0*0141
C.
5 × 5 × I
7.1164
0'0571
D.
5×5×2
7*0806
0'0358
E.
5 × 5 × 4
7'0483
0*0323
Apedale, No. 2.
Hot Blast.
Blaenavon, No. 1. Calder, No. 1
Cold Blast.
Hot Blast.
3. Moulding in Loam.-The introduction of dry-sand
moulding (moulding in baked sand) has to a great extent
limited the application of moulding in loam.
This mode of moulding differs from those before described
in not using any patterns, but forming the moulds by the
assistance of templets. The interior shape of the casting is
always formed first.
Loam moulding is chiefly employed to form cores for tubes
and hollow castings, and in such cases when making a
pattern would be too expensive, or when the size of the
casting is of such magnitude as to render too difficult the
MOULDING IN LOAM.
661
carriage of the moulding frames and cores, which is the case
with large pans, bells, large guns, &c.
In Ure's "Dictionary of Arts," ii., p. 391, the mode of
fabricating loam moulds is illustrated by forming a sugar
pan; it is described as follows:
Fig. 193 is the pan. There is laid upon the floor of the
foundry an annular platform of cast-iron, a, b, Fig. 194, and
FIG. 193.
FIG. 194.

upon its centre, c, rests the lower extremity of a vertical
shaft, adjusted so as to turn freely upon itself, whilst it
makes a wooden pattern, e, f, Fig. 195, describe a surface of
FIG. 195.
FIG. 196.
FIG. 197.



U
d
m
d P
revolution identical with the reversed internal surface of the
boiler intended to be made. The outline, e, g, of the pattern
is fashioned so as to describe the surface of the edge of the
vessel. Upon the part a, d, b, d, Fig. 195, of the flat cast-iron
ring, there must next be constructed with bricks laid either
flat or on their edges, and clay, a kind of dome, h, i, k, Fig.
195, from 2 to 4 inches thick, according to the size and
weight of the piece to be moulded. The external surface of
the brick dome ought to be everywhere at least 2 inches
distant from the surface described' by the arc, e, f. Before
building up the dome to the point, i, coals are to be placed.
within it upon the floor, which may be afterwards kindled
for drying the mould. The top is then formed, leaving at i,
round the upright shaft of revolution, only a very small
outlet. This aperture and some others left under the edges
662
IRON.
of the iron ring enable the moulder to light the fire when it
becomes necessary, and to graduate it so as to make it last
without more fuel, until the mould is quite finished and dry.
The combustion should always be extremely slow.
Over the brick dome a pasty layer of loam is applied and
rounded with the mould, e, g, f; this surface is then coated
with a much smoother loam by means of the concave edge
of the same mould. Upon the latter surface the inside of
the sugar pan is cast, the line, e, g, having traced in its
revolution the ledge, m. The fire is now kindled, and as the
surface of the mould becomes dry, it is painted over by a brush
with a mixture of water, charcoal powder, and a little clay,
in order to prevent adhesion between the surface already
dried and the coats of loam about to be applied to it. The
board, g, e, f, is now removed and replaced by another,
g', c', ƒ', Fig. 197, whose edge, e', f', describes the outer sur-
face of the pan. Over the surface, e, f, a layer of loam is
applied, which is turned and polished so as to produce the
surface of the revolution, e', f', as was done for the surface,
e, f, only in the latter case the line, e', g', of the board does
not form a new shoulder, but rubs slightly against m.
The layer of loam included between the two surfaces,
e, f, c', f', is an exact representation of the sugar pan. When
this layer is well dried by the heat of the interior fire it must
be painted like the former. The upright shaft is now re-
moved, leaving the small vent hole through which it passed
FIG. 198.

to promote the complete combustion of the coal. There
must be now laid horizontally upon the ears of the plat-
form, d, d, Fig. 196, another annular platform, p, q, like the
former, but a little larger, and without any cross-bar. The
MOULDING IN LOAM.
663
relative position of these two platforms is shown in Fig. 198.
Upon the surface, e', f', Fig. 197, a new layer of loam is laid,
2 inches thick, and the surface is smoothed by hand. Then
upon the platform, p, q, Fig. 198, a brick vault is constructed
whose inner surface is applied to the layer of loam. This
contracts a strong adherence with the bricks, which absorb
a part of its moisture, whilst the coat of paint spread over
the surface, e', f', prevents it from sticking to the preceding
layer of loam. The brick dome ought to be built solidly.
The whole mass is now to be thoroughly dried by the
continuance of the fire, the draught of which is supported
by a small vent left in the upper part of the new dome; and
when all is properly dry the two iron platforms are adjusted
to each other by pin points, and p, q, is lifted off, whilst care-
fully kept in a horizontal position. Upon this platform are
removed the last brick dome, and the layer of loam which
had been applied next to it and representing exactly on its.
inside the mould of the surface, e', f', that is, of the outside
of the pan.
The crust contained between c, ƒ and e', f' is
broken away, an operation easily done without injury to the
surface, e, f, which represents exactly by its inside the inner
surface of the pan, or only to the shoulder, m, corresponding
to the edge of the vessel. The top aperture through which
the upright shaft passed must now be closed; only the one
is kept open in the portion of the mould lifted off upon p, q ;
because, through this opening the melted metal is to be
poured in the process of casting. The two platforms being
replaced very exactly one above the other, by means of the
adjusting pin-points, the mould is completely formed and
ready for the reception of the metal.
When the object to be moulded presents more complicated
forms than the one now chosen for the sake of illustration, the
workman constructs his loam moulds by analogous processes,
but the modes of effecting many things which at first sight
appear scarcely possible must be left to his sagacity. Thus,
when the forms of the interior and exterior do not permit
the mould to be separated into two pieces, it is divided into
several, which are nicely fitted with adjusting pins. More
than two cast-iron rings or platforms are sometimes necessary.
664
IRON.
When oval or angular surfaces are to be traced instead
of those of revolution, no upright shaft is used, but wooden
or cast-iron guides made on purpose, along which the cut-
out pattern board is slid according to the drawing of the
object. Iron wires and claws are often interspersed through
the brickwork to give it cohesion. The core, kernel, or inner
mould of a hollow casting is frequently fitted in when the
outer shell is moulded.
The case of a gas retort, Fig. 199, will illustrate this
subject.
FIG. 199.

17:
d
E
The core of the retort ought to have the form e, e, e, e, and
be very solid, since it cannot be fixed in the outer mould for
the casting, except in the part standing out of the retort
towards m, 111. It must be modelled in loam, upon a piece
of cast-iron called a lantern, made expressly for the purpose.
The lantern is a cylinder or a truncated hollow cone of cast-
iron, about half an inch thick, and differently shaped for
every different core. The surface is perforated with holes
about half an inch in diameter. It is mounted by means of
iron crossbars, upon an iron axis which traverses it in the
direction of its length. Fig. 200 represents a horizontal
section through the axis of the core; g, h is the axis of the
FIG. 200.

lantern, figured at i, k, k, i; o, i, i, o is a kind of disc or dish,
perpendicular to the axis, open at i, i, forming one piece with
the lantern, whose circumference, 0, 0, presents a curve
similar to the section of the core, made at right angles to its
MOULDING IN LOAM.
665
axis.
We shall see presently the two uses for which this dish
is intended. The axis, g, h, is laid upon two gudgeons, and
handles are placed at each of its extremities to facilitate the
operation of making the core. A hay band as thick as the
finger is wound around the whole surface of the lantern, from
the point h, to the collet formed by the dish. Even two or
more coils may be applied, as occasion requires, over which
loam is spread to the exact form of the core, by applying
with the hand a board against the dish, o, o, with its edge
cut out to the desired shape; and also against another dish,
adjusted at the time towards h, while a rotatory movement
is given by means of the handles to the whole apparatus.
The hay interposed between the lantern and the loam,
which represents the crust of the core, aids the adhesion of
the clay to the cast-iron of the lantern and gives passage
to the holes in its surface for the air to escape through in
the casting.
When the core is finished and has been put into the drying
stove, the axis, g, h, is taken out; then the small opening
which it leaves at the point, h, is plugged with clay. This
is done by supporting the core by the edges of the dish, in a
vertical position. It is now ready to be introduced into the
hollow mould.
The mould, executed in baked sand, consists of three
pieces, two of which, absolutely similar, are represented
(Fig. 201) at p, q, the third is shown at r, s. The two similar
parts p, q, present each the longitudinal half of the nearly
cylindrical portion of the outer surface of the gas retort; so
that when they are brought together the cylinder is formed;
r, s contains in its cavity the kind of hemisphere which
forms the bottom of the retort. Hence, by adding this part
of the mould to the end of the two others, the resulting
apparatus represents, in its interior, the exact mould of the
outside of the retort; an empty cylindrical portion, t, t, whose
axis is the same as that of the cylinder, u, u, and whose sur-
face, if prolonged, would be everywhere distant from the
surface, u, u, by a quantity equal to the desired thickness of
the retort. The diameter of the cylinder, t, t, is precisely
equal to that of the core, which is slightly conical, in order
666
IRON.
that it may enter easily into this aperture, t, t, and close it,
exactly, when it is introduced into the collet or neck.
The three parts of the mould and the core being prepared
the two pieces, p, q, must first be united and supported in an
upright position; then the core must be let down into the
FIG. 201.
FIG. 202.


t
u
pe
14
9
17
S
opening, t, t, Fig. 202. When the plate or disc, o, o, of the
core is supported upon the mould it is evident that the end
of the core is everywhere equidistant from the edge of the
external surface, u, u, and that it does not go too far beyond
the line q, q. Should there be an inaccuracy it must be cor-
rected by slender iron slips placed under the edge of the disc,
o, o; then, by means of a cast-iron cross and screw bolts,
v, v, the core is fixed immovably. The whole apparatus is
now set down upon r, s, and by means of screw bolts the
plane surface, q, q, are fixed upon r, r, and the melted metal
is introduced by an aperture, z, which has been left at the
upper part of the mould.
When, instead of the illustration now selected, the core of
the piece to be cast has to project beyond the mould of the
external surface, as in the case of a pipe open at each end,
the process is more simple, because we may easily adjust and
fix the core by its two ends.
In casting a retort the metal is poured into the mould set
upright. It is important to maintain this position in the
two last examples of casting, for all the foreign matters
which may soil the metal during its flow, such as sand, char-
coal, gases, and scoriæ, being less dense, rise constantly to
the surface.
CASTING IN CHILLS.
667
Casting in Metal Moulds or Chills.*-These moulds
being good conductors of heat quickly cool the iron which
is poured into them and chill it on the surface; the moulds
are usually brushed with black-lead, tar, or lime. Mottled
pig-iron, and grey iron with an inclination to become white
are best adapted for this method of casting, as they combine
the qualities of hardness, toughness, and liquidity.
thicker the metal moulds the harder will be the castings.t
The
Hard rollers or chilled rollers, the gudgeons of which
are moulded in dry sand as they require to be soft, are cast
in a vertical position, whilst the iron is introduced from
below. Some sorts of cast-iron contract so much in metal
moulds as to cause the castings to crack on their surface,
owing to the expansion of the liquid iron which the casting
contains in its centre, whilst the surface has already solidified.
One attempt to avoid this disadvantage was made by casting
the roller round an iron core and by employing moulds of
more or less thickness.
Cast-iron containing phosphorus contracts but very slightly
and is thinly liquid, whilst iron containing sulphur is thickly
liquid. In order to obtain a clean surface of the casting
the iron must be introduced into the mould in a rotating
movement, which may be produced by joining the runners to
the mould in the direction of a tangent. To prevent the
mould from bursting it must be warmed before the fluid iron.
is introduced into it. The larger the roller to be cast the
more mottled must be the iron used.
In this country chilled castings have been brought to a
great perfection by Messrs. Ransome, of Ipswich. Their
chilled ploughshares and chilled railway chairs are cast in
moulds of such a construction that the melted iron comes
in contact with iron in those parts of the moulds where it is
required to be chilled.
*
B. u. h. Ztg., 1848, p. 521; 1853, p. 860. Bgwkfd., xiii., 440. DINGL.,
Bd. 127, P. 47.
+
B. u. h. Ztg., 1848, No. 1; 1853, No. 8.
KARST., Arch., 2 R., vii., 3; viii., 254. Bgwkfd., xviii., 549. Polyt. Centr.,
1837, No. 5. TUNNER, Jahrb., 1854, p. 254. Berggeist, 1860, No. 99. Oesterr.
Ztschr., 1855, p. 367; 1857, p. 188.
653
IRON.
With regard to improvements in moulding the attention
of founders has been turned-
Ist. To the methods by which the labour of making moulds
in sand may be reduced.
2nd. To the introduction of improvements in the mode
of constructing patterns and moulds; and
3rd. To the manufacture of metallic moulds for those
purposes for which they could be applied.
A great progress has been made during the last 20 years
in all these branches of the treatment of iron.
Machine Moulding.-In large works devoted to the pro-
duction of cast-iron pipes for the conveyance of water and
gas, machinery has been applied, so that the operation of pipe
moulding is performed almost without manual labour, with
great rapidity and precision. The cost of pipes at the present
time is only about £2 per ton above the value of pig-iron out of
which they are made—a very small sum when it is considered
that the iron has to be re-melted, an operation involving both
the expense of fuel and a loss of from 5 to 20 per cent of the
iron in the cupola. An ingenious machine for moulding in
sand, spur and bevel wheels of any pitch or diameter has
been employed in Lancashire, the advantage being that the
machine moulding-tool acts directly upon the sand without
the intervention of any pattern or mould. In any large foundry
there is an enormous accumulation of costly wheel-patterns,
taking up a great deal of space, and these can now be dis-
pensed with by substituting the wheel moulding machine.
Railway chairs are moulded in a machine, and ploughshares
which, although only weighing a few pounds each, are sold at
the low rate of £8 a ton, are moulded in the same manner.
Plate Casting.-Under the next class of improvements
the introduction of plate casting has been the most productive
of good results.
One great source of expense and trouble in a foundry is
the injury done to patterns and to their impressions in the
sand by the necessity under the ordinary system of moulding,
of striking the pattern, or pushing it first in one direction and
* Dr. URE's Dictionary of Arts, &c., vol. ii., p. 386.
FINISHING THE CASTINGS.
669
then in another in order to loosen it. Now the object of the
machinist is to construct all his spindles, bearings, bolts, and
wheels of specified sizes, and then to cast the framing of the
machine so accurately that the working parts may fit into the
frame without any manual labour. In order to effect this,
every projection and every aperture in the casting must be at
an exact distance, and this can only be obtained by employing
such a system as plate casting, where the pattern is firmly
attached to a plate, and it is impossible for the moulder to
distort or injure the impression. Plate casting has been long
known, but was practically confined for many years to the pro-
duction of small articles, such as cast nails and rivets.
Shell Casting.-A circular aperture is made in a horizontal
planed plate of iron two inches thick. Through this a sphere.
of iron of the same diameter as the aperture is pushed until
an exact hemisphere appears above the plate. The lower
flask is put on to the plate and sand filled in; the lever being
relieved the sphere falls by its own weight, the lower flask
is removed and the upper flask put on the plate; the sphere
is pushed through the plate as before, and sand filled in with
great rapidity and accuracy.
The sand cores for filling up that part of the shell which
is to be hollow are also carefully and quickly made at
Woolwich. The halves of the core-mould open and shut
with a lever, so that the bad plan of striking the core-mould
is avoided as completely as that of striking the pattern is in
the process of moulding shot and shell.
Finishing the Castings.
Some of the castings are ready for sale when taken from
the mould and after the ragged edges have been trimmed
with a hammer, and have been chipped by a chisel when
quite cold; but a number of castings require a further
treatment, either mechanical or chemical.
The mechanical treatment* of castings may consist of the
following manipulations:--
*
KARMARSCH, mechanische Technologie, i., 420, 493,
HARTMANN, Eisen-
giesserei, 1863, p. 510. Mitthlgn. des Hannov. Gewerbe-Vereins, 1835, Lief. 3,
P. 192.
670
IRON.
1. Some castings, such as smoothing irons, anvils, &c.,
are ground on grindstones, which are usually set in motion
by water power.
2. Shot, &c., are polished by putting them in a rotating
cask, in which they rub against each other.
3. Cannons, machine castings, &c., require to be bored,
turned, planed, &c.,
4. Many castings are blackened, which prevents their
oxidation and improves their appearance.
For this purpose large castings are heated to about 100° C.
and brushed with tar, and then allowed to dry, or the
castings are dipped into boiling tar, which is a quicker way.
Small castings are coated with an admixture of linseed oil
varnish, and lampblack; this coating is applied either at
the common temperature or after having previously heated
the castings.
Sometimes the castings are repeatedly smoked with burning
resinous wood and brushed, or they are brushed over with
linseed oil, heated, and brushed after the flame has dis-
appeared.
Upon dissolving ½ lb. of asphalt and lb. of resin in 4 lbs.
of linseed oil at a high temperature, a lac or varnish is ob-
tained, also well fitted for coating larger kinds of castings.
After cooling, the lac may be preserved in a bottle, and
when required for use some linseed oil must be added
to it.
In this country, castings brushed with linseed oil are sus-
pended 8 or 10 inches above a strongly smoking wood fire.
After an hour the castings are lowered, but so that they do
not come into contact with the glowing coals, and 15 minutes
later they are dipped into cold oil of turpentine. The coating
thus obtained is bright, resists oxidation, and even the
action of weak acids.
5. In some cases the castings have* a coating of oxide arti-
ficially formed on their surface, thus protecting them against
the influence of the atmosphere. This process is effected
DINGL., Bd. 96, p. 20.
TEMPERING.
671
either by brushing the castings with dilute nitric acid and
drying them in the open air, or by some other means (butter
of antimony); but this process is more applied to articles
made of wrought-iron and steel than to those of cast-iron.
Many proposals have been made for the protection of iron
against oxidation. A simple method consists of brushing
articles of iron or steel with a solution of wax in benzol, and
allowing the benzol to evaporate. The solution may be prepared
by putting small pieces of wax into cold benzol till saturation.
6. Welding Cast-Iron.-This process is used to correct
defects in castings. A hole is bored in the defective part
of the casting, which is then strongly heated. A wall of
sand is formed round the deficient part, and liquid iron
gradually poured into the hollow formed by the sand wall; the
liquid iron then dissolves the red-hot iron of the casting with
which it comes into contact; gudgeons of large rollers, &c.,
have been thus welded.†
The chemical treatment of castings comprises the following
operations :-
1. Tempering.-This operation is sometimes applied to
hard or brittle castings which are unfit for the mechanical
treatment of turning, boring, &c., and is performed by heating
the castings for from 12 to 20 hours in closed cast-iron
vessels, which are filled either with sand or a mixture of coal
dust and bone ash, thus rendering the iron softer (page 289).
For the purpose of heating, the vessels containing the
castings are placed on a railway running beside a grate in
brick vaulted chambers. Upon heating castings for several
days at a higher temperature and in contact with substances
evolving oxygen (brown stone, red iron ore, oxide of zinc),
they attain more or less malleability, like steel or wrought-
iron (malleable cast-iron) (page 278). The heating may be
effected in crucible furnaces|| or in reverberatory furnaces.§
*
DINGL., Bd. 82, p. 75; Bd. 94, p. 46. Bericht. der 13, Versammlung
deutschr. Archit., Hanover, 1863, p. 107.
+ Polyt. Centr., 1861, p. 1160.
Bgwkfd., x., 14, 39. DINGL., Bd. 65, p. 155.
|| Bgwkfd., ii., 93, 510; ix., 43, 142; xi., 760; xiii., 77. DINGL., Bd. 26, p.
315; Bd. 29, p. 150; Bd. 62, p. 376. Polyt. Centr., 1849, pp. 13, 34; 1853
P. 951.
§ DINGL., Bd. 102, p. 220.
672
IRON.
White pig-iron is better adapted for this purpose than grey
iron.
2. Bronzing.*-The castings are scoured till they become
bright; they are then brushed with a solution of copper-
vitriol and polished after being dried. The copper precipi-
tated by the iron gives it a bronze-like appearance.
Sometimes the galvanic method is employed, and now and
then the mechanical one of coating the castings with a var-
nish and with a green colour, upon which thin finely pul-
verised bronze colours are partially laid.
3. Coating Castings with other Metals.-a. Gilding+
cast-iron by means of gold amalgam is very difficult, as the
amalgam does not stick to the iron. It is therefore neces-
sary to brush the well cleaned surface of the iron with a
concentrated solution of copper vitriol, and to apply the
amalgam to the precipitated copper. As under certain cir-
cumstances the coating of copper is injurious, Böttger coats
the articles direct with mercury by means of the electro-
positive zinc in the following manner:-The article to be
gilded is well cleaned and boiled in a porcelain vessel together
with 12 parts of mercury, I part of zinc, 2 parts of iron vitriol,
I parts of muriatic acid of 1*2 specific gravity, and 12 parts
of water; in a short time a layer of mercury will deposit
upon the iron, and upon this the gold amalgam may be uni-
formly distributed.
The gilding may also be effected upon polished iron in the
following manner :-If a nearly neutral solution of chloride
of gold be mixed with sulphuric ether and agitated, the
ether will take up the gold, and float above the denser
liquid. When this auriferous ether is applied by a camel-
hair pencil to brightly polished iron or steel, the ether
evaporates and the gold adheres. It is fixed by polishing
with a burnisher. This gilding is not very rich or durable;
in fact, the affinity between gold and iron is feeble compared
to that between gold and copper or silver.
*DINGL., Bd. 47, p. 313; Bd. 52, p. 319. Bgwkfd., xiii., 559.
†DINGL., Bd. 99, p. 158.
BOTTGER'S Beiträge, &c., 1847, Hft. 3, p. 14. Polyt. Centr., viii., 384.
COPPERING CAST-IRON.
673
Gilding of cast-iron by the galvanic way* is also difficult,
and is successful only if the article is perfectly clean. It is
advisable previously to coat the article with copper or silver.
Polished iron may also be gilded with heat by gold leaf.
b. Silveringt cast-iron. Iron to be silvered is first pro-
vided with a coating of copper, upon which the silver is
applied either by means of amalgam or silver leaf.
Cast-iron can be well silvered by the galvanic way‡ without
a previous coppering.
c. Coppering.-Cast-iron is easily coppered by simply im-
mersing it in a solution of copper vitriol, but the coating of
copper thus produced does not adhere to the iron. The copper
will adhere to the iron when employing the galvanic current,
chiefly when the cast-iron has been previously coppered or im-
mersed in a solution of cyanide of potassium and copper.
The great advantages which would arise from the perfection
of a plan by which iron could be coated with copper at a cheap
rate induced Messrs. Elsner and Philip, of Berlin, to undertake
a series of experiments to ascertain if the coppering could
not be effected more economically than by employing cyanide
of potassium, and in this they have been successful. To
coat iron the article must be well cleaned in rain or soft
water and rubbed before immersion in the solution, which
may be either chloride of potassium or chloride of sodium con-
taining a little caustic ammonia added, or tartrate of potash,
with a small portion of carbonate of potash. At the ex-
tremity of the wire, in connection with the copper or nega-
tive pole of the battery, is fixed a thin flattened copper plate,
and the article to be coated is attached to the wire from the
zinc or positive pole, and both are then immersed in the
solution, the copper plate only partially. The liquid should
be kept at a temperature of from 15° to 20° C., and the suc-
cess of the operation depends on the strength and uniformity
of the galvanic current. When the chlorides are employed,
the coating is of a dark, natural copper colour, and with
*
1856.
ELSNER, die galvanische Vergoldung und Versilberung, 3 Aufl., Leipzig,
† KARMARSCH, mechan. Technol., i., 477-
ERDM., J. f. pr. Chem., xxix., 264.
+
VOL. II.
2 X
674
IRON.
tartrate of potash it assumes a red tinge, similar to the red
oxide of copper. When sufficiently covered, the article is
rubbed in saw-dust, and exposed in a current of warm air to
dry, when the metal will take a fine polish and resist all
atmospheric influence.
d. Tinning.-A coating of tin is frequently applied to cer-
tain kinds of castings, chiefly to cooking utensils, thus pre-
venting them from rusting, and also preserving the food to
be cooked from taking a black tinge. The tin applied must
be free from lead, or the food is liable to become poisoned.
The articles to be tinned are first tempered (method of
Weinberger and Girardt), then turned in a lathe, or other-
wise well cleaned, and washed with dilute muriatic acid of
8° or 10° B., or with sulphuric acid. Delesse and Thomas‡
mix the acid with some organic substances, and Sorel || mixes
it with copper and tin salts. The iron pots are now dried,§
and heated up to the melting point of tin; the fluid tin is
then rubbed either with a cork or a ball of cotton on the
bright surface of the iron to be tinned. Too low a tempera-
ture of the pots causes too thick a coating, and too high a
temperature prevents the tin from adhering to the iron.
Sal-ammoniac or chloride of zinc and ammonium¶ is em-
ployed in the operation to keep the surface of the metal free
from oxidation.
The tinning of iron by Mr. Morris Stirling's patent process
is described in Ure's "Dictionary," vol. iii., p. 925, thus:-
"For this purpose the sheet, plate, or other form of iron pre-
viously coated with zinc, either by dipping or by depositing
from solutions of zinc, is taken, and after cleaning the sur-
face by washing in acids or otherwise, so as to remove any
oxide or foreign matter which would interfere with the per-
fect and equal adhesion of the more fusible metal with which
it is to be coated, it is dipped into melted tin, or any suitable
alloy thereof, in a perfectly liquid state, the surface of which
*B. u. h. Ztg., 1857, p. 305.
† Polyt. Centr., 1851, p. 1267.
DINGL., Bd. 107, p. 446; Bd. 111, p. 271.
| Ibid., Bd. 112, p. 121.
Trockenvorrichtung, Bgwkfd., x., 118.
DINGL., Bd. 106, p. 73.
ENAMELLING CAST-IRON.
675
is covered with any suitable material, such as fatty or oily
matters, or the chloride of zinc, so as to keep the surface of
the metal free from oxidation, and such dipping is to be con-
ducted in a like manner to the process of making tin plate
or of coating iron with zinc."
Iron may be tinned also by the galvanic current (method
of Roseleur and Bouchy).*
Tinned iron articles which are deficient in tin oxidise
more rapidly than iron without any tin coating, owing to a
galvanic reaction caused by the contact of tin and iron. A
coating of zinc, on the other hand, more effectually protects
the iron from oxidation, even if this coating is only partial,
but as zinc is readily dissolved in acids, salt brine, &c., iron
vessels coated with it cannot be used for cooking purposes.
Iron may also be coated with zinc by the galvanic
current.t
4. Enamelling for cast-iron hollow wares, for saucepans,
&c. Cooking vessels are frequently enamelled to prevent
them from oxidation and the action of weak acids; but the
uniform and durable coating of such vessels with an enamel
perfect in lustre, colour, hardness, &c., can only be performed
with difficulty.
This difficulty arises from the following circumstances :—
a. The enamel sticks less to iron rich in graphite than to
iron free from it, therefore strongly mottled or white iron
produced from bog-ores is best fitted for enamelling; this
white iron requires a mere preparatory scouring, whilst grey or
mottled iron must be submitted to a corroding and scouring in
order to remove the graphite from the surface of the vessels.
b. As cast-iron has a different rate of expansion (0'0011445)
when heated from o° to 100° C. to enamel (0'00080833), the
enamel cracks and flies off when heating the enamelled vessels,
if the enamel has been applied to the vessels as a single vitreous
coating. For this reason iron vessels'are first coated with a
ERDM., J. f. pr. Ch., lxv., 250.
† Polyt. Centr., 1861, p. 76.
VOGELSANG, Lehrbuch der Eisenmaillerkunst, 1851. KARMARSCH und
HEEREN, technisches Wörterbuch der Gewerbskunde, 1854, i., 719. KARMARSCH,
mechan. Technologie, i., 483. DINGL., Bd., 133, p. 256. Berggeist, 1861, p. 460.
2 X 2
676
IRON.
mass which frits upon heating and sticks to the iron, and
upon this surface the enamel is melted. The first coating is
somewhat porous, and contracts and expands along with the
iron, and thus preserves the enamel from flying off.
c. The proper temperature for fixing the first coating to
the iron. The mass applied must neither melt nor remain in
its original state; this operation is of the greatest impor-
tance and can only be properly understood after long practice;
in many places it is still considered a secret.
d. The composition and preparation of the enamel, which
is also mostly kept secret, and only to be ascertained by ex-
periments.
The mass for the first coating consists chiefly of silica,
boracic acid, alumina, and alkalies; the enamel, or second
coating, contains the same components but a larger amount
of alkali, which renders the mass more easily fusible, and also
some oxide of tin, rendering it opaque and white. Concerning
their chemical composition both mixtures are variable within
certain limits. Some enamel also contains lead; it then may
be fixed to the iron at a lower temperature, and, therefore,
more cheaply, and it is also more durable; but it is less
white in appearance and it is liable to be injurious when used
for cooking vessels.
Eulenberg ascertained this enamel to contain-
Silica.
•
Oxide of lead
Phosphoric acid
Phosphate of lime
43 38
39°12
3'51
2.61
The production of enamels free from lead has lately been
aimed at.
The following are the chief materials used in the prepara-
tion of the enamel :-
a. Silica, in the form of quartz, flint, or elutriated sand.
The quartz or flint is ignited, quenched in water, broken,
sifted, and ground.
b. Felspar, containing on an average—
•
Silica
Alumina
Potash.
65'4
18.0
16.6
ENAMELLING CAST-IRON.
677
and is chiefly effective for the purpose on account of the
amount of silica and alkali it contains. It is treated like the
quartz, and sometimes digested with muriatic acid if con-
taining peroxide of iron.
c. Pieces of Glass, containing on an average—
Silica.
Alumina.
Soda.
Lime.
53
70
14
3
12
are sometimes used as substitutes for silica or felspar.
d. Pieces of Porcelain, of the average composition-
Silica.
Alumina.
Potash
Lime.
are used like pieces of glass.
e. China Clay, containing on an average-
60
33
4
3
Silica.
Alumina.
Water
47
37
16
This must be elutriated, dried, and sifted. It is used as
an admixture in the mills for grinding the mass for the first
coating, thus rendering it more pasty and opaque after
burning. When forming the second coating, China clay
enters into the composition of the mass.
f. Borax, containing-
Soda.
Boracic acid
Water
16.37
36°53
47'10
The boracic acid is sometimes replaced by phosphoric acid,
and lately also by the cheaper arsenious acid.
g. Calcined Soda, containing 58.69 parts of soda and
41°31 parts of carbonic acid; it is usually used in the mass
for the second coating only.
h. Saltpetre, containing 46°59 parts of potash and 53°41
parts of nitric acid, used for the enamel only.
i. Magnesia, as basic carbonate, obtained by decomposing
sulphate of magnesia by means of carbonate of ammonia,
4MgO,CO₂+ MgO,HO,
containing 50.85 parts of magnesia, 41°59 parts of carbonic
678
IRON.
acid, and 456 parts of water. It is used in both coatings;
in the first coating merely as a mechanical admixture, and
in the second coating as a mechanical admixture and also
as part of the composition of the melted mass. Magnesia
occasions a better fixing of the masses to the iron. Sulphate
of magnesia, whose sulphuric acid is expelled by the silica
upon heating, is chiefly used in admixture when grinding the
enamel, as it facilitates the fritting of the mass and its adhe-
sion to the iron; it is not applied when melting the compo-
sition for forming the mass. Chalk, fluor spar, and heavy
spar are used in rare cases only.
k. Carbonate of Ammonia is sometimes used as an
admixture of the mass when grinding, to render it pasty and
to prevent a cracking of the enamel.
1. Litharge and White Lead are used when producing
plumbiferous enamels.
m. Oxide of Tin is produced by heating pure tin in a
clay vessel up to its melting point, and continually removing
the oxide formed from the surface; a strong draught is liable
to carry oxide away. The oxide of tin is a yellowish or
greyish white powder, and must be heated several times
with an admission of air, in order more highly to oxidise the
dark suboxide of tin which may be present; if necessary, an
addition of saltpetre is given. A little saltpetre contained
in the enamel thus has a decolourising reaction. The pure,
ground, and washed oxide of tin is a white or yellowish white
powder, becoming yellow upon heating.
When it is wished to produce a plumbiferous oxide of tin,
an alloy, consisting of four parts of tin and one part of lead,
is heated in an iron pan covered with a muffle, which is pro-
vided with a draught hole; the formed crust of oxide is then
continually removed into an iron box, allowing the metal
which may have been removed along with the oxide to settle
to the bottom.
Oxide of zinc and bone-ash are sometimes used.
The great number of recipes* given for the preparation
* ERDM., J. f. pr. Chem., xiii., 12. DINGL., Bd. 78, p. 40; Bd. 79, p. 3.
Bgwkfd., vii., 208. Polyt. Centr., 1840, p. 326; 1847, p. 488; 1854, p. 1378;
1855, p. 700. KARST., Eisenhüttenkunde, iii., § 846.
ENAMELLING CAST-IRON.
679
of enamel do not differ from each other very essentially, and,
according to them, the mass for the first coating contains
from 65'92 to 77'05 per cent of silica; the enamel for the
second coating from 25'59 to 43'71 per cent of silica.
The following may be relied upon-
Mixture for the First Coating.-No. I contains 50 parts
of quartz, 30 parts of borax, and I part of magnesia as com-
ponents to be melted; the melted mass is then ground with
an admixture of 15 parts of quartz and 13 parts of clay;
thus containing-
Silica.
Boracic acid
Alumina.
Potash
Magnesia
77'05
11.87
5'21
5°44
0'55
For the preparation of the same mass, No. 2, 30 parts of
quartz, 30 parts of felspar, and 25 parts of borax are melted and
ground together with an addition of 6 parts of felspar, 1.25 parts
of magnesia, and 10*75 parts of clay; thus containing—
Silica.
Boracic acid
Alumina.
Potash
Soda.
Magnesia
65'92
10*30
11*76
6.72
4'60
0'71
The enamel for the second coating is prepared in the
following manner :—
No. 1.—10˚5 parts of glass, 9 parts of porcelain, 13.5 parts
of borax, 4°5 parts of saltpetre, 3 parts of magnesia, 1'5 parts
of sulphate of magnesia, and 9 parts of oxide of tin are
melted together, and ground with an addition of 11'85 parts
of quartz, 178 parts of magnesia, and 3'95 parts of oxide
of tin; thus containing-
Silica.
Boracic acid
Alumina.
43'71
8.76
5'56
Potash
Soda.
6'98
4'50
Lime.
2.72
Magnesia
4'76
Oxide of tin.
23'01
680
IRON.
No. 2 is formed of 15 parts of quartz, 37°5 parts of felspar,
10 parts of clay, 40 parts of borax, 20 parts of soda, 15 parts
of saltpetre, 7°5 parts of magnesia, and 25 parts of oxide of
tin; thus containing—
(6
Silica.
Boracic acid
Alumina.
Potash
Soda .
Magnesia
25°59
12 75
9'12
II°53
15.95
3°33
21.81
Oxide of tin
The following are recipes for enamels given in Ure's
Dictionary of Arts, &c.," vol. ii., p. 211 :—
Grey Mixture Enamel.
lbs.
Oz.
Sand.
ΙΟ O
Red lead
33
O
Boracic acid
20
Cullet
114 O
Bicarbonate of soda
16 0
Nitre.
I 2
Manganese.
o 8.5
194 10'5
Another is-
lbs. Ozs.
Flint
•
36
.0
Boracic acid.
24
Bicarbonate of soda
24
O
Nitre
18
O
102
White Mixture Enamel.
Cullet
Boracic acid.
lbs.
Ozs.
II
O
7
Bicarbonate of soda
Phosphate of lime .
Oxide of antimony.
O
3
4
8
О
2
21
14
ENAMELLING CAST-IRON.
681
of
The Preparation of the Mass for the First Coating.
The components being weighed and intimately mixed, are
heated in a Hessian crucible provided with an aperture
an inch wide at the bottom. The temperature must be so
high that the mass melts and flows through a corresponding
opening in the support of the crucible into a vessel filled with
water, or an arrangement must be made allowing a tapping
off the mass from time to time, whilst the upper crucible is
charged anew. Sometimes the crucibles used for melting
the enamel for the second coating are used for this operation
without permitting the mass to run out; after cooling, the
crucible is broken. The result will be reddish or yellowish in
colour, and greenish yellow if containing any lead.
The melted mass is then pounded, sifted, and ground
either by itself or together, with the additions already men-
tioned. The finely ground mass mixed with water is passed
through a fine hair-sieve into a copper or enamelled iron
vessel, and allowed to rest for a week or two to render it
more homogeneous.
Preparation of the Enamel for the Second Coating.
This process is taken from the recipe followed in the pre-
paration of the previous mass, except that the melting is
repeated two or three times in order to obtain the enamel
free from blisters. In order to mix the enamel intimately
with the oxide of tin, it is made to run out through an
opening 1-12th of an inch in diameter. The melted mass
must be white, of a glassy, greasy lustre; it must have a
conchoidal fracture, and be harder and more compact than
common glass.
The process of enamelling necessitates the following mani-
pulations:-
a. Cleaning the Vessels by exposing them to the action
of dilute sulphuric acid (1 volume of sulphuric acid to 50 of
water) for from 12 to 18 hours.
b. Scouring with Sand and washing, first with cold water,
and afterwards with boiling water, thus facilitating the
drying of the vessels and preventing oxidation.
682
IRON.
c. Fixing the First Coating.-The mass being brought
to the consistence of cream by mixing with boiled luke-warm
water and a little carbonate of ammonia, or sometimes caustic
ammonia, is next brushed over the vessel, which must be
heated to 50° C., some is then poured into the vessel, and
the latter swung about in order to divide the mass uniformly.
It is then reversed, whilst the flowing out of the mass is
facilitated by slight blows from a wooden hammer; the edges
of the vessel are then cleaned either with the finger or a stiff
instrument of leather or wood.
d. Drying the Vessels.-They are now suspended by
wire a certain distance above an iron plate, which is heated
from below. The vessels are next placed upon the plate
itself, thus heating them to a temperature of from 200° to
300° C.; this operation of drying takes about one hour.
e. Burning-in the First Coating. This manipulation
is effected in 10 or 15 minutes at a full red heat (melting
point of brass) in an iron muffle provided with a trap-door.
In order to heat the vessels uniformly they must be
several times turned. The mass for the first coating must
adhere firmly to the iron in a uniform layer without being
either pulverulent or fused.
f. Fixing the Second (Glazing) Coating.-The vessels
after having cooled and been cleaned outside from oxide, are
dipped into water of about 50° C., and the prepared enamel
is then uniformly distributed over the vessels by swinging
them without any previous brushing. The drying is effected
as after the first coating.
g. Burning-in the Second Coating.-The vessels are
now exposed to a red heat for about 12 minutes, being fre-
quently moved and turned, and the cold air excluded as far
as possible, thus melting the enamel and fixing it to the first
coating. The proper vitrification is indicated when the
enamel shows a lustre; if heated beyond this point the enamel.
becomes full of bubbles. The exterior of the vessel before its
complete cooling is then blackened by brushing it with tar or
with an admixture of pine-oil and soot.
In December, 1799, a patent* was obtained for the process
* Dr. URE's Dictionary of Arts, &c., vol. ii., p. 209.
ENAMELLING CAST-IRON.
683
of enamelling iron by Dr. Samuel Sandy Hickling. His
specification is subdivided into two parts:—
1. The coating or lining of iron vessels, &c, by fusion
with a vitrifiable mixture, composed of 6 parts of calcined
flints, 2 parts of composition or Cornish stone, 9 parts of
litharge, 6 parts of borax, I part of argillaceous earth, I part
of nitre, 6 parts of calx of tin, and I part of purified
potash.
Or, 2ndly, 8 parts of calcined flints, 8 red lead, 6 borax,
5 calx of tin, and I of nitre.
Or, 3rdly, 12 of potters' composition, 8 borax, 10 white
lead, 2 nitre, I white marble calcined, I argillaceous earth,
2 purified potash, and 5 calx of tin.
Or, 4thly, 4 parts of calcined flints, I potters' composition,
2 nitre, 8 borax, I white marble calcined, argillaceous
earth, and 2 calx of tin.
Which ever of the above compositions is taken it must be
finely powdered, mixed, and fused; the vitreous mass is to be
ground when cold, then sifted and levigated with water. It
is then made into a paste with water or gum-water. This
paste is smeared or brushed over the interior of the vessel,
dried, and fused in a muffle at a proper temperature.
Calcined bones are also used as an ingredient of the flux.
The fusibility of the vitreous compounds is varied according
to the heat to be applied to the vessel, by using various pro-
portions of the siliceous and fluxing materials. Colours and
also gilding may be given.
The second part or process in the specification describes
certain alloys of iron and nickel, which he casts into vessels
and lines or coats them with copper precipitated from its
saline solutions. It also describes a mode of giving the
precipitated copper an enamel surface by acting upon it with
bone ashes and zinc with the aid of heat.
A factory of these enamelled hollow wares was carried on
for some time, but it was given up for want of due encourage-
ment.
A patent was granted to Thomas and Charles Clarke on the
25th of May, 1839, for a method of enamelling or coating the
inner surfaces of iron pots and saucepans in such a way as to
684
IRON.
prevent the enamel from cracking and splitting off from the
effect of fire. This specification directs that the vessel shall
first be cleaned by exposure to the action of very dilute sul-
phuric acid (sensibly sour to the taste) for three or four hours,
next boiled in pure water for a short time, and then the com-
position is applied, consisting of 100 lbs. of calcined flints,
50 lbs. of borax calcined and finely ground with the above.
This admixture is to be fused and gradually cooled.
40 lbs. weight of the above product is to be taken with
5 lbs. weight of potters' clay to be ground together in water
until the mixture forms a pasty consistent mass, which will
leave or form a coat on the interior surface of the vessel
about one-sixth of an inch thick. When this coat is set by
placing the vessel in a warm room, the second composition
is to be applied. This consists of 125 lbs. of white glass
(without lead), 25 lbs. of borax, 20 lbs. of soda (crystals) all
pulverised together and vitrified by fusion, then ground,
cooled in water and dried. To 45 lbs. of that mixture 1 lb.
of soda is to be added, the whole mixed together in hot
water, and when dry, pounded, then sifted finely and evenly
over the inner surface of the vessel, previously covered with
the first coating or composition whilst this one is still moist.
This constitutes the glazing. The vessel thus prepared is
to be put into a stove, and dried at the temperature of 212°
Fahr. It is then heated in a kiln or muffle like that used
for glazing china. The kiln being brought to its full heat,
the vessel is placed first at its mouth to heat it gradually,
and then moved into the interior to fuse the glaze.
practice it has been found advantageous also to dust the
glaze powder over the fused glaze and apply a second fluxing
heat in the oven; by this double application the enamel
becomes much smoother and sounder.
In
Messrs. Kenrick, of West Bromwich, produced in their
factory some excellent specimens of enamelled saucepans of
cast-iron. Dr. Ure exposed the finely enamelled saucepans
of Messrs. Kenrick to very severe trials, having fused even
chloride of calcium in them and found them to stand the
fire very perfectly without chipping or cracking.
5. Coating Iron with a Glass-like Mass.-The well-
PRODUCTION OF WROUGHT-IRON.
685
cleaned articles are brushed with a solution of gum arabic,
and then a powder prepared by melting 130 parts of glass
free from lead, 20 parts of soda, and 12 parts of borax is
strewn over them. The articles are then dried in a stove at
a temperature of 100° C., and afterwards heated up to redness
and gradually cooled. This method is used in France* (fer
controxidé or inaltérable) and in America.t
Wyatt‡ recommends a glass composed of 3 parts of white
lead, 2 borax, and I flint.
PRODUCTION OF WROUGHT-IRON.
Wrought or malleable iron is seldom produced direct from
the iron ores, but usually from pig-iron, and it is the purest
of the different sorts of iron produced metallurgically. Carbon
(from 0.1 to 0'5 per cent) forms an essential component, and
deprives the iron of the undesirable softness possessed by pure
iron. But wrought-iron may also be contaminated either by
substances chemically combined, such as sulphur, phosphorus,
silicon, earthy metals, manganese, antimony, arsenic, &c.,
or by mechanical intermixtures, as oxidised iron, slag, &c.;
sometimes it possesses a higher amount of carbon, as we
have before shown, or its carbon is unevenly distributed. All
these circumstances modify the properties of the wrought-iron
(strength, property of welding, hardness, &c.). Moreover,
the influence of temperature, electricity, or sudden shocks
may cause a molecular change and thus deteriorate the
quality of the iron. The appearance of the fracture and the
outside of the iron bars, their behaviour at higher tempera-
tures, and their resistance to shocks usually show what in-
fluences have caused the modification of the iron, and
accordingly wrought-iron is called by different names (red-
short, cold-short, burnt, &c.).
Wrought-iron differs from pig-iron and steel chiefly in its
behaviour, as explained on page 264.
Bgwkfd., xiii., 653, 798. DINGL., Bd. 112, p. 391. Polyt. Centr., 1850,
p. 818.
+ B. u. h. Ztg., 1859, p. 51. .
Polyt. Centr., 1851, p. 60.
686
IRON.
The Properties of Wrought-Iron.
The following are the chief properties of wrought-iron :—
Colour. Wrought-iron varies in colour between silver-
white and a lightish grey, with different.lustres; a light colour
with little lustre, and a greyish colour with a dark lustre
characterise a good iron; a bluish-white colour and a strong
lustre indicate a burnt iron; a silver-white colour and a
strong lustre a cold-short iron; a dark colour and little lustre.
red-short iron; very little lustre indicates that the iron is
brittle and soft.
Texture. The original texture in every good wrought-
iron which has only been hammered from a ball to a large
bar is a jagged grain, the grains varying in size, and according
to the amount of carbon present, the iron upon further
forging becomes either uniformly fine-grained* or fibrous.
The fibrous iron is softer and tougher than the grained iron,
and its toughness increases with the length and fineness of
the fibres. The finer the grains and the darker the colour
of the iron, the richer in carbon, the more steel-like and the
harder it is, and the coarser its grains the more easily will it
become fibrous.
According to Gurlt,† soft wrought-iron poor in carbon,
when forged or rolled to a long bar, more easily attains the
fibrous structure than the harder sort richer in carbon, because
it is more inclined to crystallise, and because it consists
of an aggregation of small cubes lying beside each other.
These cubes are drawn out in length by the manipulation of
forging, but they retain their tendency to reassume their
original position when externally influenced by long-continued.
vibration, hammering, high temperature, or the galvanic
current. As these influences enlarge the volume of the iron.
the resulting product is less strong than the original fibrous
iron; fine-grained wrought-iron with a larger amount of
carbon is therefore preferred for purposes where these in-
fluences may affect it, as it is less inclined to form fibres.
* TUNNER'S Leob. Jahrb., 1859, viii., 161.
Berggeist, 1860, No. 19. B. u. h. Ztg., 1860, p. 502.
THE PROPERTIES OF WROUGHT-IRON.
687
According to V. Mayrhofer and Gurltt the additionst
(nickel) which have been given to the iron, to prevent the
conversion of the fibres into crystals, merely deteriorate the
quality of the iron. The modified structure of the fibrous
iron is not an act of crystallisation which could be prevented
by alloying the iron with nickel, but it is the result of the
elasticity of the iron; the crystals of the iron distorted by
hammering or rolling have a tendency to. reassume their
original form, owing to the elasticity of the iron. According to
Eggertz|| wrought-iron with o˚5 per cent of carbon is already
somewhat hard and changes into steel (fine-grained iron); the
smallest amount of carbon found in a variety of Bessemer
iron was o'08 per cent. The manner in which wrought-iron
is broken has a considerable influence on the appearance of
the fracture;§ upon breaking an iron bar suddenly by a hard
blow its fracture will appear grained or crystalline, whilst
the same bar will show a more or less fibrous texture when
broken slowly by gradually increasing force. Lüders dis-
covered file-like markings on the fracture of broken bars.
Rühmkorff states that soft iron when rendered magnetic
is more difficult to be attacked by the file, i.e., harder, than
when not in connection with a magnet.
Wrought-iron in a soft weldable state is amorphous, but
sometimes it occurs in cubes or in combinations of the
octahedron, cube, and garnet octahedron, in the puddling
process for instance, when pig-iron has been kept for a long
time in the liquid slag.
Fine-grained iron approaches steel on account of its
uniformity, strength, and hardness. As its production re-
quires a quicker process with a smaller admission of air than
the fibrous iron, the employment of a better sort of pig-iron is
at first required to remove the injurious substances; how-
ever, some sorts of pig-iron are better adapted for the
*
Oesterr. Ztschr., 1861, No. 47. B. u. h. Ztg., 1862, p. 200.
† Berggeist, 1861, No. 24.
HARTM., Fortschr., v. 185. Polyt. Centr., 1863, p. 1172.
B. u. h. Ztg., 1863, p. 374.
Ibid., 1863, p. 140. LEOB., Jahrb., 1859. viii., 162.
¶DINGL., Bd. 155, p. 18. B. u. h. Ztg., 1861, p. 392.
688
IRON.
Some sorts of pig-
are also well adapted
production of fine-grained iron than other sorts, but the
reason is not thoroughly understood. At Low Moor,* iron
workers use refined pig-iron which has been produced from
very good ores with coke and cold blast.
iron rich in carbon and manganese
for the production of fine-grained iron.
that the presence of sulphur induces a
wrought-iron; pig-iron intended for conversion into fibrous
wrought-iron is therefore produced from an ore mixture
containing some sulphur, and grained iron from a mixture
free from sulphur, with a larger addition of limestone.
V. Mayrhofer‡ states
fibrous texture of the
Whilst good wrought-iron shows a uniform jagged grain
moderate in size, the grains of cold-short iron (owing to the
presence of phosphorus) are arranged in a somewhat strati-
fied manner; wrought-iron rendered cold-short by a continued
cold hammering (the aggregation of the iron being thus dis-
arranged), shows a fine grain with a strong lustre. Burnt
iron has coarse, uneven, angular grains of strong lustre ;
iron containing particles of crude metal, uneven, coarse and
fine grains; iron which is brittle owing to the presence of
silicon, a mixture of grained and fibrous texture.
The specific gravity of wrought-iron is from 7.6 to 77, and
increases with the purity of the iron.
The strength varies with the foreign admixtures and the
mechanical treatment of the iron. Upon heating a strong
good iron and cooling it slowly or quickly without submitting
it to a mechanical treatment, its strength will be impaired,
and the strength will increase the more finely the iron is
drawn out. Whilst an iron rod of 1 square inch section
bears from 500 to 600 cwts. this weight may be increased to
from 1000 to 1100 cwts. per square inch upon drawing out
the rod so as to form wire. According to V. Burgs the
HARTM., Fortschr., vi. 223.
† Ibid., iii., 262.
LEOB., Jahrb., 1861, x., 388.
ZIUREK'S technol. Tabellen, 1863, p. 142. B. u. h. Ztg., 1857, p. 165;
1861, p. 170. DINGL., Bd. 149, p. 394; Bd. 155, p. 394. Berggeist, 1861,
•
No. 22.
B. u. h. Ztg., 1859, p. 257.
THE PROPERTIES OF WROUGHT-IRON,
689
strength of iron upon heating increases to a certain degree
of temperature, but beyond that it decreases.
Fine-grained iron is stronger than fibrous iron and becomes
less easily crystalline and fragile (page 687). According to
Clay the brittleness of fibrous iron is prevented by heating
and slowly cooling it.
The tensile and transverse strength is greater than that of
pig-iron, but not its crushing strength. Steel is stronger
than iron. Fine grained iron is harder than fibrous iron, and
its hardness increases with the amount of carbon present,
but it may also be increased by other substances (manganese,
&c.); in the latter case the hardness remains the same
whether the iron is cooled slowly or quickly, whilst iron that
has been rendered hard by the presence of carbon approaches
steel (when containing above 0'5 per cent of carbon) and
becomes harder upon heating and quenching in water. Iron
which is soft at the common temperature remains soft at
higher temperatures; it welds easier, and when hot it can
be more easily treated than hard iron. Wrought-iron made
hard by a sudden cooling may, at the same time, be brittle,
and this brittleness may be also induced by the presence of
certain substances, chiefly earthy metals; but manganese
effects this in a less degree.
Upon heating bright wrought-iron at a gradually rising
temperature, from 200° to 400° C, it shows a succession of
characteristic colours (pale yellowish, pale straw yellowish,
orange, violet, darkish violet, dark blue, light blue, green,
pale green, and, at last, colourless); hard iron attains those
colours sooner than soft iron. Next the glowing heats
appear—namely, at 525° C. begins the red heat, at 700° the
dark red heat, at 1000° the perfect cherry-red heat, at 1100°
dark orange, at 1200° a light glowing heat, at 1300° C., the
white heat, at 1400° a strong white heat, at from 1500° to 1600°
a dazzling white heat, and at from 1900° to 2120° C. fusion.
According to Clement and Desormes, wrought-iron fuses
at 2118° C., and, according to Elsner,† in the highest heat of
VOL. II.
*
B. u. h. Ztg., 1857, p. 191.
↑ Polyt. Centr., 1859, p. 317.
2 Y
690
IRON.
a porcelain furnace; the wrought-iron produced in Bessemer's
furnace may be cast in moulds. Pouillet gives the fusion
point of wrought-iron at 1600° C. only. Before melting, the
iron becomes pasty and weldable. Hard iron welds sooner
than soft iron.
Behaviour of Wrought-Iron towards other
Substances.
Oxygen. When exposed to the influence of moist air
iron oxidises, and red-short iron more quickly than cold-
short iron; iron in a red-hot state oxidises on the admission
of air and becomes coated with a dark oxide (iron scale) which,
according to Berthier,* consists of 4FeO, Fe₂O,. This oxide
causes loss of iron, and adheres less to the iron beneath,
according as it has been formed quicker and at a higher
temperature. The surface of wrought-iron treated at lower
temperatures has a reddish appearance.
༢.
The scale is more perfectly removed from iron by
hammering than by rolling, and it renders iron unsound if it
is allowed to remain with the iron during certain further
treatments.
Iron exposed to a continued high temperature with an
admission of air loses its carbon, assumes an uneven coarse-
grained texture of a white colour and strong lustre (burnt
iron), and is friable either cold or hot. Wrought-iron may
acquire a similar crystalline texture without essentially losing
carbon when kept for a longer time at a higher temperature
with atmospheric air excluded (page 688); wrought-iron
made cold-short by too long a hammering when cold shows
a uniform finer grain. All these modified varieties of iron
may be improved if they are raised to a welding heat whilst
under a coating excluding the atmospheric air, and when in
this hot state they are condensed by a hammer or rolls.
Wrought-iron, cold-short owing to the presence of phosphorus,
cannot be improved in this manner. In order to protect the
iron from burning, sand is strewn over it at the welding heat,
thus forming a liquid cover of slag; chloride of sodium is
* KARST., Arch., 1 R., xi., 433; ix., 508; xiii., 365.
THE PROPERTIES OF WROUGHT-IRON.
691
sometimes added to the sand, or some glass, and, occasionally,
even borax; Schwarz brushes the iron with a concentrated
solution of water-glass. The fusing iron scale also forms a
protection against burning.
Nitrogen.-Iron combines with nitrogen in different.
proportions. Wrought-iron produced on a large scale only
contains a small amount of this substance. According to
Boussingault, iron wire contains from o'oco075 to o'000086
of nitrogen. Jullien* states that burnt iron contains nitrogen,
and, according to Fleury,† grained iron is produced by the
liberation of nitrogen, which then combines with carbon.
Sulphur is said by Karsten to render iron red-short if
present in it to 1-10,000th part; this, however, is exag-
gerated, as good varieties of wrought-iron may contain as
much as 3-10,000ths; and some thousands of sulphur
are required to render the iron decidedly red-short, and the
more so the more imperfect its conversion into wrought-iron.
A well and uniformly converted wrought-iron containing
0'04 per cent of sulphur was capable of being punched, whilst
wrought-iron with such an amount of sulphur is usually
inferior in quality.
The different parts of one and the same iron bar some-
times contain varying amounts of sulphur.
Phosphorus lessens the strength of iron at the common
temperature, and makes it cold-short, but it greatly in-
creases its property of welding. Almost every kind of
wrought-iron contains some phosphorus, which, according to
Karsten, is harmless if it does not exceed o˚5 per cent. Eg-
gertz‡ states that wrought-iron containing from 0.25 to
0'30 per cent of phosphorus shows cold-shortness, but can
be applied to the manufacture of nails, wire, and other fine
forge articles. The more phosphorus the iron contains, the
softer it is when heated. Upon cooling, it takes a crystalline
texture, and becomes brittle if not forged immediately after
heating; and the crystalline texture and the cold-shortness
will be lessened the longer it is under the hammer.
B. u. h. Ztg., 1862, p. 148.
† Ibid., 1861, p. 439.
† Ibid., 1860, p. 418.
2 Y 2
692
IRON.
Silicon, usually not less than o'05 per cent, renders the
iron brittle at common and higher temperatures; the fracture
of the iron shows black fibres and grains, and often cleavage
planes.
Aluminium, calcium, and other earthy metals* act like
silicon.
Tin, Antimony, and Arsenic render wrought-iron cold-
short, and, according to Wehrlet and Mrazek,‡ arsenic
especially lessens the property of welding (page 696). Mrazek
found the composition of a cold-short wrought-iron which
could only be welded with difficulty, to be as follows:-
Silicon
Carbon .
Phosphorus
Arsenic.
Cobalt
Nickel
Sulphur
Copper
0·630
0'055
0*290
0*375
0'017
0'021
trace
trace
1
The English commission for investigating the quality of
iron with regard to its applicability to railway purposes found
that best Dundyvan wrought-iron in bars of I square inch
broke when burdened with 24°33 tons; 4 cwts. of this iron
containing 4 lbs. of zinc broke at a burden of 25.86 tons, and
the same quantity, containing 1 lb. of tin, at a burden of
23°39 tons.
Antimony renders wrought-iron hard and crystalline.
Nickel and Cobalt.-Rubach|| found in a cold- and red-
short wrought-iron 153 per cent of nickel, o'63 per cent of
cobalt, and o*19 per cent of carbon, but no copper, arsenic,
phosphorus, sulphur, or silicon, which is somewhat contra-
dictory when compared to the statements made on page 318.
Manganese purifies the iron on its conversion into wrought-
iron, and but little of it enters the iron; in this case the iron
is rendered harder. According to Tunner, o'5 per cent of
manganese scarcely effects the properties of wrought-iron.
*KARST., Arch., 1 R., ix., 417.
+ B. u. h. Ztg., 1862, p. 408.
+
WEHRLE, Hüttenkunde, ii., 23.
|| DINGL., Bd., 117, p. 395.
ASCERTAINING THE QUALITY OF WROUGHT-IRON. 693
Copper, when contained in wrought-iron to some tenths,
renders it very red-short. A small amount of copper is more
injurious in steel than in wrought-iron. List* states that
copper forms an impediment when puddling the pig-iron,
whilst Schafhäutlt asserts that the puddling process is
not impeded even when a great deal of copper is added to
the iron.
Ascertaining the Quality of Wrought-Iron.‡
1. For this purpose the iron bars are submitted to the fol-
lowing tests :-
a. The bars are thrown from a certain height upon a small
faced anvil, when cold-short iron, burnt iron, or iron that has
been hammered in a cold state will break; this breakage is
facilitated when the bars are tested in cold weather.
b. A weight is allowed to fall upon the bars, which are
placed horizontally upon two supports about 3 feet apart.
The weight and the height from which it falls vary with the
size of the bars.
c. The bars are bent to and fro so as to form angles of 90°
or 180°, the angles varying with the thickness of the bars.
The bars should bend to a definite amount without fracture;
cold-short iron and iron that has been hammered in a cold
state will often break at the first bending. Soft iron bends
without making any noise, whilst hard iron on bending
vibrates and crackles.
2. The exterior appearance of the bars to a certain degree
indicates the quality of the iron. Good iron bars are sharp
and clean at the corners and have a smooth surface. Bars
of bluish tint and very bright appearance have been
hammered when in a cold state. Some bars show on their
surface spots of intermingled cinder or hammer slag. Burnt
iron is blunt at the edges. Iron bars showing cracks cross-
wise towards the edges are either red-short or are still
* B. u. h. Ztg., 1860, p. 52.
+ Berggeist, 1860, Nos. 28, 29.
f
ERDM., J. f. ök. u. techn. Chem., 1833, No. 2. Freiberger Jahrbuch,
1841, p. 1.
694
IRON.
partially in the state of crude iron; and cracks lengthwise
indicate the iron to be very red-short or very crude.
3. The appearance of the fracture as regards texture and
colour (page 685). To obtain a clean fracture the bar is
notched some lines deep with a sharp chisel and broken. A
fracture obtained by repeatedly bending the bar is almost
invariably fibrous.
4. The behaviour at higher temperatures, namely:-The
capability of welding and throwing out coruscations; beha-
viour of the iron after being quenched in water, and at its
mechanical treatment; drawing out the bars to smaller
dimensions and bending them on their narrow side; punching;
forging a horse-shoe; pointing the iron either cold or
hot, &c.
5. Corroding the bars with dilute acids; when hard and
polished iron will show a darker colour than cold-short iron,
and a lighter colour than soft and red-short iron. A bar of
varying hardness has a spotted appearance.
Wrought-iron may be classified according to the following
characteristics, and these characteristics are more obvious
the smaller the dimensions of the iron bars :—
1. Fine-Grained, Hard Iron.-This iron is of a uniform.
fine grain, between tin-white and lead-grey in colour; it is
moderately bright; its exterior is smooth and has sharp
edges; the iron is very hard, strong and very tough, and
elastic. Upon welding it throws out fine red sparks, and
only yields a little fine hammer slag. The welding is per-
formed at a lower temperature than is required in the case
of soft iron, but as it is forged with more difficulty, owing to
its hardness and density, the welding operation must be con-
ducted properly,* and great care must be bestowed upon the
heating of the iron. With regard to its treatment in
the fire and in forging this iron stands between raw steel
and wrought-iron produced in fineries. Its amount of car-
bon is but slightly lessened even after a repeated treatment
at a higher temperature. Upon tempering the hot iron
* WERLISCH, Verhalten des Rothehütter Puddeleisens, B. u. h. Ztg., 1859,
pp. 19, 36.
DIFFERENT SORTS OF WROUGHT-IRON.
695
in water it assumes a finer grain, a lighter colour, and
On
increases in hardness; it is also more inclined to warp.
treating the iron with a file small filings result, and the file
only attacks the iron very slightly.
2. Soft, Fibrous Iron.—It is tough, its exterior shape is
easily changed, but it does not break easily; it is not very
stiff, wears off easily, and, upon bending, it shows fine, long
fibres of a lightish grey colour. Continued shocks, cold
hammering, long heating, &c., render it crystalline and
brittle. The bars are smooth and acute angled. It can only
be welded at the highest temperatures, when it emits fine
bright sparks; when tempered in water it remains soft and
retains its texture. The file produces coarse filings from it,
and cuts it deeper than the preceding iron.
3. Red-Short Iron (owing to the presence of sulphur and
sometimes of copper) has a dark, rather dull fracture, and
may be fibrous or granular; it may be bent several times
without breaking. Very red-short iron has grey coarse fibres
without lustre. The bars have cracks on the edges and
cracks lengthwise if extremely red-short.
At the common temperature the iron is strong and flexible,
easily oxidising; at the welding heat it is very tractable and
emits large sparks, but it cracks upon treatment at a red
heat, especially if forged out into thin bars, or if bent or
punched in a red-hot state. An increase of the red-shortness
lessens the property of welding, the more so the harder
the iron. Red-short iron cannot be improved by exposure to
a welding heat like—
4. Raw-Short Iron, which is the result of an imper-
fect conversion of the pig-iron into wrought-iron, when the
foreign components are insufficiently or unequally extracted.
This iron shows separate coarse and bright grains of finished
wrought-iron and higher carbides (steel or pig-iron); it is
hard, brittle, difficult to weld and to forge; it shows cracks
on the edges, chiefly on the ends of the bars, and is both cold-
and red-short.
This sort of iron is oftener produced by the finery process.
than in puddling furnaces.
5. Cold-Short Iron, comprising several kinds of hard iron
696
IRON.
which weld easily, and behave well at a high temperature, but
easily crack and break when bent, thrown down, &c., at the
common temperature. They have a white grain varying in
brightness.
a. Wrought-Iron containing Phosphorus has bluish
white, very bright schistose grains, which are arranged in
strata. It has the properties stated on page 691, and is not
improved by treatment at the welding heat. It is very fusible,
slightly oxidisable, takes a bright polish, and upon filing
yields short, rough, white filings.
Antimony and arsenic have the same influence upon iron
as phosphorus; iron containing those substances emits a
greyish vapour when heated up to white heat.
b. Iron Hammered when Cold. This hammering is
sometimes performed with iron the temperature of which
has been lowered too far below the glowing heat, whilst the
faces of hammer and anvil are kept wet. The resulting iron is
very bright, with a bluish tint on its surface; it has a small,
compact, bright grey grain, and is hard and brittle at the
common temperature, which distinguishes it from fine-
grained iron, of which it only possesses the hardness; expo-
sure to a moderately high temperature may transform it into
soft iron. Iron which has been hammered when too cold
shows a greater hardness and brittleness than its other
qualities would appear to indicate.
6. Burnt Iron which has been exposed for a longer time.
to a dry heat with an admission of air, and has thus been.
rendered poor in carbon, has an irregular, very coarse-grained
texture. The coarse and small angular grains show a strong
lustre; they are irregular and loosely arranged, showing a
yellowish white colour. This iron is cold and red-short, but
may be improved by exposure to a strong welding heat.
Fibrous iron, transformed by continued heating, vibrating
motion, &c., behaves in the same manner.
The influence of silicon upon wrought-iron has been
mentioned on page 300. The resulting iron is termed in
German faulbrüchig. It has short thick fibres of a dark
colour, without lustre, and with intermixed grains. At a
higher temperature it is flexible, but breaks at the common
MAKING WROUGHT-IRON DIRECTLY FROM THE ORE. 697
temperature more or less easily according to its amount of
silicon.
Some iron bars are faulty, owing to bad welding, erroneous
mechanical treatment, or to particles of slag and hammer
scale which they contain in admixture. They may be
improved by suitable exposure to welding heats.
METHODS OF MAKING WROUGHT-IRON DIRECT
FROM THE ORES.
At present this process is seldom used on account of its
numerous disadvantages. It requires pure, rich, and easily
fusible ores, and is performed in interrupted operations; much
iron is scorified. The consumption of fuel is very large, and
lastly, the product is seldom uniform, and is mixed with slag,
which can only be removed by repeated welding.
The process is carried on either in hearths similar to the
common forges (bloomeries), or in a small kind of furnace
(Stücköfen).
The process in hearths has been carried on differently in
different countries, and although the differences are very
slight, yet they are the cause of the following classification
of the process:
I. The German Process in Bloomeries.-The hearth is
lined with charcoal dust or breeze and coated with a glaze
of easily fusible ore mixture, and then filled with charcoal.
Next the ore is charged with a shovel, and the ore charge
repeated when the former charge has smelted and passed
through the coal; the blast is introduced by a tuyere in a
horizontal position. The resulting ball or lump of iron is then
either re-melted or cut into pieces, and forged out at the
smelting of a new ball, if found to be sufficiently malleable.
Frequently only half the iron contained in the ore is ex-
tracted, and 80 cubic feet of charcoal are consumed for the
production of I cwt. of iron.
2. Catalan or French Process.-The ore is either roasted
and reduced in one hearth and smelted in a second one, or
both operations are performed in the same hearth; the
hearths vary in dimensions in different countries. The
smallest hearths are 20 inches long and broad, 16 inches
698
IRON.
deep, and the tuyere is placed 9 inches above the bottom.
The ore charge for this kind of hearth is 3 or 4 cwts. Other
hearths used in Navarre and Guipuscoa are 30 inches long,
24 inches broad, and 24 inches deep, the tuyere lying 14 or
15 inches above the bottom; 5 or 6 cwts. of ore are charged.
The largest hearths are those employed in Biscay and in
Navarre; they are 40 inches long, 30 or 32 inches broad,
and from 24 to 27 inches deep. The tuyere is placed
18 inches above the bottom, and 7 or 8 cwts. of ore are
charged. The hearth is also lined with a coating of coal
dust, and a wall of ore in large pieces is erected in the hearth
on the side opposite to the tuyere. This wall is covered
with a mixture of coal dust and small ore moistened with
water, and the space between the side of the tuyere and the
ore walling (amounting to about two-thirds of the hearth) is
then filled up with charcoal. For the first one and a half
or two hours blast of very low pressure is employed, and
during this time the pieces of the ball, or result of the last
operation, are heated in the hearth at the same time to draw
them out. During this period the reducing gases penetrate
the ore wall without smelting it. The blast is then turned
full on, blowing against the lower part of the ore wall, and
owing to the higher temperature produced, separates the
liquid slag from the reduced iron. After tapping off the
slags, the ore wall is moved by means of iron bars somewhat
nearer to the tuyere; the softened foot of the wall is gradually
broken off and brought before the tuyere, thus causing the
wall gradually to sink. The resulting spongy masses of iron
in the hearth are worked into a lump, which is lifted out and
carried to the hammer to be forged and condensed; it is then
cut up into several pieces and re-heated during the following
operation. The character of the slag greatly influences the
process; if it is pasty, the necessary fluidity is imparted
by the addition of finely divided ore, which is allowed to
dissolve in the slag, thus rendering it easily fusible, and
facilitating its removal during the subsequent forging. The
easily fusible slag containing much protoxide of iron, also
reacts upon the reduced masses of spongy metal, and pre-
vents the formation of cast-iron.
MAKING WROUGHT-IRON DIRECT FROM THE ORE. 699
About 50 cubic feet of coal are consumed for the produc-
tion of I cwt. of iron. One operation takes six hours, and
from 70 to 80 cwts. of bar-iron are produced weekly when
working ores yielding 33 per cent of iron.
The blowing machine usually employed is the trompe,
which is represented in Figs. 203 and 204.
A is a water
FIG. 203.
F
K
T
FIG. 204.

T
reservoir; B, B are rectangular throats or water pipes pro-
vided with holes or aspirators, g, g, which incline inwards
and downwards. These throats connect the reservoir, A,
with the the wooden case or wind chest, G, provided with
the opening, D, for the exit of the water, and the tube, F,
receives the blast. F is connected with the blast-pipe, T, by
means of a sheepskin tube, K. The vertical water-pipes, B, B,
are provided on the top with conical adjoined tubes for
700
IRON.
admitting the water, and which may be closed by the
wedges, m.
In some parts of America,* particularly in the States of
Vermont and New Jersey, the Catalan forge is extensively
employed for smelting the rich magnetic ores which abound
there. The form of this forge is thus described by Overman :-
The whole is a level hearth of stone work from 6 to 8 feet
square, at the corner of which is the fire-place, from 24 to
30 inches square, and from 15 to 18, often 20, inches deep.
Inside it is lined with cast-iron plates, the bottom plate being
2 or three inches thick. Fig. 205 represents a ground plan,
and Fig. 206 a cross section through the fire-place and tuyere,
FIG. 205.

α
commonly called tue-iron. d represents the fire-place, which,
as before remarked, varies in its dimensions. The tuyere, b,
is 7 or 8 inches above the bottom, and more or less inclined
according to circumstances. The blast is produced by wooden
bellows of the common form, or by square wooden cylinders,
worked by water wheels. The crystallised magnetic ore is
usually employed; this ore very readily falls to a coarse sand,
and when roasted the grains vary in size from a pea to the
finest grain; sometimes the ore is used without roasting.
In the working of these fires much depends on the skill and
experience of the workmen. The result is subject to con-
siderable variation, according to whether economy in coal
or in ore is the object. Thus modifications are required in
the construction either of the whole apparatus or of parts.
The manipulation varies in many respects. One workman,
by inclining his tuyere to the bottom, saves coal at the
*URE'S Dictionary of Arts and Mines, &c., vol. ii., 691.
MAKING WROUGHT-IRON DIRECTLY FROM THE ORE. 701
expense of obtaining a poor yield. Another, by carrying his
tuyere more horizontally at the commencement, obtains a
larger amount of iron, though at the sacrifice of coal. A
good workman will pay great attention to the tuyere, and
alter its dip according to the state of the operation. The
general manipulation is as follows:-The hearth is lined
with a good coating of charcoal dust, and the fire-plate, or
the plate opposite the blast, is lined with coarse ore, if any
FIG. 206.

ཚགས་ ར ་ས་ ་ ་
is at disposal. If no coarse ore is employed, the hearth is
filled with coal, and the small ore piled against a dam of
coal dust opposite the tuyere. The blast is at first gently
urged and directed upon the ore, while the coal above the
tuyere is kept cool. 400 lbs. of ore are the common charge,
two-thirds of which are thus smelted, and the remaining
third, generally the finest ore, is kept in reserve to be thrown
on the charcoal when the fire becomes too brisk. The char-
coal is piled to the height of two, sometimes even three and
four feet, according to the amount of ore to be smelted.
When the blast has been applied for an hour and a half or
two hours, most of the iron is melted, and forms a pasty mass
at the bottom of the hearth. The blast may now be urged
702
IRON.
more strongly, and if any pasty or spongy mass yet remains,
it may be brought within the range of the blast and melted
down. In a short time the iron is revived, and the scoriæ
are permitted to flow through the tapping-hole, so that only
a small quantity of cinders remains at the bottom. By
means of iron bars the lump of pasty iron is brought before.
the tuyere. If the iron is too pasty to be lifted, the tuyere
is made to dip into the hearth; in this way the iron is raised
from the bottom, directly before the tuyeres or to a point
above them, until it is welded into a coherent ball 12 or 15
inches in diameter. This ball is brought to the hammer or
squeezer, and shingled into a bloom, which is either cut
into pieces to be drawn out by a hammer, or sent to the rolling
mill to be formed into marketable bar iron. An admixture.
of fibrous iron, cast-iron, and steel is the result of the above
process. The quality of the iron depends entirely on the
quality of the ore, for there are no opportunities for the exer-
cise of any skill to create improvements in the process.
Poor ores cannot be smelted. In Vermont, where the rich
magnetic ores are employed, a ton of blooms costs about
40 dollars; 4 tons of ore and 300 bushels of charcoal are
required to produce 1 ton of blooms.
The method used in Corsica and on the west coast of
Italy requires more time and fuel than the Catalan process,
as the reduction and melting are not performed in one opera-
tion. According to this method enough ore for four meltings
is roasted and reduced at a time.
Very low Stücköfen are at present used in East India and
Finland. Small cupola furnaces* were formerly used in
Carinthia, &c., the interior in the form of a double crucible.
They were usually from 10 to 16 feet high and 24 inches
wide at the top and bottom, and measuring about 5 feet
at their widest part. There are generally two tuyeres,
both on the same side; the breast is open, but during the
smelting operation it is closed by bricks. The furnace is
heated previous to closing in the breast, after which charcoal
and ore are thrown in; the blast is then turned on.
* Dr. URE's Dictionary of Arts, &c., vol. ii., 692.
As
MAKING WROUGHT-IRON DIRECT FROM THE ORE. 703
T
soon as the ore passes the tuyere, iron is deposited at the
bottom of the hearth; when the cinders rise to the tuyere a
portion is suffered to escape through a hole in the dam; the
tuyeres are generally kept low upon the surface of the
melted iron, which thus becomes whitened; as the iron
rises the tuyeres are raised. In about 24 hours one ton of
iron is deposited at the bottom of the furnace, the blast is
turned off, and the iron, which is in a solid mass, stück or
wolf, as the Germans call it, is loosened from the bottom by
crowbars, taken by a pair of strong tongs, which are fastened
to chains suspended on a swing crane, and then removed to
an anvil, where it is flattened by a tilt-hammer into slabs
4 inches thick, cut into blooms, and, finally, drawn out into
bar iron by smaller hammers. Meanwhile the furnace is
re-charged with ore and coal, and the same process is
renewed.
The following are newer methods for the production of
wrought-iron direct from the ores :-
V. Gersdorff* roasts sparry iron ore in reverberatory fur-
naces, and heats the roasted ore together with coal in
crucibles. Clayt heats ore and coal in a retort, and treats
the reduced iron in a puddling furnace. Renton‡ reduces
the iron ores in vertical slightly heated tubes by means of
carbonic oxide gas, and forms the reduced iron into balls in
a puddling furnace. Chenot|| submits the ores to a reducing
roasting to transform them into magnetic oxide, which he
finely crushes, and by means of an electro-magnetic appa-
ratus extracts the magnetic components; he then reduces
the ore with carbonic oxide gas, grinds the resulting spongy
iron, mixes it with soda, presses it into a cylindrical shape,
and at a suitable temperature draws it out into bars. Rogers
heats the iron together with coal in a rotating cylinder, and
forms the balls in a puddling furnace.
None of these methods seem to have met with any practical
success.
* B. u. h. Ztg., 1843, PP. 320, 577.
HARTM., Fortschr., i., 252.
B. u. h. Ztg., 1856, pp. 133, 180.
TUNNER'S Bericht über d. Pariser Industrie-Ausstellung, 1855, p. 46.
B. u. h. Ztg., 1862, p. 341.
704
IRON.
Gurlt's method for the production of cast-iron by means
of gases (page 424) has been advantageously applied in
Biscay by Clark* for the direct production of wrought-iron
in a cupola furnace (page 332).
PRODUCTION OF WROUGHT-IRON FROM PIG-IRON.
The conversion of pig-iron into wrought-iron mainly con-
sists in the oxidation of most of the carbon contained in the
pig-iron, and in order to render the resulting wrought-iron
tractable both at low and high temperatures the foreign ad-
mixtures (sulphur, phosphorus, silicon, arsenic, &c.,) must
at the same time be separated by oxidation, forming the vola-
tile substances, whilst the fixed substances may be scorified.
The oxidising agents are chiefly the oxygen of the atmo-
sphere and fluxes rich in oxygen (oxidised iron, iron cinders),
steam is seldom used.
The oxidation process may be performed either at a red heat
(Tunner's method), or at the melting temperature (in fineries,
reverberatory furnaces, or in the Bessemer cupola). The
lumps or balls resulting when the melting heat is used are
much mixed with slag; which must be forcibly pressed out
by hammers or suitable apparatus, when the iron will be
ready to be drawn out into bars; but in most cases these
bars still contain slag and do not form a homogeneous mass
of iron, and to make them marketable they must be heated
at the welding heat and drawn out repeatedly; to allow
these repetitions the bars are cut up and formed into a pile,
which is then heated and drawn out.
The process of converting cast-iron into wrought-iron
seems to be very simple in its chemical reactions, but it is,
nevertheless, very difficult in its performance, as it requires
in the manipulator both mechanical skill and an experienced
eye. The process itself requires to be suitably modified
according to the nature of the fuel and pig-iron under treat-
ment; the quality of the pig-iron and fuel also decides the
details of this process.
Allgem. b. u. h. Ztg., 1863, p. 294.
PRODUCTION OF WROUGHT-IRON FROM PIG-IRON. 705
The following are the chief methods:-
1. The Production of Wrought-Iron at a Glowing Red
Heat. In 1828, Karsten proposed for this purpose to submit
thin pigs to a glowing heat in a continued current of air.
Tunner in his method, published in 1846, proposes to produce
malleable iron by exposing bars of cast-iron 7 or 9 lines thick
for from 15 to 35 days to a glowing red heat in boxes filled with
quartz sand, allowing a regular admission of air. Steel or
malleable iron may be produced, according to the length of
the operation, and the application of substances giving off more
or less oxygen (peroxide of iron, oxide of zinc, brown stone);
the resulting product is also known by the name of "malle-
able cast-iron." This iron lends itself well to the drawing
out, the piling, and the welding operations. The following
analyses by Gottlieb and Richter prove that not only is the
greater part of the carbon and sulphur removed as carbonic
oxide and sulphurous acid respectively, but also great part
of the silicon and manganese, as both become oxidised and
form silicate of manganese, which liquates from the porous
iron; or it is, perhaps, separated in the same manner as the
segregations of manganiferous pig-iron.
Analyses of Malleable Cast-Iron, &c.
C (chemically combined).
C (mechanically mixed)
Si
I.
II.
III.
IV.
0446 2217 0.8552 3′3+0
0434 0'583
}
0*409 0.951 0*2562 Ι'ΟΙΟ
trace trace
P.
Al
trace
S.
0'015
Mn
0'4470
Sand
0'502
V.
VI.
VII.
VIII.
C (chemically combined).
C (mechanically mixed)
:} 1-176
I'201 3'5700 3'420
Si
0'002
0'008 0*1300
ΟΙΙΟ
P.
trace trace
Al
S.
Mn
O'OOI
0.188
o'oogo o'008
O'210
0.6100 0*580
Vd
trace
VOL. II.
2 Z
706
IRON.
No. I is malleable cast-iron from Lancashire, analysed by
Miller; its specific gravity is 7'718. No. 2 is white cast-iron.
from Lancashire, from which No. I has been produced; its
specific gravity is 7.684. No. 3 is Tunner's steel produced
by his method, analysed by Gottlieb; it is produced from
white cast-iron, No. 4. Nos. 5 and 6 are also this steel,
analysed by Richter, produced from radiated cast-iron, Nos.
7 and 8.
·
This method of producing malleable iron is only applicable
to the purest kinds of cast-iron, and therefore is little
used.
2. The Bessemer Process.-Grey iron is used which has
already been purified for a longer time from part of its phos-
phorus, sulphur, silicon, &c. (thus preparing it for the puddling
process), either by making the blast react for a longer time
upon the liquid iron in the crucibles of blast furnaces, or by
treating the liquid iron in refineries. Bessemer, however,
conducts the liquid iron from the blast furnaces into special
apparatus in which he burns the carbon of the pig-iron by blast
and without the application of fuel. The temperature thus
produced, together with that of the burning and liquid iron,
is sufficiently high to liquefy even malleable iron. It is very
difficult to separate the carbon to the required limit, even when
employing a pure pig-iron, on which account a friable burnt
iron, poor in carbon, is likely to result; and this iron will be
red- and cold-short at the same time when treating impure
cast-iron, as phosphorus and sulphur cannot be sufficiently
extracted, owing to the rapidity of the operation. This cir-
cumstance and the great loss of iron are the reasons why
this method has not been generally introduced; but it is
applied most advantageously and extensively in the manu-
facture of steel, when it is more easy to fix the point of
decarbonisation.
Oestlund* has tried to improve Bessemer's method of
producing malleable iron by conducting the blast, together
with carbonic oxide gas, over the surface of the fused pig-
iron and by hammering the iron when in a pasty state.
B. u. h. Ztg., 1861, p. 296; 1860, p. 206.
PRODUCTION OF WROUGHT-IRON FROM PIG-IRON. 707
3. The Production of Malleable Iron in Hearths or
Fineries, using charcoal and blast, which, together with
ferruginous slags, serve as oxidising agents. The compres-
sion and extension of the lumps or balls are effected with
hammers.
4. The Production of Malleable Iron in Reverberatory
Furnaces (puddling process) is effected with raw fuel or fuel
in the form of gas; employing, at the same time, larger
additions of cinders, chiefly with an oxidising reaction.
The balls are compressed by various machines (hammers,
squeezers, rolls, &c.), and usually drawn out by rolls, and, in
very rare cases, under hammers.
5. Combined Reverberatory and Finery Process.-
Sometimes the process in reverberatory furnaces is combined
with the process in fineries, in this manner :-The pig-iron
is first treated in puddling furnaces so as to be brought
more or less to the state of malleability, and this product is
completely converted into malleable iron by a subsequent
treatment in fineries.
The methods mentioned under 3 and 4 are those most
commonly used, and we will therefore describe them more in
extenso. A comparison of both methods gives the following
results:
1. When working an impure iron a better product may be
obtained by the puddling process than by treating the iron
in fineries, supposing the puddling process to be suitably
modified, as the latter lies open to the eye and its regulation.
is not limited to the sense of touch; furthermore, added
fluxes are more effective, and a reducing and oxidising reac-
tion can more easily be employed when required.
The process in fineries depends entirely on the diligence
and skill of the workman, and his manipulations are less.
under control than those of the puddler.
2. The puddling process combined with rolling mills
admits of a larger and cheaper production, as pig-iron of
inferior quality (coke-iron), raw fuel, or fuel in the form
of gas may be employed, and also less fuel is required by
this process, than charcoal for the finery process. Another
advantage is that the process takes a shorter time than the
2 Z 2
708
IRON.
finery process. In the finery process 100 lbs. of wrought-iron
are produced from about 139 lbs. of cast-iron with a con-
sumption of from 100 to 133 lbs. of charcoal. In the puddling
process from 125 to 140 lbs. of cast-iron yield 100 lbs. of
wrought-iron with a consumption of about 100 lbs. of mineral
coal, and a puddling furnace treats about three times as
much pig-iron as a finery hearth.
3. As the puddling process is carried on at a lower tem-
perature but with a larger quantity of slag than the finery
process, it requires stronger apparatus and a repeated heating
to compress the balls, and hammers are more effective for
this purpose than rolls or other contrivances. Corbin Des-
boissières* attributes the superior quality of the iron produced
in fineries to the higher temperature employed, as causing
a more perfect welding of the particles of iron and thus
yielding a more homogeneous metal. When employing
hammers for compressing the balls, and rendering them homo-
geneous, it is of no consequence whether the subsequent
drawing out of the balls is effected by rolls or by hammers,
as both contrivances then yield an equally good product ;
rolls, however, turn out a much larger quantity, and their
working requires less skill than that of hammers, and there-
fore they are preferable.
Tunner states that the working of hammers requires two
or three times as many workmen, four or five times as much
motive power, and two or three times as many repairs as
rolls, supposing the production to be the same in quantity in
both cases.
4. The finery process is sometimes preferred to the puddling
process when a large and cheap production is not so desirable
as the production of a superior wrought-iron, for the chief
reason that the finery fires treat in the first instance a purer
cast-iron and melt it gradually, so that it drops down before
the tuyere; less cinders are also employed, and the iron is
repeatedly brought before the tuyere and worked up whilst
employing a higher temperature, thus more perfectly sepa-
rating the foreign admixtures; the iron in puddling furnaces,
* B. u. h. Ztg., 1859, p. 13.
PRODUCTION OF WROUGHT-IRON FROM PIG-IRON. 709
on the other hand, is more porous, rich in cinder and slag,
and the intermixed cinders can only be expelled by a re-
peated heating and mechanical treatment.
The iron resulting in fineries is more compact and homo-
geneous, owing to its smaller amount of slag, and its drawing
out by hammers admits of the observation and correction of
the faulty places in the bars. Formerly the finery iron excelled
the puddling iron by its granular texture and greater hardness,
but fine-grained iron is now also produced in puddling fur-
The wrought-iron from impure cast-iron containing
phosphorus and sulphur is not so good when produced in
fineries as in puddling furnaces, as the direct contact with
coal facilitates the reduction of the scorified components.
Local circumstances have led to the following combined
contrivances for the production of wrought-iron.
naces.
1. Finery Fires and Hammers, using the same fire for
the conversion and the re-heating of the iron, thus saving
charcoal (Oberhartz).
2. Finery Fires, Hollow Fires, and Hammers.-The
conversion and re-heating of the iron is performed with
charcoal in countries abounding in wood, in Polonia for ex-
ample; in order to economise the charcoal, the re-heating
operation is usually performed with coke (Pontypool, Ystali-
fera, Neustadt-Eberswalde, Rybnik), or with coal in hollow
fires (Upper Silesia, Thuringia, formerly also in Siegen and
in the Mark). The better the quality of the pig-iron the
better will be the resulting wrought-iron (Siegen, Thuringia),
whilst an inferior cast-iron (Upper Silesia) is better adapted
for the puddling process.
Finery Fires, Re-heating Furnaces and Hammers (sel-
dom rolls) for a larger production and a wrought-iron of a
good quality. The re-heating furnace is heated with mineral
coal (Rybnik in Silesia, Erlahammer in Saxony), with brown
coal (Styria), with charcoal or gases from wood and turf
(Eckmann's furnaces in Sweden), or with the waste gases of
the fineries (Donnersbach in Styria).
4. Puddling Furnaces, Hollow Fires, and Hammers,
chiefly employed in countries where the want of charcoal has
led to the replacement of the fineries by puddling furnaces;
710
IRON.
the comparatively low production necessitating the pre-
servation of the hammers (Siegen, Nassau, Eifel, Thuringia,
Königshuld, Ilsenburg, Upper Hartz). The puddling fur-
naces are fed either with compact fuel or with fuel in the
form of gas, the hollow fires either with coal or with coke.
With regard to the quality of the wrought-iron, authorities.
are divided as to whether it is preferable to convert the iron
with charcoal in fineries and to re-heat the balls with coke
in hollow fires, or to puddle the iron and re-heat it with
charcoal (Blechhammer in Upper Silesia).
5. Puddling Furnaces, Re-heating Furnaces, and
Rolling Mills, employed in establishments of large produc-
tion, which are situated near the collieries, employing steam
as the motive power and working a cheap coke pig-iron
(England, Belgium, France, Westphalia, &c.). Brown coal
(Prävali in Carinthia, Maximilianshütte near Regensburg,
Hachenburg in Nassau) and turf (Kallich in Bohemia) are
sometimes used instead of coal; when turf is used charcoal
pig-iron is also employed.
To save fuel or to make a superior quality of wrought-
iron the finery and puddling processes are combined in
different manners, especially in Upper Silesia, where the
cast-iron, without refining it, is first puddled, the resulting
balls are compressed under hammers and treated again in
fineries to render them perfectly malleable. In this manner,
at Rybnik for example, a superior iron at less loss was ob-
tained, at a saving of 42 per cent of charcoal.
I. PRODUCTION OF WROUGHT-IRON IN HEARTH
FINERIES OR OPEN FIRES.
The finery process is carried on with various modifications,
according to the quality of the pig-iron, constituting so many
different methods, which we will describe after reporting
upon the materials and apparatus used in the process.
Materials used in the Finery Process.
a. Cast-Iron.-Those kinds of cast-iron which easily lose
their carbon in the finery process are termed in German
THE FINERY PROCESS.
711
gaarschmelzig, and the other kinds rohschmelzig. The amount
of graphite in the iron has the greatest influence in this
behaviour of the cast-iron, graphite being slightly com-
bustible; and the iron becomes, when fused, more or less
thinly liquid, and is thus exposed to the influence of the
oxidising agents for a longer or shorter time. Certain pre-
paratory processes, chiefly with the object of effecting a
strong oxidation of the foreign components and the trans-
formation of the graphite into chemically combined carbon,
permit the improvement of impure rohschmelzige kinds of cast-
iron so as to make it more fit for the finery process; the ob-
jection to these processes is that they involve extra expense.
But the gaarschmelzige iron, i.e., iron which comes quickly
to nature, is not always the best. The more impure the iron
the longer the finery process must be carried on for a
thorough removal of the impurities. Other circumstances,
such as a suitable construction of the finery hearth, the in-
clination of the tuyeres, the application of suitable fluxes,
&c., also tend to prolong the process.
1. White Cellular Iron (pages 276 and 561).—Owing to
its small amount of chemically combined carbon and its
pasty state when fused, it is quickly converted into wrought-
iron; it is also very pure, and contains a little manganese,
facilitating the process. The finery process takes place in
the shortest time when working the "gekrausten flossen"
(page 562), which are very poor in carbon, and may be added
to the finery process at its first period, or when employing
hot blast, supposing them to be very pure. This iron has a
granular texture, great strength, and has many holes of a
dull appearance; "kleinluckige flossen," also with a radiated
texture and a larger amount of carbon, refine somewhat
slower. Grossluckige flossen" have a still larger amount of
carbon. The two latter kinds of cast-iron are the best ma-
terial for the production of malleable iron, and grossluckige
flossen are employed for the production of steel as they do not
come too quickly to nature. Some kinds of iron from the
preparatory process have the nature of the cellular iron.
b. Flowery or Radiated White Iron (pages 276 and 561)
is richer in carbon than the preceding kinds; it is less easily
712
IRON.
fined, and is more used in the manufacture of steel than
of malleable iron.
c. Spiegeleisen (pages 269 and 569).—Owing to the larger
amount of chemically combined carbon which it contains,
and its thin liquidity when melted, spiegeleisen takes longer
to refine than the preceding kinds, but it fines more quickly
than grey graphitic pig-iron. It is very pure, containing
manganese, and therefore yielding an excellent malleable
iron. But as spiegeleisen is too expensive, it is generally used
in the finery process together with cheaper sorts of iron, or
it is worked into steel. Spiegeleisen is usually more impure
the further it is in colour from silver-white.
d. White Iron of the Regular Process (pages 279 and
562). It melts in a more or less pasty state according to its
amount of carbon, but it is less pure than the iron we have
already described, and therefore only applicable in admixture
with grey cast-iron, or when the finery process is artificially
prolonged. Owing to its amount of chemically combined
carbon, this iron fines more quickly than grey iron. A cer-
tain amount of manganese is desirable.
e. White Cast-Iron of the Irregular Process (pages
280 and 563) is more impure than the preceding and than
grey cast-iron. The rapidity of its fining depends on the
amount of carbon present, and the more quickly it fines the
less applicable it is, as then no time is allowed for the sepa-
ration of impurities. This iron cannot always be sufficiently
purified by a preparatory treatment, which also causes loss
of time, iron, and increases the consumption of fuel.
This inferior white iron is usually of a dirty greyish white
colour, and a more granular or scaly texture, whilst the good
kinds of white iron are silver-white in colour and of a radiated
crystalline texture.
f. White Chilled Iron of Radiated Texture produced
by chilling grey iron, contains the whole of the carbon of the
grey iron, but mostly in a chemically combined state. It
therefore requires for fining a longer time than the radiated
white cast-iron which has been produced direct in the blast
furnace, and contains a smaller amount of carbon. Only
the mottled and grey kinds of cast-iron produced from pure
MATERIALS USED IN THE FINERY PROCESS. 713
easily fusible ores yield, when chilled, the artificial white
radiated iron; whilst grey iron produced from ore mixtures
difficult to fuse becomes lighter, and not radiated in texture
when chilled.
g. The Varieties of Grey Iron (pages 282 and 551).—
Owing to the presence of graphite and their thin liquidity,
these require more time for fining, and cause more loss of
metal than the white. irons. This behaviour is desirable
if the grey iron also contains other impurities, but it is not
desirable if the iron is free from sulphur and phosphorus,
and does not contain too much silicon (page 299). Pig-iron
rich in silicon, chiefly the darkish grey kind produced with
hot blast* from siliceous admixtures, requires a high tempera-
ture for melting. It then becomes very thinly liquid, and
takes a long time to fine, causing a great loss of fuel, time,
and iron,† as one part of silicon scorifies about six times its
weight of iron.
The state of combination of the silicon seems to have con-
siderable influence upon the finery process. Richtert has
called attention to the difficult combustibility of the crystal-
lised silicon contained in pig-iron, and Schafhäutl|| to that
of carbide and of nitro-carbide of silicon.
A larger amount of silicon may be desirable in iron con-
taining much sulphur and phosphorus, as it prolongs the
finery process, thus allowing the removal of those impurities;
it may finally also be separated.
In comparison with the varieties of white cast-iron rich
in carbon the grey iron is softer and stronger. It is more
difficult to break, as it consists of a white iron poor in carbon
with an admixture of graphite. Usually grey iron is better
the stronger it is, as its strength decreases when it combines
with certain substances (silicon, earthy metals, &c.) Its
darkish grey colour with strong lustre then assumes more of
an ashen grey colour with an earthy dull texture. But this
rule also has its exceptions, as, for example, a cast-iron with
*
Bgwkfd., iv., 145, 420.
↑ Oesterr. Ztschr., 1853, p. 376.
+
B. u. h. Ztg., 1862, p. 320.
|| Ibid., 1861, p. 38.
714
IRON.
a small amount of sulphur and copper, which do not impair
its strength, will yield a red-short wrought-iron. Very grey
iron with a considerable amount of graphite, which greatly
weakens the iron, may be freer from other noxious substances
than a stronger iron containing less graphite, as generally
the quantity and the size of the graphite scales are larger
the more easily fusible the ores, the higher the temperature
in the blast furnace, and the more slowly the iron has been
cooled.
Preparation of the Pig-Iron.
White cast-iron rich in carbon and impure grey iron (for
instance, coke pig-iron) are frequently submitted to a prepara-
tory process, if such an expensive operation cannot be avoided
by a suitable modification of the finery or puddling process.
The preparation of the cast-iron has for its object the oxi-
dation of part of the carbon or foreign substances (silicon,
phosphorus, sulphur, &c.,) and this may be effected in dif-
ferent ways, either in the crucible of the iron blast furnace,
or in special apparatus.
1. Preparation of the Cast-Iron in Blast Furnaces.
a. By feeding* the furnace through the tuyeres with iron
cinders or iron ores (page 546).
b. By conducting the blast† upon the mottled or radiated
iron in the crucible of the blast furnace, the crucible being
filled with iron up to the tuyere, and the slag being tapped
off. A boiling slag will then be produced, causing the for-
mation of cellular (luckiges) iron, which is then tapped off.
This operation wastes the furnace hearth considerably, and
Kelly has improved upon it.
c. By forming the cast-iron which has been tapped off into
a pit into thin plates by throwing water upon it, each time
removing the solidified iron on the surface (page 544).
d. The process of granulating the iron is still more effec-
tive, but it is seldom used.
*
KARST., Arch., 1 R., xiii., 207.
† Ibid., 1 R., vii., 14.
+ B. u. h. Ztg., 1860, p. 11I.
PREPARATION OF THE PIG IRON.
715
2. Raising the Pig-Iron to a Glowing Heat.-The pig-
iron being cast into discs or plates of an inch thick, and
cooled with water, is raised to a glowing heat either in
special apparatus or in a hearth joined to and forming the
flue of a finery hearth, whilst introducing a little blast. The
iron is placed between small coal; the temperature is regu-
lated by the small coal with which the iron is covered, and
the blast must be very limited, as otherwise too much iron
scale will be formed. From 30 to 100 cwts. require from
12 to 36 hours for this operation.
The heated pieces of iron consist essentially of three layers,
namely, one of iron scale, one of more or less malleable white
iron, and a core of grey iron; the lamellar texture has given
place to a granular aggregation. In Carinthia, where they
use this process, the carbon is transformed into carbonic
oxide, sulphur into sulphurous acid, and manganese and
silicon into an easily fusible silicate which liquates from the
iron. Pure white or mottled iron is usually submitted to
this process to lessen its amount of carbon. The layer of
scale formed in the operation facilitates the removal of carbon
in the finery process.
3. Refining the Iron by Treating it in the Liquid State
in Hearths or Finery Fires.-These methods of purifying
or refining the iron are more effective than the preceding
methods. As, however, the iron is usually melted in the
refineries whilst in contact with coal, a reduction of oxidised
and scorified foreign substances (phosphorus and sulphur,
but not silica) is likely to take place, and on this account
reverberatory furnaces are preferable for this purpose. These
furnaces also enable the process of oxidation to be carried
on for a longer time. The refining process is sometimes
carried out on hot fluid iron run direct from the hearth of
the blast furnace; this method allows a considerable saving
of time and fuel.
Analyses have proved that upon re-melting (refining) the
iron in hearths silicon is first oxidised, and afterwards man-
ganese and phosphorus; the amount of sulphur in the iron
decreases very little and may even be increased by the
sulphur of the fuel, as it becomes but slightly oxidised. One
716
IRON.
of the chief reactions of the refining process is the conversion
of the graphite into chemically combined carbon; this,
according to Drassdo,* is caused by the high temperature of
the process, and by the separation of the silicon. It seems
that chemically combined carbon and silicon exclude each
other, as cast-iron is of a deeper grey (and therefore richer in
graphite) the more silicon 'it contains, and, on the other
hand, it is whiter the less silicon it contains. The separa-
tion of the silicon in the refining process increases the
affinity of the carbon to the iron chiefly at high temperatures,
and the graphite becomes transformed into chemically com-
bined carbon. A stronger reaction of the blast increases the
purification of the iron, and a decarbonisation also takes
place by an addition of slags rich in oxygen. The direct
action of the air separates little or no carbon; any separa-
tion is effected by substances containing oxygen, chiefly by
oxide of iron, which is formed in large quantities by the
prolonged action of the blast.
Various plans have been patentedt for the employment of
fluxes to assist the removal of the impurities of cast-iron,
both in the refining and puddling furnaces. Thus, in 1855,
Mr. Hampton patented a flux prepared by slaking quick-lime
with the solution of an alkali or alkaline salt. In a patent
secured in 1856, MM. du Motay and Fontaine propose to
purify and decarbonise iron in the refining and puddling fur-
nace by using fluxes prepared from the scoria of the
puddling furnace, from oxides of iron and silicates or carbo-
nates of alkalies or other bases. Mr. Pope (1856) proposes
to add the residue obtained by the distillation of Boghead or
Torbane mineral to such fuel as is employed in the
refining of iron. Mr. Sanderson, of Sheffield (1855), em-
ployed for the refining of iron such substances as sulphate
of iron, capable of disengaging oxygen or other elements,
which will act upon the silicon, aluminium, &c., con-
tained in the metal. These and various other schemes.
have been suggested with the object of lessening the
enormous waste which pig-iron undergoes on its passage
* Preuss. Ztschr., xi., 170.
URE'S Dictionary of Arts, &c., vol. ii., 728.
REFINING THE PIG-IRON.
717
through the refinery; for, as the process is at present
conducted, the partial elimination of the carbon, sulphur,
phosphorus, &c., is only effected at the expense of a
large quantity of iron, which is oxidised by the blast and
passes into the slag in the form of silicate; the desideratum
is the discovery of some method of reducing the oxide of
iron and substituting for it some other base, which will form
with silica a sufficiently fusible silicate.
Mr. Blackwell suggests that the decarburation of pig-iron
may be effected by re-melting it in a cupola furnace, either
alone or with minerals containing nearly pure oxides of iron;
the oxide of iron would be reduced by the carbon of the pig-
iron, while the silicate of the fuel, with the silica, alumina,
and other easily oxidisable alloys eliminated from the cast-
iron, would be separated in the form of a fusible earthy glass.
The employment of steam as a purifying agent for cast-
iron has been patented by several persons. In 1854, Mr.
Nasmyth obtained a patent for the treatment of iron in the
puddling furnace with a current of steam, which, being intro-
duced into the lower part of the iron, passes upwards, and
meeting with the highly heated metal undergoes decomposi-
tion, both elements acting as purifying agents. The steam
employed is at a pressure of about 5 lbs. per square inch,
and passes into the metal through a species of hollow rabble,
the workman moving this about in the fused metal until the
mass begins to thicken, which occurs in from five to eight
minutes after the introduction of the steam; the steam-pipe
is then removed and the puddling finished as usual. The
advantages are said to consist in the time saved at each.
heating or puddling operation (from ten to fifteen minutes),
the very effective purification of the metal, and the possibility
of treating highly carbonised pig-iron at once in the puddling
furnace, thus avoiding the preliminary refining. In October,
1855, Mr. Bessemer patented a somewhat similar process.
for the conversion of iron into steel; highly heated steam,
or a mixture of air and steam, being forced through the
liquid iron run from the furnace into skittle pots; steam was
used only at an early stage of the process, and the treatment
finished with heated air.
718
IRON.
In the early part of the same year Mr. Martien, of New
Jersey, obtained a patent for a partial purification of cast-
iron, by causing air or steam to pass up through the liquid
metal as it flows along gutters from the tap-hole of the fur-
nace or finery forge; and he subsequently proposed to include
with the air or steam other purifying agents, such as
chlorine, hydrogen, coal gas, oxides of manganese and
zinc, &c. Other methods of treating cast-iron with air and
steam were patented by Mr. Bessemer in December, 1855,
and January, 1856. In October a patent for the employment
of steam in admixture with cold blast in the smelting fur-
nace and fining forge was taken out by Messrs. Armitage
and Lee, of Leeds, and, in August, a patent was taken out
by Mr. George Parry, of the Ebbw Vale Iron Works, for the
purification of iron by means of highly heated steam. The
fluid iron is allowed to run into a reverberatory furnace pre-
viously heated, and the steam is made to impinge upon it
from several tuyeres, or to pass through the metal.
Steel is to be obtained by treating highly carburetted iron
with the steam and then running it into water, and fusing it
with an addition of purifying agents, or adding to it in the
furnace a small quantity of clay and afterwards about 10 or
15 per cent of calcined spathose ore.
Mr. Parry observed that when steam was sent through the
molten iron, as in Mr. Nasmyth's process, the iron quickly
solidified, and he conceived the idea of communicating a
high degree of heat to the steam by raising the steam-pipe
a couple of inches above the surface of the metal, so that it
might be exposed to the intensely heated atmosphere of the
furnace; he also inclined the jet at an angle of 45° so as to
give the molten mass a motion round the furnace while the
pipe was maintained in the same position at a little distance
beyond the centre; in a few minutes the iron began to boil
violently, the rotatory motion of the fluid bringing every
part of it successively into contact with the highly heated
mixture of steam and atmospheric air, and no solidification
took place. Having thus ascertained the proper way of using
steam as a refining agent, it occurred to Mr. Parry that, as
the presence of silicon in the pigs for puddling affects to a
REFINING THE PIG-IRON.
719
remarkable degree the yield of iron as well as the strength,
the silicon should be removed as completely as possible pre-
vious to the puddling operation; the steaming of the iron
would probably therefore be more profitably applied in the
refinery than in the puddling furnace. Pig-iron containing
3 per cent of silicon gives 6 per cent of silica, which would
require from 10 to 12 per cent of iron to form a cinder suffi-
ciently fluid to allow the balling up of the iron; and this can,
of course, only be obtained by burning that amount of iron
in the puddling furnace after the expulsion of the carbon
and while the mass is in a porous state. The superheated
steam is injected on the surface of the iron, in the refinery
by water tuyeres, similar to those used for hot blast in
smelting furnaces; they are inclined at an angle of about
45°; some are inserted at each side of the door of the fur-
nace, and are pointed so as to cross each other and give the
iron a circulating motion in the furnace. The tuyeres are
from to an inch in diameter; a little oxide of iron or sili-
cate in a state of fusion on the surface of the iron accelerates
the action, as in common refineries, and increases the yield
of metal, but to a much greater extent than when blast of
air is used. The steam having been turned on, the mass of
iron begins to circulate around the inclined tuyeres and soon
boils, and the action is kept uniform by regulating the flow
of the steam. The most impure oxides of iron may be used
in this process, such as tap cinder or hammer slag from
puddling furnaces, without injury to the quality of the refined
metal; the large quantities of sulphur and phosphorus which
they contain being effectually removed by the detergent
action of the heated steam. When 4 cwts. of cinders are used
to the ton of pig, I ton of metal may be drawn, the impurities
in the pig being replaced by refined iron from the cinders.
The following analyses of the cinders and metal fully bear
out these statements :-
Graphite
Silicon
Slag.
•
Pig-Iron.
Refined Metal.
2'40
0°30
2.68
0°32
0'68
0'00
Sulphur
0'22
O'IS
Phosphorus
0°13
0'09
Manganese.
0.86
0°24
720
IRON.
Sulphur
Phosphoric Acid .
Forge Cinders
thrown into
the Refinery.
I'34
2.06
Cinders run
out of the
Refinery.
0'16
O'129
A ton of grey iron may be refined in half an hour, using
seven jets of steam of an inch in diameter, and with a
pressure of from 30 to 40 lbs. ; the temperature of the steam
is from 600° to 700° F., and the orifices of the tuyeres are
2 or 3 inches above the surface of the iron. As the fluidity
of the metal depends upon the heat which it receives from
the combustion of the fuel in the grate and not on the heat
generated in it by the action of the steam, it is evident that
the supply of steam must not exceed a certain limit in a
given time, or the temperature of the fluid iron will become
reduced below that of the furnace. This, however, regulates.
itself and does not require much nicety in the manage-
ment, for if too much steam is given the ebullition becomes
so violent as to cause the cinders to flow over the bridges,
warning the refiner to slacken his blast. The "forge cinders"
used in the steam refinery contain 66 per cent of iron, the
"run out cinder" contains only 26 per cent, therefore, about
40 per cent of iron is converted into refined metal, and the
resulting cinder is as pure as the ordinary Welsh mine with
its yield of 25 per cent of iron. The following is the result
of one week's work of the steam refinery :—
Pigs used.
cwts.
qrs. lbs.
Metal made
Loss
Yield
396
0 15
393
3
I
2 I
14
20
14
The quantity of cinder (puddling) used was 3 cwts. per
ton of pig. When 1 cwts. of cinders were used to 1 ton of pig
the yield was invariably 20 cwts. over a make of about 100 tons.
The refinery fire most frequently used is represented in
Figs. 207 and 208, and consists of a rectangular hearth, the
bottom, s, is formed of sand or slag resting upon the
refractory brick-work, s. Three sides of the hearth are en-
closed by cast-iron hollow troughs, E, through which a current
REFINING THE PIG-IRON.
721
of cold water is made to pass. The front plate, F, closes
the hearth and contains the tapping hole, o, conducting the
refinery products into the gutters, x and y. f is the working
plate, g is a side plate with notches bearing the iron bar, h, h',
for supporting the iron tool used in opening the tapping
FIG. 207.

P
L
M
TH
HO
hole. H are iron plates with openings for the tuyeres. The
plates, L, enclose the uppermost part of the hearth, and are
fixed to the pillars, M, which support the chimney. P are
cast-iron bearings resting upon the pillars, M, and bearing
the chimney. E' are cast-iron troughs filled with water for
cooling the working tools. Tare the tuyeres cooled with
water from the troughs, N. 9 is a wind box, distributing the
blast in the tuyeres.
The fires have generally four, sometimes only two tuyeres,
3 A
VOL. II.
722
IRON.
+
and in rare cases but one. The semicircular tuyeres have an
inclination of from 25° to 35°, according to the impurity of
the iron, and project 5 or 6 inches into the fire; they are
1 or 2 inches wide and about 1 inch high, whilst the blast
pipes are 1 or 1 inches in diameter. The refinery fires
I
I
M
FIG. 208.

Q
C
ERBUR
h
E
H
H
IN 12 6 0
لسيليسيا
x
0
2
L
M
To
У
GO:00121
SIENT
MEINDER
4
5
FT
D
are usually 4 feet long, 3*17 feet broad, and from 8 to 12 inches
deep. Small fires with two or three tuyeres require 600 cubic
feet of blast per minute, larger fires with from four to eight
tuyeres, 800, 1000, or 1200 cubic feet. The pressure of the
blast is from 1 to 2 lbs. per square inch.
The refractory bottom of the refinery fire is next covered
with a layer of quartz-sand 3 inches thick, and then
with small coal. The refinery is filled with coke, upon
which the pigs are placed above the tuyeres, with an addi-
tion of cinders. The smelting is then commenced, and
enough cast-iron is subsequently charged to fill the refinery
with iron up to the tuyeres; the blast is made to react upon
the surface of the fluid iron, the duration of the reaction
varying. As the iron becomes more refractory as its amount
of carbon diminishes,, the end of the process is seen by the
REFINING THE PIG-IRON.
723
consistence of the metal, which, when tapped off, throws thick
sparks without flame, and is either of a radiated texture or
slightly cellular. The refined iron together with the slag is
usually tapped into cast-iron moulds, which are brushed over
with clay cr lime water, and sometimes direct into finery
fires. The slag contains from 60 to 70 per cent of iron, is
thinly liquid, light red in colour, and, towards the end of the
process, adheres less to the iron tools with which the iron is
sometimes stirred. The following analyses shows the com-
position of some of these slags :-
I.
II.
III.
IV.
SiO 3
27.6
32°2
13.69
22.76
FeO
61.2 66°5
73°12
61.28
MnO .
0°9
3'58
Al2O3 ·
4°0
7°30
Fe₂03
13.09
PO,
7°2
I'7
Not
determined.
CaO
MgO
S.
3°41
0'76
0.46
No. 3,
No. I is a slag from Dudley, near Birmingham, analysed
by Berthier. No. 2, slag from Firmy, by the same.
slag from Stourbridge, by Rammelsberg-
2(6FeO,SiO3)+FeO,Fe₂O₂
or 3FeO (SiO,,Fe₂O3)+6FeO(SiO3, Fe₂O3),
an aggregation of regular octahedrons. No. 4, slag from
Bromford, near Birmingham, by Percy.
The usual charges of refineries are from 20 to 25 cwts. of
iron, which require about 2 hours for melting, and for 1 or 1½
hour the blast is made to react upon the fluid iron; the loss
of iron in the process is from 8 to 14 per cent. 100 lbs. of
refined iron consume from 40 to 50 lbs. of coke which lies
loosely, and from 60 to 65 of compact lying coke. Fuel is
economised by running the liquid iron from the smelting
furnaces direct into the refineries, or by heating the cast-
iron by the waste gases of the refinery before charging it;
this is more effective than the application of hot blast.
Cassel's and Morton's method of previously melting the iron.
with an addition of slag in cupola furnaces and then running
3 A 2
724
IRON.
it into the refineries also economises fuel. Clay proposes
to introduce the iron into the refineries in a very thin stream.
A modification of this refinery process is that (Hartzer-
rennen) used in Carinthia, chiefly for iron intended for the
manufacture of steel. Grey or mottled pig-iron, with an
addition of oxidising fluxes, is melted in a hearth lined with
small coal (Lösch), charcoal being used as fuel. The slag
is removed from the fluid iron, and, according to its nature,
(liquidity, colour, &c.), oxidising fluxes, from the size of a nut
to the size of the fist, are immersed in the iron by means of
a wooden pole. Water is then thrown upon the metal, and
the chilled crust stripped off. From six to eight thin plates
or discs are thus obtained from a melting of 5 cwts. of iron.
These plates are sometimes submitted to the process of
'braten' (page 714). 100 lbs. of refined iron consume from
4 to 6 cubic feet of charcoal.
At Mariazell, a refinery fire with two tuyeres is used for
the operation of "Hartzerrennen." It is provided with a
hearth for the previous warming of the pig-iron, &c., and its
construction is shown in Figs. 209 and 210.
FIG. 209.

D
C
B
IN: 12
Eututului
h
Ρ
A d
Ъ
3
5
ப
0
J FI
A is the refinery hearth; B, the hearth for heating the pig-
iron; and the space, c, is used for heating wrought-iron
balls of the finery forges for the purpose of hammering them;
m
REFINING THE PIG-IRON.
725
D is the chimney provided with a damper.
a is the bottom
formed of fire-bricks, upon which a layer of quartz sand, b,
3 inches thick, is beaten down; c, d, and e are cast-iron.
troughs, through which a stream of cold water is made to
circulate, to prevent their being fused by the heat. Together
with the front stone, f, they form the enclosure of the hearth.
The stone, ƒ, is provided with the tapping-hole, g; h is a
FIG. 210.

ૐ
P
Ъ
q
B.
e
}
0
cast-iron plate containing openings for receiving the tuyeres,
u, o; i and k are iron plates similar to h, and enclosing the
hearth above the tuyeres; is an iron plate suspended by a
hook, m, in the front opening of the hearth; n is refractory
brickwork; are cast-iron bearings supporting the brick-
work, n; the plates, h, i, k, are screwed to them; q, q are
side plates with grappling irons, enclosing the brickwork, n.
The fire is 32 inches broad, 36 inches long, and 9 or 10
inches deep from the tuyere to the sand bottom. The tuyeres
are 9 inches distant from each other, 1 inches wide, and
1 inches high, projecting 6 inches in the fire, with an incli-
nation of 15°. The blast-pipe is 1 inch in diameter; the
pressure of the blast is 18 inches of water.
From 80 to 90 cwts. of refined iron are produced in 24 hours
at a loss of from 3 to 9 per cent of iron, according to the
quantity of cinders (of the finery process) added. The con-
sumption of charcoal amounts to from 4 to 6 cubic feet per
100 lbs. refined iron.
726
IRON.
The double refinery fire provided with a hot blast apparatus,
and shown in Figs. 211, 212, 213, is also used at Mariazell.
FIG. 211.

α
h.
i
a
யாம்
2
4.
6
8
FT
a is the hearth; b are the tuyeres cooled with water; c are
the cast-iron troughs enclosing the hearth and cooled with
water; d is the tapping-hole; e, the front or working side;
f are flues leading into the channel, i; g, the hot blast appa-
ratus; h, the chimney, 54 feet high.
4. Refining in Reverberatory Furnaces with raw fuel
(coal, brown coal, turf, and wood) or gas. The refining by
gas, which is most frequently used in Germany and first
applied in Upper Silesia, is thus described in Ure's "Dic-
tionary of Arts and Mines," vol. ii., 730 :—
"The most simple form of gas reverberatory furnace is that
known as Eck's furnace (Figs. 214 to 224) which is employed
at the Government works of Gleiwitz and Königshütte
(Upper Silesia) for refining iron made on the spot. The
following description and plan of this furnace is extracted
from a report to the Secretary of State for War from the
REFINING THE PIG-IRON.
727
Superintendent of the Royal Gun Factories, Colonel Wilmot,
R.A., and the Chemist of the War Department, Professor
Abel :-
FIG. 212.
JOL
FIG. 213.

Ъ
Z
==
C
ď
d
The gas generator (which replaces the fire-place of the
ordinary reverberatory furnace) is an oblong chamber 3 feet
FIG. 216. FIG. 214.

T
FIG. 215.
728
IRON.
9 inches wide, and the height from the sole to the com-
mencement of the sloping bridge is 6 feet 4 inches. It
tapers slightly towards the top so as to facilitate the descent
of the fuel, which is introduced through a lateral opening
near the top of the generator. Its cubical contents are about
44 feet.
The air necessary for the production of the gas is supplied
by a feeble blast, and enters the generator from the two
openings or tuyeres of a long air chest of iron plate (Figs.
214, 215, 216) fixed at the back of the chamber near the
bottom. The space between the air chest and the sole of
the chambers serves as a receptacle for the slag and ash of
the fuel.
There are openings on the other side of the chamber
opposite the tuyeres, which are generally closed by iron plugs;
they are required when the tuyeres have to be cleaned
out. There is an opening below the air chest through which
fire is introduced into the chamber when the furnace is set
to work, and which is then bricked up, until at the expiration
of about fourteen days it becomes necessary to let the fire
die out, when the slag and ash which have accumulated on
the sole of the chamber are removed through this opening.
The hearth of the furnace is constructed of a somewhat
loamy sand; its general thickness is about 6 inches; its form
is that of a shallow dish, with a slight inclination towards
the tap-hole. The iron is prevented from penetrating
through the hearth by the rapid circulation of cold air below
the fire-bridge and the plate of the hearth.
Figs. 217 and 218 represent the upper oblong air chest
provided with a series of tuyeres, which enter the top of the
furnace just over the fire-bridge at an angle of 30°. The air
forced into the furnace through these tuyeres serves to in-
flame and burn the gases rushing out of the generator, and
the direction of the blast throws the resulting flame down
upon the metal on the hearth in front of the bridge. This
air chest, like the other one, communicates by pipes with
the air accumulator of the neighbouring blast furnace.
The pressure employed is about 4 lbs., but the supply
of air, both to the generator and the inflammable gases,
REFINING THE PIG-IRON.
729
admits of accurate regulation by means of valves in the con-
necting pipes. There is an opening in the arch at both sides
of the furnace, not far from the bridge, into which, at a cer-
FIG. 217.

O
FIG. 218.
tain stage of the operations, tuyeres are introduced (being
placed at an angle of 25°) also connected with the blast
apparatus, and provided with regulating valves.
FIG. 219.
Gas Reverberatory Furnace.-Front View.

The refining process is conducted as follows :-The hearth
of the furnace having been constructed or repaired, a brisk
coal-fire is kindled in the generator through the opening
at the bottom, which is afterwards bricked up. About
730
IRON.
20 cubic feet of coal are then introduced from above, and the
necessary supply of air admitted to the generator through
the lower air chest. When these coals have been thoroughly
FIG. 220.
Longitudinal Section.

ignited, the generator is filled with coals, and a very moderate
supply of air admitted through the tuyeres below (for the
generation of the gas) and those over the bridge (for its com-
bustion) until the furnace is dried, when the supply of air at
both places is increased, so as to raise the hearth to the
temperature necessary for baking it thoroughly, upon which
about 40 cwts. of iron are introduced, the metal being dis-
tributed over the whole hearth as uniformly as possible, and
the size of the pieces being selected with a view to exposing
as much surface as possible to the flame. The fusion of the
charge of metal is effected in about three hours, the coal
REFINING THE PIG-IRON.
731
used amounting to about 33 cubic feet per hour. The gas
generator is always kept filled with coal, and the supply of
air admitted from below is diminished by a regulation of the
valve whenever fresh coal is supplied, as the latter, at first,
FIG. 221
Plan.

C
D
A
B
always yields gas more freely. The arrangement of the
upper row of tuyeres effects the combustion of the gases just
as they pass from the generator on to the hearth. The hottest
portion of the furnace is, of course, near the fire-bridge, i.e.,
where the blast first meets with the gases. During the
melting process the iron is shifted occasionally, so that the
cooler portion near the flue may in its turn be melted without
loss of time. When the iron is ascertained to be thoroughly
732
IRON.
fused, about 5 lbs. of crusted limestone are thrown over its
surface for the purpose of converting the dross which has
separated into a fusible slag. The two side tuyeres are now
introduced into the furnace through the openings above
FIG. 222.
Cross Section at A, B on Plan.

alluded to, the width of the nozzle employed depending upon
the power of the blast used. The air rushing from these
FIG. 223.
Cross Section at E, D on Plan.

REFINING THE PIG-IRON.
733
tuyeres impinges with violence upon the iron, and the two
currents meeting, an eddying motion is imparted to the fused
metal; in a short time the motion produced in the mass is
considerable. The supernatant slag is blown aside by the
FIG. 224.
End View.

of
blast, and the surface of the iron thus exposed undergoes.
refinement, while it changes continually, the temperature of
the whole mass being raised to a full white heat by the action.
of the air. The iron is stirred occasionally in order to insure
a proper change in the metal exposed to the action of the
blast. A shovelful of limestone is occasionally thrown in
(the total quantity used being about 1 per cent of the crude
iron employed). The slag produced is exceedingly fusible,
and is allowed to remain in the furnace until the metal is
tapped, and on cooling it separates completely from it.
The duration of the treatment in this furnace after the
734
IRON.
metal is fused varies from two hours and a half to five hours,
according to the product to be obtained. For the prepara-
tion of perfectly white iron the treatment is carried on for
five hours. A sample is tapped to examine its appearance,
when it is believed to be sufficiently treated.
When the charge is to be withdrawn from the surface,
the side tuyere nearest the tap-hole is withdrawn, so that
the blast from the opposite tuyere may force the metal
towards the hole. The fluid iron, as it flows from the tap-
hole is fully white hot, and perfectly liquid; it chills, how-
ever, very rapidly, and soon solidifies. A few pails of water
are thrown upon those portions of the metal which are not
covered with the slag which flows out of the furnace, the
object being to cool it rapidly, and thus prevent the oxidation
of any quantity of iron. The loss of metal during the treat-
ment is said not to exceed 5 per cent.
The purification the iron undergoes in the gas reverberatory
furnace appears to be confined chiefly to the elimination of
carbon and silicon, the proportion of sulphur and phosphorus.
varying but slightly, as the following analysis shows (Abel):-
•
Silicon
Phosphorus
Sulphur.
Pig-Iron.
Refined Metal.
4.66
0.62
0*56
0'50
0'04
0'03
Nevertheless the iron thus refined is highly esteemed for
all castings which require iron possessing unusual powers of
resistance. Some experiments made to ascertain the com-
parative strain borne by the refined metal, and by the same
metal as obtained from the blast furnace, showed the strength
of the former to be greater by one-half than that of the latter."
When refining iron for the puddling process the operation
is carried on until the metal assumes a radiated or cellular
texture. The longer the fused iron is exposed to the oxi-
dising blast the more perfect is the oxidation of, first, silicon
and manganese together with iron, and afterwards more or
less phosphorus and sulphur. The graphite is transformed
as in refinery fires into chemically combined carbon, and
partly oxidised by the oxide of iron of the slag if the process
FUEL.
735
is continued long enough, thus allowing the fine metal to be
obtained in either the flowery or cellular state.
100 lbs. of pig-iron yield about 90 lbs. of fine metal at a
consumption of 14 cubic feet of coal in lumps, and 1 lb. of
limestone per 100 lbs. of the fine metal.
In the refinery fires the yield of fine metal was somewhat
less, and the consumption of coke 2727 cubic feet per
100 lbs. of fine metal.
b. Fuel.
Charcoal is chiefly used in the finery process, and its
quality greatly influences the result. Compact hard charcoal
(of beech wood, for instance), produces a greater heat than
soft charcoal, and is sometimes employed; it is then necessary
to modify the process, to construct the fires suitably, and to
increase the pressure of the blast. But the less compact soft
charcoal (of pine and fir wood) is preferred, as, owing to its
greater heating power, the hard charcoal induces a quicker
melting of the pig-iron, thus prolonging the finery process.
Hard charcoal also lies more compactly in the finery hearth,
preventing a free passage of the blast, and decrepitates
when quickly raised to a high temperature, &c. Hard char-
coal is desirable when the pig-iron is to be quickly melted
or converted into a thinly fluid state for the separation of
foreign substances. Hard and soft charcoal is sometimes.
applied to the finery process in a certain combination; for
example, hard charcoal is used for welding the forge iron,
whilst the pig-iron is melted with soft charcoal, when both
operations are not carried on at the same time. In the German
method, &c., the molten pig-iron is converted into malleable
iron, and the lump formed by soft charcoal.
Charcoal for the finery process must be dry, well charred,
free from sand, in pieces the size of an egg or the fist, and
kept in store for some time, as age heightens its effect. Fresh
or newly charred coal burns more quickly, is liable to de-
crepitate, produces a higher temperature and is liable to
burn the iron, it must therefore be slightly moistened.
Raw charcoal, i.e., charcoal which has not been well
charred, burns with a flame, and impedes the observation of
736
IRON.
the process, which is allowed by the passage of the flame
through the coals. Coarse soft charcoal requires but little
pressure of the blast, which also comes more easily into con-
tact with the smelting mass, thus promoting the fining.
Other fuel, such as wood*, turff, brown coal‡, mineral
coal,|| peat charcoal,§ charred brown coal,¶ and coke**
has been used with varying success; fuel in the form of gastt
has been also proposed.
In some French iron works‡‡ the application of air- and
kiln-dried wood, either by itself or in admixture with char-
coal, with hot blast of higher pressure has caused a saving of
15 or more per cent of fuel. Wood is less adapted for heating
the lumps to draw them out than for smelting the pig-iron,
and to convert it in the malleable state.
Charred turf is used at Underwiller,|||| consuming 1100
parts of turf for the production of 1000 parts of wrought-iron.
At Boo§§ in Nericia (Sweden) turf is used for heating the
blooms for the purpose of forging; I cwt. of the latter con-
sumes o'42 cwt. of turf containing 2.88 per cent of ash.
A considerable saving of fuel is effected by employing hot
blast and hearths for previously heating the pig-iron; these
hearths are heated with the waste gases of the finery fires.
c. Fluxes and Agents.
These are required in the finery process chiefly for regu-
lating the process, for extracting foreign substances from the
iron, for lessening the loss of iron, and for the treatment of
scrap-iron of different kinds.
1. Fluxes reacting upon the finery process.
Bgwkfd., i., 356. B. u. h. Ztg., 1842, p. 424; 1843, P. 441.
+ LE BLANC, iv., 55.
‡ Bgwkfd., ii., 462.
+
|| KRAUS, Jahrb., 1855. KARST., Arch., 1 R., iii.,
§ Bgwkfd., i., 476.
107.
B. u. h. Ztg., 1847, p. 422.
**
Bgwkfd., viii., 523.
tt DINGL., Bd. 120, p. 272.
‡‡ Allgem., B. u. h. Ztg., 1862, p. 497.
++
B. u. h. Ztg., 1862, p. 264.
§§ Ibid., 1863, p. 167.
CINDERS PRODUCED IN THE FINERY PROCESS.
737
a. Slags or Cinders produced in Finery Processes.*—
According to the relative proportions of oxide of iron which
these contain they are divided into two classes, poor slags
and rich slags. The poor slags are formed in the first opera-
tion of smelting down the pig-iron, and the rich slags in the
later operations.
Rich slags consist essentially of mono-silicate of iron,
3FEO,SiO3, with a certain amount of intermixed magnetic
oxide, and have an oxidising reaction upon the carbon, sul-
phur, silicon, &c., contained in the pig-iron; the peroxide of the
magnetic oxide then losing some of its oxygen, and becoming
converted into protoxide or metallic iron, whilst the protoxide
of the magnetic oxide enters the mono-silicate of iron, trans-
forming it, for instance, into 6FeO,SiO,. The latter com-
pound also has an oxidising reaction, becoming decomposed
thus:
6FeO,SiO3 3FeO,SiO3 + 3Fe + 30.
These slags have an iron-black colour, little lustre, a com-
pact texture, high specific gravity, and are sometimes of
crystalline structure. They are less thinly liquid, and are
whiter in the fluid state than the poor slags of the same pro-
cess; they also solidify more slowly, and when solidified have
a less fused appearance.
Their composition is shown by the following analyses :—
I.
II.
III.
IV.
V.
SiO3.
21'0
19*15
5.6
10*25
20.31
FeO.
75°5
65°12
85'5
77'00
71'10
MnO
3'75
3'00
CaO.
2'4
1'75
0'92
3°5
I 25
O'I
3'00
Fe₂03
10°03
2'01
PO,
4.7
I'75
4'11
C
0'75
No. I is a rich slag from Silbernaal near Clausthal, ana-
lysed by Metzger; formula of the slag-
4(6FeO,SiO3) + 3(2FeO,SiO;) + 3(3 FeO,SiO3).
VOL. II.
B. u. h. Ztg., 1860, p. 449.
3 B
738
IRON.
No. 2 is slag from Sollinger Iron Works (Hanover), analysed
by Lampadius; formula-
3
5(6FeO,SiO3)+2 Fe₂O,,SiO₂ = 7(4FeO,SiO3)+2FeO‚Fe₂O3:
4(6FeO,SiO), + 3FeO,2SiO3 + FeO,SiO3 + 2FeO,Fe₂O3
6FeO(SiO3,Fe₂O3) + 8[3FeO(SiO3,Fe₂O3)].
No. 3 is slag from Torgelow resulting from phosphoric
pig-iron, analysed by Berthier.
Nos. 4 and 5 are slags from Lauchhammer, analysed by
Lampadius.
The average composition of rich slags from the finery pro-
cess may be considered to be—
6FeO,SiO3, and 3(2FeO,SiO3)+8(6FeO,SiO3)+2FeO,Fe₂O3.
The poor slags consist of pure mono-silicate of iron or of
a mixture of mono- and bi-silicate; they have not an oxidising
reaction. The composition of some of these slags is shown
by the following analyses :-
Sio,
FeO
MnO
MgO .
CaO
A12₂O
Fe₂03
KO
PO5
C.
•
2.4
I'O
T
I.
II.
III.
IV.
V.
32°4 32°35 28.0
17.2
40'15
60:2
62'04
70'0
61.3
49'21
2.65
0'5
O'25
I'41
O'I
0°4
2.7
0.8
0*2
0'45
I'04
0°29
trace
16.5
5°22
3°03
0'9
2'0
No. I is a poor crystallised slag from Silbernaal near
Clausthal, analysed by Metzger.
No. 2, a slag from the iron works at Gettelde (Hartz), beauti-
fully crystallised like hyalosiderite, analysed by Walchner.
No. 3 is a slag from Pissos in France, consisting of inter-
woven large crystals in the form of peridote, analysed by
Berthier.
No. 4, a slag from Torgelow in Pomerania, resulting from
a phosphoric pig-iron, analysed by Berthier.
No. 5 is a slag from Lauchhammer, by Berthier.
A comparison of these and other analyses of this kind of
slag shows that these slags have a composition intermediate
between-
3 FeO,SiO3, and 3FeO (SiO,,Fe₂O3) + 3(2FeO) (SiO3,Fe2O3) ;
PURIFYING FLUXES.
739
the former compound is formed at the beginning of the
melting of the pig-iron, and the latter at the end of the opera-
tion. When treating easily fusible pig-iron rich in silicon
even a poor slag of the composition—
may be formed.
2(FeO,SiO3)+2FeO,SiO¸
The richer this slag becomes in oxide of iron, the more
magnetic oxide will intermix with it, the less fusible it is and
the stronger is its oxidising reaction.
Iron scale or hammer slag, formed when annealing
wrought-iron with an admission of air, in forging, &c., has
an action similar to the rich finery slag. According to
Berthier* hammer slag has the composition 4FeO,Fe₂O3•
Numerous experiments on the formation of iron scale have
been made by Mr. Smitht in Dr. Percy's laboratory.
b. Sand and Clay are more frequently employed for
fluxing when producing steel in finery fires than when pro-
ducing wrought-iron. They facilitate the formation of a
thinly liquid slag, similar to the poor slags of the finery
process.
c. Water thrown upon the coal prevents its burning on
the outside, and has an oxidising action on the pig-iron;
upon moistening the small coal at the bottom of the finery
fire, the fining operation is also accelerated; and likewise
when cooling the iron bottom plate by water, which is allowed
to flow below it.
2. Purifying Fluxes.
a. Carbonate of Lime (Limestone).-When working
grey pig-iron, from 2 to 5 per cent of limestone is added, for
the extraction of phosphorus and sulphur. The addition of
limestone is made when the iron is broken up, when the
imperfectly fined masses are raised from the bottom of the
hearth to the tuyere in order to re-subject them to the influ-
ence of the blast. Additions of limestone and repeated
breaking-up of the iron are, at the same time prolonging the
oxidating reaction, the best means of removing phosphorus.‡
KARST., Arch., I R., xi., 433; ix., 508; xiii., 36.
† PERCY'S Metallurgy, on Iron and Steel, 1864, p. 21.
DINGL., Bd. 65, p. 443.
3 B 2
740
IRON.
Fuchs* suggests aluminous limestone as most effective for
this purpose.
Limestone also facilitates the fining process by forming a
more thickly liquid slag, some of which is, however, liable
to remain in the wrought-iron and to deteriorate it.
b. Schafhautl's Powder,† consisting of 3 parts brown-
stone, 6 parts common salt, and I part potters' clay. Upon
heating this mixture a thinly liquid slag (silicate of manga-
nese, alumina, and soda) is formed, which easily separates.
from the iron; oxygen and chlorine are then liberated, and
combine with sulphur, arsenic, silicon, and phosphorus if
present. Manganese (page 311) also combines with sulphur
and silicon, or even with phosphorus. This powder is more
frequently used in puddling furnaces than in finery fires, as
the blast of the fineries is apt to blow it out of the fire, and
it is applied only when the iron is to be fined and broken up.
Some other agents have the same reaction as Schafhäutl's
powder; chloride of iron, for instance, according to
Augustin||; Duclos§ uses chloride of manganese and lime;
and Fontaine¶ recommends alkaline, earthy chlorides, and
some slags of artificial composition.
3. Scraps of different kinds of wrought-iron in small quan-
tities are in some cases added after the melting-down of the
pig-iron, thus accelerating the fining process.
Apparatus Used in the Finery Process.
a. The finery fires or hearths are usually of a rectangular
form and lined with iron plates 2 or 3 inches thick.
In rare
cases they are round and walled up with brickwork. The
front of the hearth is often formed of brasque, which is covered
with an iron plate. At the level of the hearth bottom the
brasque has an aperture and a cast-iron gutter for tapping off
the cinder. When closing the front of the hearth with an
*
KARST., Arch.. 1 R., xv., 3.
+ Bgwkfd., i., 129; ix., 166. DINGL., Bd. 65, p. 154.
+
ERDM., J. f. pr. Ch., 1860, No. 6, p. 344.
|| KARST., Eisenhüttenkunde, iv., 332.
§ Bgwkfd., i., 133.
¶ B. u. h. Ztg., 1857, p. 98.
APPARATUS USED IN THE FINERY PROCESS.
741
iron plate it must also be provided with one or several holes
for running out the cinder. The bottom of the hearth is
usually formed of an iron plate, sometimes of a stone plate,
or of beaten down small coal (Lösch), and occasionally of
rich cinders from the same process. In the latter case the
natural humidity of the ground* affects the finery process, as
the lowest parts of the fused iron are likely to be cooled too
much and the regularity of the process disturbed. Bottom
plates of stone, although they concentrate the heat more
thoroughly, crack more easily than plates of iron. Some-
times an iron plate is placed lowest, and a plate of stone
upon it, or the bottom of the hearth is formed of a cast-iron
box filled with ash. When it is required repeatedly to break
up the iron, the bottom of the finery hearth is usually formed
of an iron plate with the middle (where it is mostly exposed
to high temperature) hollow, allowing it to be cooled from
below either by water or cold air.
The hearth is surrounded with brickwork from 6 to 10 feet
long, 3 or 4 feet broad, and 12 or 16 inches high. This
brickwork is covered with iron plates, and serves to keep the
coals upon it, &c.
FIG. 225.

X
The general construction of the finery hearths is to be
seen in Figs. 225 and 226, which represent one of the
hearths common in the Hartz.
Bgwkfd., v., 366.
742
IRON.
The illustration is taken from the Mandelholz Works, in
the neighbourhood of Elbingerode. Fig. 226 is an elevation
of this forge. D is the finery hearth, provided with two pairs
FIG. 226.

b
D
C
of bellows. Fig. 225 is a vertical section, showing the con-
struction of the hearth in the finery forge. c is an overshot
water-wheel, which gives an alternate impulse to the two
bellows, a, b, by means of the revolving shaft, c, and the
cams or tappets, d, f, e, g.
The hearth, D, is lined with iron plates. Through the
pipe, l, cold water may be introduced under the bottom
plate, m, in order, when necessary, to lower the temperature
of the hearth and facilitate the solidification of the bloom.
The orifice, n, Figs. 225 and 226, allows the melted cinders.
or slag to flow off from the surface of the melted metal. A
copper pipe or nose piece conducts the blast of both bellows
into the hearth, as shown at b, x, Fig. 226.
I
With a view to economy in fuel, finery hearths covered
with a vaulted roof were employed in France (Nivernais)
at the commencement of the present century; this im-
provement was not introduced into the Franche Comté and
the Champagne until 1830. At Audincourt the waste heat
of finery hearths has been used for heating iron plates since
about 1825; and ten years later this source of heat was used
for previously heating the pig-iron intended for the finery
process, with the flue of finery hearth suitably constructed.
The waste heat of the finery hearth was then also used for
heating the blast. But these improvements have been in
APPARATUS USED IN THE FINERY PROCESS.
743
general use only since 1834, and even at the present time
they are neglected in many finery forges.
The closed hearths have two disadvantages; they reflect
the heat, which injures the workmen, and a mixture of slag
and small coal (Lösch) sticks to the roof; these deposits,
however, upon removal, must not fall into the hearth.
The waste heat* of finery fires is also used in the puddling
process (Karr'st method) to heat boilers, &c.
When constructing a finery hearth the hole or pit intended
for it is cleaned and its bottom levelled; the tuyere side
plate and the opposite side plate are then put in their places
and propped, until the back plate is placed between the two
former plates and fixed to them with wedges. After removing
this prop the hearth bottom of brasque is levelled, and as
much of the brasque removed, about 1 inch high, as is
necessary to form the hollow space below the bottom, plate.
The brasque supporting the bottom plate, and the ground
adjoining it, must be of a loose consistence in order to admit
the water for cooling. The hole formed for admitting the
cooling water is then also examined, and the iron bottom
plate is slid into the hearth on two iron bars and placed on
flat pieces of iron with its four corners levelled. All the
joints are then filled up with pieces of iron and luted with
loam, and after placing the tuyere in the right position with
regard to inclination and projection into the hearth, the front
wall of brasque is covered with an iron plate when the
hearth is ready for operation.
The iron hearth bottom is usually placed horizontally,
except when treating grey pig-iron, when it is placed with an
inclination towards the tuyere side plate, as the largest
quantity of iron collects before the tuyere, and more rich
decarbonising cinder will accumulate here. The tuyere side
plate is usually in a slanting position, the upper part pro-
jecting somewhat into the hearth, and is thus better able to
support the tuyere and preserve the plate from being too
quickly burnt. The opposite side plate and the back plate
* GURLT, in Berggeist, 1860, p. 572. TUNNER, Stabeisen und Stahlbereitung,
ii., 256. Allgem., B. u. h. Ztg., 1862, p. 496.
+ B. u. h. Ztg., 1854, p. 12.
744
IRON.
incline obliquely the other way, thus facilitating the removal
of deposits and the raising of the fined iron out of the hearth.
The tuyere is usually made of copper, and sometimes
hollow, so as to allow a circulation of water, and its position.
greatly affects the finery process. It projects 3 or 4, and in
rare cases, 6 inches into the hearth, and its inclination varies
from 5° to 16°, and in rare cases more, according to the nature
of the pig-iron under treatment; the decarbonisation of the
iron increases with the inclination of the tuyere. The less
the tuyere projects into the hearth the nearer the focus will
approach the tuyere side plate, causing a cooling of the
hearth. By shortening one side of the tuyere nozzle the
blast may be expanded partially, that is, carried more to one
side of the hearth than to the other.
I
The tuyere nozzle is usually semicircular, and either of
the same width as the blast-pipe nozzle or somewhat larger.
The blast-pipe is from 1 to 1 inches wide, and its nozzle
within is from 3 to 5 inches of the tuyere nozzle. When
treating white pig-iron narrower tuyeres (13 inches broad
and 1 inches high) are employed than when fining grey
pig-iron; in the latter case the tuyeres are 2 inches broad
and 1 inches high. The tuyeres are placed 8 or 10 inches.
distant from the back plate, and the smaller this distance
the more decarbonising will be the reaction of the blast, as
most of the iron is accumulated before the back plate by the
manipulations in the hearth.
I
Two or three tuyeres are occasionally employed in larger
hearths to increase the production, but one is the usual.
number.
When two tuyeres are used, they are placed either
opposite each other or side by side, but the anticipated
advantages have been less realised than in the refining of
the pig-iron.
The size of the hearths, and therefore the quantity of blast
required, depend chiefly on the amount of the production
and the nature of the pig-iron to be treated. Upon charging
from 2 to 3 cwts. of pig-iron, from 1 to 2 cwts. of wrought-
iron are produced in from two and a half to five and a half
hours, or 50 or 60 cwts. weekly, at a loss of from 25 to
APPARATUS USED IN THE FINERY PROCESS. 745
28 per cent. When treating grey pig-iron, which requires a
great deal of manual labour in the hearth, the hearth is made
longer than it is broad, the breadth of the fire being the dis-
tance from the tuyere side plate to the opposite side plate.
Some of these hearths are 32 inches long and from 24 to 26
inches broad, and when working white, easily refined, pig-iron
the hearth is made of less length and greater breadth.
In some cases the hearth is square in form, 22 inches
square on the bottom and 23 inches at its mouth. A narrow
hearth concentrates the heat more perfectly.
At Königshütte (Hartz) the finery hearths are constructed
of the following dimensions. I. A hearth for the production
of wrought-iron intended for the wire manufacture, from good
grey pig-iron. 2. A hearth for the production of common
wrought-iron from pig-iron containing some sulphur.
Length from the front to
the back.
Breadth from the tuyere
side to the opposite side
on the top.
Total depth
on the bottom
Depth from the bottom to
the lowest edge of the
tuyere
Projection of the tuyere
Inclination
""
Width of the semicircular
tuyere nozzle.
Height of the tuyere ·
Distance of the tuyere from
the back plate.
Breadth of the blast-pipe
I.
II.
23
22
inches
243/3
231
2 2 H
+3+
14
23 3/1/2
~ ~ H
22
I 2
݂ܕ
10-101
91
4
15-17
4
12-13
degrees
I
H H
со
831-8343
inches
I
"
8 3/31/1
nozzle
Height of it
color
H
I
I
I
""
Section of it
1*3435 1*3435 square inches
The depth of the fire has most influence upon the finery
process, and usually varies between 6 and 10 inches, mea-
suring from the hearth bottom to the lowest edge of the
tuyere nozzle. The following circumstances regulate the
depth :-
746
IRON.
a. The quality of the pig-iron. White iron requires a
deeper fire than grey pig-iron.
b. The quality of the fuel. When using hard charcoal the
hearth requires less depth than when using soft coal.
c. The inclination of the tuyere, which must be in such
proportion to the depth of the fire as to prevent too quick a
melting of the partly decarbonised iron on the bottom of
the hearth. The more inclined the tuyere and the shallower
the hearth, the quicker will be the decarbonisation.
The Blowing Machines are usually bellows or cylinder
machines, and they must be so constructed as to deliver from
400 to 600 cubic feet of blast per minute. The pressure and
quantity of the blast depend on various circumstances, such
as the quality of the pig-iron and fuel, the largeness of the
production, the method of fining, the construction of the
hearth, &c. When, for instance, the melted iron keeps
too long in a thinly fluid state, less blast is employed in
order to lower the temperature, and consequently induce a
more pasty consistence of the iron, in which it fines more
quickly. A similar effect, that is, a decrease of blast, is pro-
duced when employing a tuyere with a larger nozzle, and
placing the blast-pipe farther back in the tuyere. As a
general rule, a good grey-iron requires from 140 to 150 cubic
feet of blast to melt it, and a white pig-iron from 160 to 180
cubic feet. These quantities of blast are at the commence-
ment of fining operations increased to 200 or 210 cubic feet,
and later on to 240 or 250 cubic feet.
The following quantities of blast are required at Königs-
hütte (Hartz) when employing the German method of fining,
and a tuyere of 1°3435 square inches in section :—
Common Wrought-Iron.
Pressure of the
Blast.
Lines-Mercury.
Wrought-Iron for the Manu-
facture of Wire.
Blast.
Lines-Mercury.
Quantity of
Blast.
Cubic Feet.
Quantity of
Blast.
Cubic Feet.
Pressure of
Forging the blooms and
melting the pig-iron
Breaking up the iron.
15-17
140-156
16
II-13
119-133
10-12
144
119-133
Formation of the ball
gradually increased
22-28
168-191
18 -24
156-174
**
Hot Blast* was first employed in France (page 742); it
Bgwkfd. ii., 230; iv., 94, 104, 417; iii., 463; viii., 442. KARST. Arch. 2 R.,
x., 703; xi., I7I.
APPARATUS FOR RE-HEATING THE BLOOMS.
747
gives a saving of 15 or 20 per cent of fuel. When combining
the application of the hot blast with a preparatory heating
of the pig-iron, a suitable regulation of the temperature and
a proper construction of the hearth, it has been found
successful in many iron works, and does not deterioriate the
quality of the resulting wrought-iron. Less favourable results
have been obtained in other establishments, as for instance
in the Iron Works of the Upper Hartz, as the hot blast im-
peded the fining, and yielded wrought-iron of inferior quality,
though, on the other hand, there was a small saving of fuel.
Upon changing from cold blast to hot blast, less pressure
must be employed in proportion to the temperature of the
blast, and the tuyeres and blast-pipes must be wider, requiring
a quicker movement of the blast machine.
The Hot Blast Apparatus* is usually heated by the waste
heat of the finery-hearths.
Steamt was tried in admixture with the blast for the re-
moval of sulphur, &c., from the iron, but the expected result
was not obtained. At the former finery forge at Silbernaal,
near Clausthal, the promotion of the fining was attempted
by mixing steam with the hot blast, which kept the iron
too long in a thinly fluid state; the process was prolonged,
and the consumption of fuel increased, whilst the yield of
iron was larger and of better quality than when employing
cold or hot blast without steam.
The most perfect of all finery-hearths is the Lancashire
forget with a closed hearth, provided with a hearth for a
preparatory heating of the pig-iron, with water tuyeres, and
a hollow bottom plate so as to allow of its cooling. This
forge is combined with re-heating furnaces, and it yields from
5 to 10 per cent more iron than the common German method
of fining, economises 33 per cent in fuel, and produces an
excellent iron.
Apparatus for Re-heating the Balls or Blooms.
The blooms, after being hammered and thus greatly con-
densed, mostly under heavy forge hammers, are re-heated for
MERBACH, Anwendung der erhitzten Gebläseluft, 1840.
+ Bgwkfd. i., 130. B. u. h. Ztg., 1842, p. 326.
Berggeist, 1858, p. 110.
748
IRON.
forging into bars. According to the method of refining
employed, the re-heating is performed either in the same
fineries, during the melting of the pig-iron, or previous to
this operation, or in special apparatus.
The employment of special apparatus is preferable, as it
allows a quicker re-heating and a better regulation of the
operation, but the consumption of fuel is larger. When the
refining and re-heating take place in the same hearth, the
refining is facilitated by those parts of the blooms which
melt and collect on the bottom of the hearth, and which.
would otherwise have to be replaced by suitable fluxes.
The special apparatus and re-heating fires are of the same
construction as the finery-hearths; they are about 2 feet
square, and are sometimes either wholly or partially formed
of iron plates, and sometimes of fire bricks with a bottom of
sand and small coal. In front they have a tapping hole.
The tuyere is often placed horizontally, but now and then
with an inclination. Since 1834 hot blast has been ad-
vantageously used.
These re-heating apparatus may be divided into—
a, Open fires, in which charcoal, coke, or coal may be used
as fuel. When coal is employed that of a caking nature is
preferable, as it allows the formation of an artificial roof of
caked coal, thus concentrating the heat. Such a roof of
sufficient consistence cannot be formed when the coal is
mixed with coke-cinder.
b, Re-heating fires with a walled roof (hollow fires); they
frequently have a hearth for a preparatory heating of the
blooms. Figs. 227 and 228 represent a re-heating fire of
this kind. A is the re-heating hearth; B, the hearth for the
preparatory heating; c, the space for heating the blast.
a is formed of sand, b of fire bricks, and d of cast-iron.
These apparatus are either fed with coke or with a mix-
ture of coal and coke cinder, and as their construction
allows an entrance at the back, they admit of the heating
of several blooms at the same time. One re-heating fire
will usually heat the blooms produced in one finery hearth.
Re-heating furnaces, heated with gas from wood or turf,
have lately been introduced for this purpose into the larger
APPARATUS FOR FORGING IRON.
749
A
establishments in Sweden, Reichenau, and Rohnitz.
description of them will be found in the chapter on fuel.
FIG. 227.

A
0
1
2
Z
B
5FI
One of these furnaces will heat the blooms of four finery
hearths. These furnaces consume considerably less fuel
than the re-heating hearths.
* - - **
d
FIG. 228.

d
b
b
し
​A
d
Ն
0
B
مر
b
C
5FI
Apparatus for Forging Iron.
The apparatus used for forging blooms produced in finery
fires are hammers of different construction, which are set in
motion either by direct engines or more commonly by water
wheels; or the motive power is communicated by means of
connecting straps, as with the lighter kinds of hammers.
750
IRON.
1. Lift Hammers frequently used in Germany and at
tin-plate works of this country, for converting blooms into
mercantile bars. Such a hammer is represented by Figs.
229 and 230.
FIG. 229.

20 F
The hammer head varies in weight from 5 to 20 cwts. ;
it is raised to the
blows per minute.
height of 2 feet, and gives about 100
According to the size, it is made either
FIG. 230.

e
KOMMU
S
JЪ
α
al
d
10 FI
of wrought-iron or of cast-iron, and lately also of cast-steel ;*
in order to render the cast-iron hammer heads more durable
Zeitschr. d. Ver. deutscher Ingenieure, iii., 206.
LIFT HAMMERS.
751
the eye is sometimes lined with a wrought-iron ring.*
Brown has replaced the wood on hammer shafts by shafts
of iron plates of an elliptic section, and Rittinger has capped
the wooden shaft, as they suffer most on their heads.
Instead of the faces of the hammer being smooth and
convex, they are frequently made in the form of a cross, and
they are fixed to the hammer in sockets, so as to be easily
replaced.
The frame is constructed either of wood with some parts
of iron, or altogether of cast-iron, and consists essentially.
of the following parts :-
Two pillars, a, to which the hammer shaft is fixed by
means of an iron ring, provided with trunnions which move
in iron blocks with which the pillars are provided. Behind
these pillars two other pairs of pillars, c and d, are erected
containing two wooden beams, e, in a nearly horizontal
position above the hammer shaft. By lifting the hammer
its shaft is thrown against these beams, which act like
springs, preventing the hammer from being thrown up too
high, and accelerating its fall. When the whole frame
is constructed of wood a heavy wooden beam is placed upon
these three pairs of pillars and fixed to them, thus con-
tributing to the stability of the frame. The hammer is
raised by the four cams, g, g, g, g, on the shaft, f.
The anvil, i, is a solid piece of white cast-iron, resting
upon a wooden block composed of several pieces of timber
well fitted together, with the fibre running vertically, and
surrounded with a cast-iron casing, or firmly bound with
wrought-iron hoops. This anvil block is 6 or 8 feet long
and 3 or 4 feet in diameter; it is supported on a wooden
foundation, and projects above the floor not more than
I feet.
Dawest cools the hammer and anvil by a continuous
current of cold water, and Puttmannt has combined four
hammers in such a manner that the blooms may be forged in
a horizontal and a vertical direction at the same time.
*DINGL., Bd. 129, p. 195.
† B. u. h. Ztg., 1959, p. 464.
+ Berggeist, 1856, p. 217.
752-
IRON.
2. Tilt Hammers.-A hammer of this kind is shown in
*
Fig. 231. To the left of I is the axis of the rotatory
FIG. 231.
u
13
h
N
a

1
cam, 2, 3 consisting of eight sides, each formed of a
strong broad bar of cast-iron, joined together to make the
octagon wheel. 4, 5, 6, are cast-iron binding rings or hoops.
made fast by wooden wedges; b, b, are standards of the
frame work, e, l, m, in which the helve of the forge hammer
has its fulcrum near u; h, the sole of the frame. Another
cast-iron base or sole is seen at m; n, is a strong stay to
strengthen the frame work. At r, two parallel hammers.
are placed, with cast-iron heads and wooden helves; s is
the anvil, a very massive piece of cast-iron; t is the end of
a vibrating beam for throwing back the hammer from it
forcibly by the recoil; x, y, is the outline of the water
wheel, which drives the whole. The cams or tappets are
shown mounted upon the wheel, b, g, b.
3. Forge Hammers.-The largest forge hammerst (of
the head lift form) are represented in Fig. 232. They are
made entirely of cast-iron, nearly 10 feet long, and consist
usually of two parts-the helve, c, and the head, d. The
head is forced into the helve, and is retained in its place
by wedges of iron or wood. The head consists of several
faces or planes receding from each other, for the purpose of
giving different forms to the blooms. A ring of cast-iron, a,
called the cam-ring-bag, bearing movable cams, b, b, drives.
* URE'S Dictionary of Arts, vol. ii., p. 742.
† Ibid.
FORGE HAMMER.
753
the hammer d, by alternately lifting it up round its fulcrum, f,
and letting it fall. In one iron works, this ring was found to
be 3 feet in diameter, 18 inches thick, and to weigh 4 tons.
The weight of the helve (handle) of the corresponding
hammer was 3 tons, and that of the head of the hammer,
8 cwts.
FIG. 232.

m
a
с
10%
The anvil, e, also consists of two parts; the one called
the pane of the anvil is the counterpart of the pane of the
hammer; it likewise weighs 8 cwts. The second, g, named
the stock of the anvi., weighs 4 tons. It is in form a parallelo-
piped with the edges rounded. The blooms, or rough balls,
from the puddle furnace, are laid and turned about upon it,
by means of a rod of iron welded to each of them, called a
porter. As the weight of these pieces is very great, and the
shocks very considerable, the utmost precautions should be
taken in setting the hammer and its anvil upon a sub-
stantial mass of masonry, as shown in the figure, over which
is laid a double, or even quadruple flooring of wood, formed
of beams placed in transverse layers close to each other.
These beams have an elastic force, and thereby partially
destroy the injurious reaction of the shock. In some works
a 6 feet cube of cast-iron is placed as a pedestal to the anvil.
VOL. II.
3 C
754
IRON.
The Methods of the Finery Process.
The differences between the numerous methods of fining
are principally based on the quality of the pig-iron, its purity,
and the rapidity with which it is decarbonised. The pig-iron
is sometimes previously prepared for the finery process
(page 714), and in the finery process it is treated once or
twice with oxidising agents, and the resulting blooms are
re-heated either in combination with the finery process
(page 748) or in special apparatus (page 748). There are
also some modifications of the finery process with a view to
economy in fuel, a larger production, or a decreased loss of
iron. With regard to these circumstances the Lancashire
forge is by far the most perfect (page 747).
The methods of the finery process may be classified ac-
cording as the metal is once melted down, twice melted down, or
several times melted down, which is done in the German process.
In each of these methods the process may run into the
two following extremes, namely:-(1) the melted-down pig-
iron congeals more or less quickly to a pasty mass of malle-
able iron, whilst a rich cinder is formed; and (2) the pig-iron
comes to nature—that is, it is converted into the malleable
state-with difficulty, and the formed slag is more thinly
liquid and poorer in iron. The finery process must be regu-
lated according to the purity of the pig-iron under treatment,
and according to the wrought-iron (richer or poorer in carbon)
to be produced. A poor slag is chiefly formed when the
fused iron in the hearth may be felt to be liquid by means of
an iron rake, and when upon removing the rake, thinly
liquid slag containing intermixed grains of iron adheres to it
and on cooling easily flies off. A rich cinder is indicated if
the fused iron when pierced with the rake appears as a pasty
mass; and if the iron shows a more or less glaring white and
not a yellowish colour, a thin crust of iron will then adhere to
the rake.
The construction of the finery hearth is the chief means
of inducing a rich cinder; the hearth ought to be shallow
(page 745), the hearth bottom to be inclined towards the
tuyere side plate (page 743), the distance to be small from
METHODS OF THE FINERY PROCESS.
755
the tuyere to the back plate (page 744), there should be a
decided inclination of the tuyere (page 746), a wide tuyere, the
upper part of the nozzle somewhat projecting, and a wide
blast pipe (page 744), and the tuyere should project but
slightly into the hearth (page 744). During the operation
the formation of rich cinder may be facilitated by somewhat
withdrawing the blast pipe (page 746), by decreasing the
pressure of the blast (page 746), by employing blast of a
lower temperature (page 747), by frequently tapping off the
cinder, and slowly melting down the pig-iron, by employing
decarbonising fluxes (page 736), and soft charcoal in
large pieces (page 735), by a modification of the mani-
pulations, &c.
Very different modifications are introduced when aiming
at the production of a poor slag. For example, when treating
very pure white pig-iron by the German method the hearth
is constructed 9 or 10 inches deep, whilst a slight inclina-
tion is given to the tuyere, or a hearth of from 8 to 9 inches
deep and a tuyere with a greater inclination are employed;
when treating impure white iron the hearth is from 94 to
10 inches deep and the tuyere is placed in a strongly inclined
position. Very pure grey pig-iron requires a hearth 7 inches
deep and a nearly horizontal tuyere, and impure grey iron is
treated in hearths from 73 to 8 inches deep with tuyeres of
great inclination. When the metal is melted down once or
twice the construction of the hearths for the treatment of
certain kinds of pig-iron has been always the same, whilst in
the German method the construction is modified according
to the nature of the pig-iron under treatment.
Besides these extremes of the process other varieties may
occur.
The process is called hot when the smelting mass becomes
thinly liquid, owing to the very high temperature; the iron
then comes more slowly to nature, the loss of iron is less,
but the consumption of time and fuel is greater, and a good
product results from less pure pig-iron. This variation of
the process is very liable to occur when employing hot blast.
or a shallower hearth, when the side plates project more
than usual (2 or 4 inches) above the tuyere, &c.
302
756
IRON.
Too low a temperature causes a modification which is called
cold, as then pig-iron and cinder mix whilst in a pasty state
and thus react more strongly upon each other; the iron comes
more quickly to nature, but much of it is scorified. Lowering
of the pressure of the blast may.give rise to this variety of
the process.
It may also happen that too much or too little slag is
formed in the process, in which case the iron refines quickly
but with greater loss.
The loss of iron* in the process may amount to as much
as 40 per cent, but it is usually not more than 26 or 28 per
cent, and frequently even less; for instance, in hearths with
an arched roof carried on with hot blast the loss may amount
to 20 per cent only.
The fining process is more perfect and the skill of the
finer greater the less fuel is consumed, the less iron is lost,
and the better the quality of the resulting wrought-iron.
I. The German or Breaking-up Process, in which the
iron is melted down several times.
This is the most complicated of all the methods, and
requires very skilful workmen.
When the construction of the hearth is properly modified
this process is suitable for working all kinds of pig-iron, but
usually a moderately impure mottled or grey charcoal iron
is treated by it.
The process is divided into the following periods:-
First Period.-Melting Down the Pig-Iron.-Pig-iron
is on fusion melted together with rich or poor slags of the
same process, according to its purity and how thinly liquid
it is. For this purpose the pigs are placed on the side plate
opposite the tuyere covered with coal and gradually moved
towards the tuyere. For a better concentration of the
heat in the hearth the side plates are covered with small
coal, which must be frequently wetted with water or the
blast will carry them away. At this melting down the blast
plays round each liquid particle of iron, when, according to
Oesterr. Ztschr., 1853, p. 378.
THE GERMAN OR BREAKING-UP PROCESS.
757
the investigations of Calvert, Johnson,* Botischew,† and
others, silicon and manganese with more or less iron will be
first oxidised by the action of the blast and slag evolving
oxygen, thus forming a cinder rich in silica, as in the puddling
process; sulphur and phosphorus are partly oxidised (phos-
phorus more than sulphur), and the graphite is converted into
chemically combined carbon in proportion as silicon, sulphur,
and phosphorus have been separated; this renders melted-
down iron white, like that produced by the refining process
(page 715). The resulting poor slag is tapped off and re-
placed by an addition of rich cinder varying in amount
according to the nature of the iron. Now begins the
+
Second Period. The Fining Proper. The pasty white
iron is broken up by means of an iron rake or paddle into
pieces, which are lifted above the tuyere and melted down
again. This operation is repeated one or more times ac-
cording to the nature of the iron, until after the last
melting down the iron may be formed into a simple lump
of malleable iron. If the lump still contains particles of
raw iron, showing themselves by a less intense glare, it may
be brought repeatedly before or above the tuyere. When
melting down the pig-iron the blast blows red and blue star-
like sparks of poor cinder out of the hearth; at the end of
this period these sparks are of a glowing white, and consist
of particles of rich cinder.
At the melting down of the broken-up iron, the iron and
manganese present are chiefly oxidised; the manganese
also assists the removal of the sulphur and phosphorus
(page 692); the poor cinder is thus converted into rich
cinder (page 737); owing to its pasty consistence it mixes
with the pasty iron and has a powerfully decarbonising re-
action, and reduces metallic iron. At this period and with
an increased quantity of blast the decarbonisation and puri-
fication of the iron are finished, and the lump of iron mixed
with very rich slag is lifted out of the hearth and formed by
* B. u. h. Ztg., 1858, p. 34.
† Oesterr. Ztschr., 1862, No. 29. HARTM., Fortschr., vi., 233.
DE MOTAY'S and FONTAINE'S Theory: B. u. h. Ztg., 1857, p. 99.
758
IRON.
hammering into blooms. The rich slag is then partly tapped
off. Slag containing a great amount of iron is very refrac-
tory and adheres closely to the hearth.
The purity of the pig-iron, its nature, and even local habits
have led to different modifications of the second period, and
we may distinguish the following varieties :
a. The Lump Fining.-The melted-down pig-iron is again
broken up and melted down to form a lump, which then is
completely refined by a repeated treatment before the tuyere.
The resulting wrought-iron is granular and richer in carbon,
and upon rolling it is less inclined to assume a fibrous tex-
ture. This variety was formerly in use in Sweden, and also
at Königshütte (Hartz) for the production of wrought-iron
intended for wire manufacture; the iron employed was a
good grey pig-iron from Steinrenne; 100 lbs. yielded from
71.8 to 73.3 lbs. of wrought-iron, 100 lbs. of which consumed
from 24.6 to 26.8 cubic feet of charcoal.
b. The Breaking-up Process.-A very impure thinly
liquid iron of almost any kind is broken into a larger number
of pieces, and each piece is fined or brought to nature by
itself, when all the pieces are melted to form one lump,
which, if required, may be broken up again. This mode
requires very skilful workmen and is apt to yield a hetero-
geneous product.
c. A Combination of the Lump Fining and the Breaking-
up Process. By this mode a moderately pure grey or
mottled pig-iron is treated. The melted down pasty iron is
repeatedly broken up into pieces, which are moved towards
the tuyere. The pieces are brought to nature separately,
and then melted so as to form one lump, which is treated as
in the lump fining process. It is advisable not to turn the
lump, but merely to lift it and to melt it down in its original
position after having cleared the hearth below the tuyere and
filled it with fresh coal. Thus the lumps rest on a decarbon-
ising base, and a better iron is produced, as sufficient time
is allowed by the slower progress of the process at the
latter period of the melting down to fine or bring to nature
the more raw and upper part of the lump. This method is in
use at the forges of the Upper Hartz; at Königshütte, for
THE BOHEMIAN PROCESS.
759
instance, for the production of common wrought-iron as well
as of wrought-iron intended for wire (page 745). When pro-
ducing common wrought-iron the iron is broken up from
three to five times, and 220 lbs. of pig-iron are fined in 5 hours,
yielding 76 per cent of wrought-iron at a consumption of
from 23 to 26 cubic feet of soft charcoal, or from 16 to
20 cubic feet of hard charcoal per 100 lbs. of wrought-iron.
In 24 hours the production amounts to from 800 to 825 lbs.
of wrought-iron. When producing malleable iron for the
manufacture of wire from pig-iron of good quality, the iron
is broken up two or three times into larger pieces, and great
care is devoted to rendering the lump as uniform as possible.
210 lbs. of pig-iron are worked in about 5 hours, yielding
69.75 per cent, and at a consumption of 26.5 cubic feet of
charcoal per 100 lbs of wrought-iron. In 24 hours the pro-
duction amounts to from 710 to 750 lbs.
The Swabian process (Schwabische Schmiede) is conducted
in a similar manner. The hearths are closed with a roof,
and constructed so as to allow the pig-iron and the blooms.
to be heated in the flue. They seldom have hot blast
apparatus (South Germany,* Silesia,† Mägdesprung, and
Sweden). According to the quality of the pig-iron, from
17 to 24 cubic feet of charcoal are consumed per 100 lbs. of
wrought-iron, and the yield amounts to from 75 to 82 per
cent.
d. The Bohemian Process (Böhmische Anlaufschmiede).
-When melting down the broken iron an iron bar is intro-
duced into the fining mass and revolved; the decarbonised
iron will then adhere to the bar. As soon as the bar has
accumulated from 10 to 20 lbs. of fined iron, this quantity
is forged out and cut off from the bar. The fined iron pro-
duced last in the hearth is inferior in quality to that which
has been removed with the bar. 200 lbs. of wrought-iron are
produced in one charge, and in from 5 to 8 hours, at a loss
of from 19 to 21 per cent of iron, and a consumption of from
Berggeist, 1857, P. 340.
† Oesterr. Ztschr., 1857, p. 285.
B. u. h. Ztg., 1854, P. 377.
760
IRON.
14'7 to 16 cubic feet of charcoal per 100 lbs. This method
is chiefly used in Silesia* and Bohemia.†
The Rohnitz Process.‡-In this process 6 cwts. of pig-iron
are melted down in a hearth with two tuyeres; the coals are
then removed, and the metal-bath is stirred up with decarbon-
ising fluxes. The mixture is now placed between two tuyeres
lying opposite each other, blast is introduced, coals are placed
upon the iron and after this fining period, an iron bar is in-
troduced into the mass for the accumulation of wrought-iron.
About 525 lbs. of wrought-iron are produced in six hours, at
a loss of from 121 to 14 per cent of iron, and at a consump-
tion of from 32 to 37 cubic feet of charcoal per 100 lbs. of
wrought-iron.
e. The French Process prevails in France and Belgium,
and it is also used in Switzerland, Germany, and Sweden.
The process is accelerated by frequent and quick manipula-
tions, and the breaking-up often takes place when drawing
out the blooms, thus deteriorating the latter operation. From
17 to 20 cubic feet of charcoal are consumed at the produc-
tion of 100 lbs. of wrought-iron, and from 20 to 25 per cent
of iron is lost.
f. The Sulu Process, sometimes used in Sweden. By
this process the broken up pieces are brought to nature and
forged singly.
g. Half Walloon Process, used in France, but not often
now in Sweden. The pecularity of this process is simply
that it heats the blooms in separate re-heating fires.
Third Period.-Hammering the Loupe.-By means of a
hook and tongs the loupes produced by one or the other
method are taken out of the hearth and placed on the anvil
of the hammer. At first the movement of the hammer is
slow, but the speed is gradually increased; thus the slag is
pressed out of the loupe, and the loupe is hammered to a
prismatic shape, and divided into several pieces (blooms) by
means of a cold chisel. To be drawn out to bars of certain
* Oesterr. Ztschr., 1857, p. 284. B. u. h. Ztg., 1858, p. 172; 1860, pp. 169,
279, 352, 371.
+ B. u. h. Ztg., 1846, p. 881; 1849, p. 714; 1855, p. 166. LEOB. Jahrb., 1855,
p. 59.
‡ A modification of the Bohemian process.
THE WALLOON PROCESS.
761
dimensions the blooms are re-heated either in the finery
hearths or in separate fires, the time depending upon the
rawness of the blooms. When intending a further oxidation
of the carbon, which also embraces an oxidation of iron, a
dry welding heat is applied, otherwise a wet (saftige) welding
heat, produced by an addition of sand, &c., is preferred; in the
latter case a slag is formed which coats the iron (page 690).
2. Walloon Process-Twice Melting-Down Process
(Zweimalschmelzerei).-The iron treated by the Walloon
process is a purer white or mottled charcoal iron poor in
manganese, sulphur, and phosphorus; if necessary, the iron
is previously refined (Eifel-Walloon process). Works
adopting the English Walloon process refine coke pig-iron
and re-heat the resulting blooms in hollow fires or re-heating
furnaces, which contributes to the production of a superior
kind of wrought-iron (Swedish cement and wire-iron, Styrian,
Eifler, and English iron for wire). The pig-iron employed
is usually as pure as the pig-iron of the German process, and
is obtained by the first melting down; the first melting down
of the Walloon process therefore corresponds to the second
melting down of the German process, and the purer iron of
the Walloon process allows the formation of lumps after
having been only twice melted down (Eifler, Styrian, Swedish
Walloon process). The employment of more impure pig-
iron, or the production of the best quality of wrought-iron
necessitates repeated meltings down (English Walloon pro-
cess), when the process is similar to the German process,
except that the blooms are re-heated in separate fires.
The chemical reactions during this melting down of the
pig-iron are like those of the second melting down in the
German process (page 757), and the reactions at the last
melting down are the same in both processes.
The Walloon process and its modifications are mostly em-
ployed, as the process yields wrought-iron so excellent in
quality as to fully repay the high cost of production, com-
pared with the cheaper puddling iron.
The following modifications of the Walloon process have
originated partly from mere local custom, and partly from
the quality of the pig-iron to be treated and the quality of
762
IRON.
the wrought-iron to be produced. That quality is likewise
influenced by the mode of re-heating, and the fuel (charcoal,
coke, coal) employed.
a. Eifel-Walloon Process.-This process is still in use
in the Eifel and Rhenish Prussia for treating charcoal pig-iron
which has been refined in the hearth of the iron blast fur-
nace (page 714). The hearth is lined with moist small coal,
and its bottom formed of very rich cinder, old iron, and
cinder separated from the lumps when hammering them.
Before all the pig-iron is melted down the first melted iron
that has collected on the bottom is broken up and some
cinder of the hammered lump is added, when from 50 to
70 lbs. of pig-iron have been melted, whilst the blast is some-
what decreased. The melting down and the fining proper
are performed as quickly as possible. The finished lump or
loupe is hammered and divided and re-heated for drawing
out in open hearths 2 feet square; II cubic feet of hard char-
coal are consumed for fining and re-heating 100 lbs. of
wrought-iron, and 25 per cent of iron is lost.
A similar process was formerly used in Styria, but owing
to the large consumption of fuel and loss of iron, it was aban-
doned in favour of the process in which the metal is only
once melted down.
b. Swedish Walloon Process, used for the production of
the well-known Dannemora iron from very pure radiated
white or strongly mottled pig-iron. In order to furnish the
hearth with a decarbonising bottom, several shovelfuls of
very rich cinder and hammer slag are melted down, and the
last bloom of the former lump is held with tongs in the focus.
of the finery hearth, thus causing part of the bloom to melt
and drop down to the bottom. The melting down of the pig-
iron is commenced at the same time. After melting the first.
portion of the pig-iron the working with the iron bar com-
mences, and is continued until the whole metal melted down
on the tuyere side is brought above and before the tuyere,
and so exposed to the action of the blast. When 70 lbs. of
the iron have been melted down the mass has come so nearly
to nature that it is only necessary to form it into a lump.
THE LANCASHIRE PROCESS.
763
The re-heating of the blooms is effected in open hearths.
The fining of 100 lbs. of wrought-iron requires 16 or 17 cubic
feet of coal, and the re-heating from 13 to 15 cubic feet, at a
loss of iron of from 4 to II and from 12 to 15 per cent
respectively.
c. English Walloon Process.-a. Lancashire Process.
-This process was introduced into Sweden from South
Wales in 1829, and nearly all the Swedish bar-iron, ex-
cept the Dannemora or Oregrund iron, is made by it.
The bottoms of the hearths are artificially cooled, water
tuyeres are used, and there is a lateral stove for previously
heating the pig-iron, which is heated by the waste heat of
the finery hearth. The process is carried on with hot blast.
The pig-iron in plates 2 or 3 inches thick, and in charges
of 200 lbs., is previously heated. Upon melting down it is
partly oxidised, and the oxide of iron thus formed, together
with the rich cinder, consisting of basic silicate of protoxide
of iron, which remained in the hearth from the last opera-
tion, decarbonise the iron to a considerable extent, and con-
sequently it becomes less fusible and more pasty. The metal
is then repeatedly broken up with an iron bar, and the raw
portions brought before the tuyere until all is sufficiently
refined; this operation lasts about half an hour. Subse-
quently all the metal is brought up to the top of the hearth
and melted down again to form a ball. The ball is then
taken out and hammered into a prismatic shape and cut into
pieces. The whole process lasts from 1 to 1 hours.
The blooms are drawn out into bars by one of the three
following methods:-
I. In a re-heating forge, with a hammer.
2. In a gas welding or re-heating furnace, heated by char-
coal, and with two hammers.
3. In a gas welding furnace, and by rolling instead of
hammering. The consumption of fuel and the loss of iron
vary according to which of these methods is employed.
This process has also been used advantageously in Nor-
way, at Feistritz* in Carinthia, &c.
PERCY, Metallurgy, ii., 595.
764
IRON.
B. South Wales Process.*-Grey coke pig-iron is the
material; it is previously fined in a coke refinery, from which
it is run into the charcoal finery. Some decarbonising cinder
is left in the finery hearth from the previous operation, and
is then mixed with the fluid iron, whilst at the same time
the blast is turned full on, after having removed the slag
which has been run into the finery together with the metal.
After the iron has become pasty it is broken up, mixed with
charcoal, brought up above the tuyere, and melted down
again; and this operation is repeated till the iron has suffi-
ciently come to nature, when it is melted down for the for-
mation of a lump. The charge for the refining process is
from 200 to 250 lbs.
The lump is then hammered out into a narrow rectangular
slab, more or less rounded off at the ends, and about 2 inches
thick, and while still hot it is nicked across to a considerable
depth in seven or eight places at about equal distances. The
nicked slab, still in a red-hot state, is plunged into cold
water, whereby the superficial coat of oxide is more or less
completely detached. The slab may then readily be broken
into pieces, which are termed " stamps." A stamp weighs
about 26 lbs., and the fractured edges are bright and crys-
talline. The ends of the slab are worked apart, as the iron
made from them is never so sound as that from the other
portions, which are assorted according to their soundness or
freedom from imperfectly converted metal.
The stamps are then re-heated in hollow fires (page 748),
and after a further treatment by hammers and rollers they
yield an excellent material for tin plates and for the manu-
facture of wire.
d. Styrian Walloon Process.-The pig-iron in the form
of pigs, plates, or scrap is quickly melted down and mixed.
with decarbonising agents, when it is broken into pieces,
which are removed from the hearth and re-melted separately,
thus forming balls; or the iron is collected on a bar of iron.
A similar process is adopted in some forges of Hungary,
Bohemia, Moravia, Norway, and Sweden.
*
PERCY, Metallurgy, ii., 583.
THE STYRIAN PROCESS.
765
In the Salzburg process, fused, annealed, and pounded pig-
iron is mixed with hammer slag, &c.
These processes yield a pure iron, but admit of only a
small production, and cause a considerable consumption of
fuel and great loss of iron.
3. The Once Melting-Down Process.-The kinds of pig-
iron employed are pure, white, manganiferous iron, or pure,
grey, and mottled sorts converted into cellular white iron by
a preparatory process. An admixture of decarbonising fluxes
causes the fining proper to take place at the melting down
of the iron, and the blooms are re-heated in the same hearth
either before or during the melting down of the pig-iron.
The chemical reactions at the melting down are nearly the
same as those of the fining proper of the German process
(page 757). Attempts have been made to separate manganese
and carbon by slowly melting them down with an abundant
addition of decarbonising fluxes at a low pressure of the
blast, and as these substances cannot thus be removed so
efficiently as in the former processes, it is attempted by
submitting the blooms several times to a full welding heat,
and to the action of decarbonising agents. The good quality
of the pig-iron allows this treatment.
The following methods are modifications of this process:-
A. Methods without previously preparing the pig-iron.
a. Austrian Slag Process.-The finery hearth is provided
with a preparatory heating furnace and a hot blast apparatus,
and its bottom is formed of rich finery cinders broken up and
stamped firmly down, from 8 to 9 inches thick. The metal
employed is cellular white pig-iron, and during its melting
down the blooms (eight of which are formed of one lump) are
re-heated so as partly to fuse and drop down on to the hearth
bottom. The charge is about 230 lbs., 14 or 15 per cent of
which is lost. The consumption of charcoal amounts to
from 15 to 18 cubic feet per 100 lbs. of finished iron.
b. The Styrian Process.-The bottom of the hearth is
composed of moistened coal (Lösch). Before melting down
the pig-iron the blooms of the former operation are re-heated
with an addition of rich cinder, and the re-heating being half
finished, the pig-iron is merely previously heated and melted
766
IRON.
down, after completing the re-heating. Portions of the iron
which are not fully converted into the malleable state are
again brought above the tuyere and melted down; one ball
is formed; 27 cubic feet of charcoal are consumed per 100 lbs.
of finished iron, produced at a loss of from 8 to 12 per cent
of iron.
c. The Siegen Process is almost now abandoned. The
hearth is of the usual construction of cast-iron plates, and
the bottom formed of cinder and hammer slag. The pig-iron
(radiated white or mottled) is melted down when the blooms.
of the previous operation are re-heated, and the fused metal,
if required, is repeatedly broken up and melted down. The
loss in making finished bars is 25 per cent on the pig-iron,
and the consumption of coal 6 cubic feet per 100 lbs. of
finished iron.
Processes differing slightly from those we have described
were formerly used in Westphalia* and Henneberg,† but
they are now obsolete.
B. Methods requiring a Preparation of the Pig-Iron.-
Flowery white or grey iron is submitted to a preparatory
process in order to decarburise it and convert it into cellular
iron, thus coming to nature when melted down once, as in
the Styrian process.
The following are modifications:-
Carinthian process.-Flowery white iron produced in
"Blauöfen," or grey pig-iron, is formed into discs (page 543),
and heated to redness for twelve or fifteen hours with a free
access of atmospheric air, thus removing the carbon. The
iron is then fined by the Styrian method. The loss of iron
amounts to 20 per cent on the pig-iron, and from 25 to 30
cubic feet of charcoal are consumed per 100 lbs. of finished
iron.
Another method used in Carinthia, and termed in German
"kartitschschmiede, " treats mottled or grey iron produced
in "Blauöfen" in a separate hearth with an addition of
decarburising fluxes, thus forming a lump (Kartitsch) which
* KARST., Arch., 1 R., xiii., 198.
↑ Ann. d. Min., 4 sér., 4 livr., de 1842, p. 258.
PRODUCTS OF THE FINERY PROCESS.
767
is taken out, broken up, and fined by the Styrian process.
The loss is 20 per cent on the pig-iron, and the consumption
of coal from 27 to 30 cubic feet per 100 lbs. of finished iron.
Another method in Styria* (zerrennschmiede) refines the
iron in a hearth with decarbonising fluxes (page 724), forms
the resulting metal into discs, submits it to the oxidising
process ("Braten "), and at last fines it by the Styrian
process.
Products of the Finery Processes.
The chief product of these processes, namely, wrought or
malleable iron, is either sent into the market as it is, or it is
submitted to a further treatment, which in the iron works
consists of the manufacture of rails, plates, wire, &c.
As in the finery processes less oxidising slags are employed
than in the puddling process, the resulting iron is richer in
carbon, more compact and closer grained, harder and stronger,
and therefore better fitted for the manufacture of cementation
steel, wire, and plate, than the fibrous puddling iron, which
frequently contains sulphur and a small amount of inter-
mixed slag.
However, the fine-grained puddling iron which has lately
been produced in large quantities has extensively replaced
the iron of the finery processes.
The resulting faulty bars, rough ends, and other scraps,
are either worked up as an addition to the pig-iron for the
finery process, or formed into piles, and treated in the finery
hearth by themselves with an addition of fluxes (Sweden).†
In some cases they are welded and drawn out, then yielding
an excellent product of great strength (Annahütte‡ at
Königsberg, Rotherhithe|| near London, &c.)
Rich cinders of the process are used as well as hammer
slag as oxidising fluxes for the same process. The cinders
are also melted direct for the production of pig-iron (method
of Lang and Frey) (page 346). .
The waste heat is used for different purposes (page 742).
* Oesterr. Ztschr., 1858, p. 251.
+ TUNNER, Stabeisen und Stahlbereitung, ii., 219.
Bgwkfd., xviii., 533.
TUNNER, Ber. über d. Londoner Industrie Ausstellung, von 1862, p. 48.
768
IRON.
THE PUDDLING PROCESS.
The puddling process possesses the following advantages
which the finery process has not:-It allows the appli-
cation of raw fuel and of fuel in the form of gas, with a
smaller absolute consumption; it gives a larger and cheaper
production, and it is possible by this means to produce
a useful wrought-iron from pig-iron of inferior quality
(page 707).
The puddling process was invented in 1784 by Henry Cort.
From England it was carried in about 1820 to Sweden (Skebo)
and France (Fourchambault), and into Rhenish Prussia at
about 1830; the process was first used in Westphalia and Upper
Silesia in about 1835, and in Siegen, from whence it was intro-
duced to the Hartz, &c., about 1840. Until 1819 coal only was
used in the process, when wood was introduced (Skebo, later
in Carinthia, Styria, Bavaria, &c.); about ten years later
turf was employed (Lauchhammer, later in Würtemberg,
Styria, Bohemia, Hanover, &c.), and since 1832 brown coal
has been in use (Bavaria, later in Styria, Carinthia, Nassau,
&c.) In 1837 Fabre du Faur invented the application of
the waste gases from blast furnaces, and since then they have
been also applied to the puddling process (Wasseralfingen,
Hartz, &c.) and also gases produced in generator furnaces from
different kinds of fuel, namely from turf (Lauchhammer,
Mägdesprung, Ilsenburg, Upper Hartz, Carinthia, Tyrol,
&c.), from wood (Carinthia, Styria, Hungary, Thuringia,
Zorge in the Hartz, &c.), from brown coal (Styria, Bavaria),
and from mineral coal (Upper Silesia, Siegen, &c.)
The following treatises will be found of value when
studying the puddling process :-
HARTMANN, der practische Puddel und Walzmeister, 1858.
E. MAURER, Maass- und Gewichtsverhältnisse der Roh-
und Zwischenproducte bei der Darstellung des
Schmieedeisens nach der englischen Frischmethode
oder durch den Puddlings- und Walzprocess. Stutt-
gart, 1861.
PUDDLING MATERIALS.
769
GRUNER et LAN, état présent de la métallurgie du fer en
Angleterre. Paris: 1862, p. 401.
ANSIAUX et MASION, Traité pratique de la fabrication du
fer et de l'acier puddlé, &c. (Atlas). Liège: 1861.
TRURAN, The Iron Manufacture of Great Britain, theo-
retically and practically considered. 2nd edition,
revised by J. Arthur Phillips and W. H. Dorman.
London: 1862.
In treating of the puddling process we propose to adopt
the plan of classification we followed when speaking of the
finery process, namely, to describe first the puddling materials,
next the apparatus, and finally, the methods.
Puddling Materials.
A. Pig-Iron.-The different kinds of pig-iron behave in
the puddling process in the same way as in the finery pro-
cess (see ante). White iron of the regular process and
flowery white pig-iron (blumige Flossen), containing about
4 per cent of carbon when fused, are more or less pasty, and,
if pure, they admit of an acceleration of the process and a
saving of fuel. Spiegeleisen upon fusion is more thinly
liquid, and is too valuable to be puddled by itself; it is,
therefore, chiefly used as an improving addition to impure
sorts of pig-iron. The cellular sorts of pig-iron, chiefly the
varieties having small pores, are seldom worked by them-
selves, as they always require, for fusion, a very high
temperature; owing to the small amount of carbon in them
and to their becoming thickly liquid, they cause a greater
loss of iron in puddling furnaces than when treated in the
finery hearth, as in the puddling furnaces they come more
in contact with oxidising substances. For this reason the
cellular varieties of pig-iron are puddled only in admixture
with other kinds of pig-iron, which become thinly liquid
upon fusion.
Strongly refined pig-iron behaves in the same
way; in a fused state it is thickly liquid, it renders the
puddling process very difficult, and causes a greater loss of
iron; the refined metal is, therefore, usually puddled in ad-
mixture with grey iron. Upon fusion, the varieties of grey
3 D
VOL. II.
770
IRON.
pig-iron are more or less thinly liquid according to their
amount of silicon and graphite; those sorts which are not
too dark, chiefly the mottled iron, are frequently employed in
the puddling process, as a certain degree of thin liquidity is re-
quired for a sufficient separation of the foreign substances con-
tained in the pig-iron; the degree of liquidity may be regulated
by the application of a more or less high temperature. The
darkish grey varieties of pig-iron, rich in silicon and graphite,
are seldom puddled by themselves, as they require too much
time to come to nature, the loss of iron is also increased, and
the formed slag is rich in silica and strongly attacks the
furnace hearth. Pig-iron, containing a larger admixture of
blast furnace slags, behaves in the same way, for which
reason it is better to cast the pigs in cast-iron chills than
in moulds of sand. Pig-iron, which can be puddled only
with great difficulty is improved by the following remedies:
a. Refining in refinery hearths or in reverberatory fur-
naces. Pig-iron, rich in silicon and graphite, and otherwise
impure, is submitted to the refining process, and thus con-
verted into a radiated or cellular iron according to its purity
and to the quality of the fuel employed. When using the
older method of "dry puddling," the varieties of grey iron were
almost invariably treated by this preparatory process, which
is very effective, but expensive. Since the introduction of
the more perfect mode of puddling ("wet puddling" or the
boiling process) the varieties of grey iron, if they are not
blackish grey, are now usually puddled direct, the puddling
process being suitably modified; for example, after having
puddled several charges of grey iron, some charges of white
iron are treated whilst the furnace hearth is suitably formed
and the iron is mixed in the following manner :—
b. Grey iron is mixed with white or refined metal. Some-
times (Hörde)* the different sorts of iron are melted together
in a cupola furnace and run into the puddling furnace in a
liquid state, thus obtaining a more uniform product which
comes more quickly to nature; but according to Kuder-
natscht this method causes a great waste of the furnace
Berggeist, 1861, No. 18.
*
† Oesterr. Ztschr., 1861, No. 48.
PUDDLING MATERIALS.
771
hearth; neither does it save anything, owing to the blast
required. Now and then the different sorts of pig-iron are
first mixed in the refining process (Low Moor).*
c. A suitable formation of the furnace hearth, with a view
to its preservation, is effected in different manners, for in-
stance, by cooling the hearth from below and on the sides,
by forming the hearth and lining its side plates with a mixture
of peroxide of iron with silica, produced by roasting tap
cinder, or with hæmatite, or magnetic iron ore, &c.
According to Listt and Andreet manganiferous pig-iron,
upon puddling, loses the whole of its manganese, and with
the increase of this amount, the waste of iron, the length of
the operation, and the consumption of fuel are increased.
Oxide of manganese, on the other hand, produces a thinly
liquid slag, from which the iron thoroughly separates,
and facilitates the removal of sulphur (page 312), silicon
(page 312), and perhaps also of phosphorus (page 312).
As
manganiferous slags have a less decarbonising reaction than
ferriferous slags, a certain amount of manganese facilitates
the formation of a steel-like fine-grained iron richer in carbon.
The thinly liquid manganiferous slags also cover the iron
more effectually, thus preserving it from contact with the
atmospheric air.
B. Fuel.-In the different operations connected with the
puddling process, the following kinds of fuel are used:
1. In the refining of pig-iron.
In reverberatory furnaces used for refining pig-iron, gas
is usually employed, sometimes coal, chiefly coke in refinery
hearths, and in rare cases charcoal. At the Dowlais Iron
Works the employment of coal in refining hearths has been
tried experimentally, but showed no advantage over coke;
about I cwt. more of coal than coke was consumed per ton
of refined metal as the refining required to be continued
longer; the production decreased whilst the work increased.
Anthracite has given good results when treated in the
* Preuss. Ztschr., iv., 217.
† B. u. h. Ztg., 1860, p. 52.
‡ Oesterr. Ztschr., 1860, No. 16.
3 D 2
772
IRON.
following manner :-In order to prevent its decrepitation it
is partly carbonised and put into the refinery hearth whilst
still hot (Ystalifera), or the anthracite is heated in a space
situated 3 or 4 feet above the hearth, blast being introduced
into the space. As soon as the anthracite is sufficiently
heated, it is passed into the refinery hearth.
2. In the puddling of iron.
a. Coal was first used in the puddling process, and owing
to its cheapness and high pyrometric effect it is still most
frequently employed. Slightly caking coal, burning with a
long flame and as free as possible from sulphur, is best
adapted to the process. When employing small non-caking
coal* or slack, the fire-place must be suitably constructed
(step-grates, conducting blast under the grate, Simencourt's
furnace, &c.). In some iron works of Upper Silesia† caking
small coal has been advantageously used by placing the
grate deeper, and conducting the air for combustion to the
upper layer of coal.
Janoyer states that the sulphurous acid produced from
sulphur contained in the coal does not essentially impair the
iron of the puddling process, as the sulphide of iron formed
upon the iron surface combines with oxidised iron, forming
oxysulphide, which is scorified. From 80 to 100 lbs. of coal
are consumed for the production of 100 lbs. of mill bars ;|
from 3 to 5 cwts. of coal per hour in one puddling furnace.
When employing anthracite a deeper grate is used, and
fresh air conducted under the grate by means of a fan blast
steam is sometimes admitted under the grate at the same
time. In these cases the ash pit must be hermetically
closed.
b. Brown Coal.-Some varieties of this coal are air-dried
before use (Maximilianshütte in Sauforst,§ Kaufing, T
*KARST., Arch., 1 R., iii., 107. Bgwkfd., vi., 33. HARTM. Fortschr., I,
264; v., 204. Schles. Wochenschr., 1859, p. 395; 1860, p. 250.
+
Schles. Wochenschr., 1861, No. 48. HARTM. Fortschr. v., 188, 204.
B. u. h. Ztg., 1855, p. 195.
|| OVERMANN, a Treatise on Metallurgy. New York, 1865.
§ LEOB., Jahrb., 1842, p. 257.; 1858, viii., 121.
iii., 225.
¶ Oesterr. Ztschr., 1857, No. 12.
HARTM., Fortschr., i., 276;
PUDDLING MATERIALS.
773
Westerwald),* and are burned either on large plane grates
Older
or on step grates with the assistance of hot blast.
brown coal is used at Leoben,† Prävali,‡ and Krems with
the application of step grates. The first experiments for the
application of brown coal were made in 1832 at Bodenwöhr
in Bavaria, and afterwards at Maximilianshütte, Hessen-
brückhammer in Hesse, later on in Carinthia (Buchscheiden,
Freudenberg, Prävali), in Styria (Leoben, Store, Judenburg),
and at Hachenburg in Nassau.
At Maximilianshütte the consumption of brown coal per
100 lbs of mill bars amounts to from 120 to 130 lbs. ; at
Prävali to 122 lbs. when using blast for combustion, and to
141 lbs. without the assistance of blast; at Krems to 116 lbs.
(with blast); and at Leoben to 150 lbs. At Kaufing 44 cwts.
of undried brown coal were found to have the same heating
power as 16 or 18 cwts. of good soft wood.
c. Turf was first used at Lauchhammer about 1820, at
Ichoux in France in 1830, and later on at Königsbronn and
Itzelberg in Würtemberg, at Rottenmann in
in Styria,
Wasseralfingen, Kallich in Bohemia, and Maximilianshütte¶
near Trautstein. The turf employed is either air dried or
kiln dried with the application of blast.
At Hammern and Ebenau** in Bavaria the direct applica-
tion of turf on plane grates with blast is considered more
advantageous than puddling with turf gas, because the fuel
was more perfectly utilised, the mode of using it entailed less
expense, and the workmen were less inconvenienced by the
heat.
C
d. Wood was first used in the puddling process in 1819
at Skebo in Sweden, later on at Frantschach in Carinthia,
Neuberg in Styria, Wolfsberg in Carinthia, in Polonia, &c.
*
B. u. h. Ztg., 1854, No. 36.
LEOB., Jahrb., 1852, ii., 246.
‡ ZERRENNER'S Gasfeuerung, 1856, p. 218.
|| Ibid., p. 221.
B. u. h. Ztg., 1843, p. 736; 1845, p. 337; 1850, p. 145. Oesterr. Ztschr.,
1853, p. 112. DINGL., Bd. 96, p. 202; Bd., 59, p. 470.
¶ Preuss. Ztschr., iv., B. 236. HARTM., Fortschr., i., 275; vi., 221. LEOb.,
Jahrb., 1861, xi., 56.
**
LEOB., Jahrb., 1861, xi., 48.
774
IRON.
The wood, either air dried or kiln dried, is usually burned on
common plane grates, and sometimes on grates with the
draught from above (Sweden, Bavaria). In a few cases the
wood is also burnt in a fire place without any grate*.
Wood sustains a high temperature better than turf; it is
also easier to dry, depending less on the state of the weather
and is more easily procured of uniform quality on the other
hand it is more expensive than turf, and seldom procurable.
in quantities sufficient for a large production.
In Russia part of the pig-iron is converted into malleable
iron in puddling furnaces with wood, but the greater part
is still treated in finery hearths.
This compact fuel is sometimes so inferior in quality as to
be unfit for a direct application, but it may produce a tem-
perature sufficient for the puddling process when converted
into combustible gases in a generator carried on either by
mere draught, or by cold blast introduced from below.
These gases are then burnt with hot blast. The firing
with gas is chiefly advantageous if the production of a very
high degree of temperature is required. For the production.
of lower degrees of temperature with pulverulent fuel, step
grates, &c., are best. The gas firing also has the advantage
of allowing a better regulation of the air for combustion and
the avoidance of the formation of smoke. When heating
the combustible gases as is effected by Siemens's regenerators,
a very high temperature may be produced. We purpose
giving a description of Siemens's furnaces later on.
similar heating of the gases may be obtained by conducting
them downwards through a layer of burning fuel, when the
intermixed steam will also be decomposed.
A
Sometimes also better qualities of fuel are converted into
gas, which is a more convenient form.
The waste gases from iron blast furnaces have been ap-
plied for the heating of puddling furnaces, but without
success.
The gases produced in generators were first employed in
1839.
* B. u. h. Ztg., 1843, PP. 448, 463.
PUDDLING MATERIALS.
775
A single gas puddling furnace requires 300 cubic feet of
blast per minute, and a double furnace requires 500 cubic
feet. The blast is usually heated up to 100 or 150° C., and
has a pressure of from 1 to 6 lines mercury.
The following are the different gases produced in the
generators :-
a. Gases Produced from Wood.*-In the puddling pro-
cess these gases produced from air-dried or kiln-dried wood
give most satisfactory results. At Lippitzbach, 4 Austrian
cubic feet of kiln-dried wood were consumed for the pro-
duction of I cwt. of mill-bars; at Neuberg, 5 Austrian cubic
feet, at Neuhüten in Bohemia, a compact mass without inter-
stices of 6 Austrian cubic feet of wood; at Thiergarten, 87
cubic feet of air-dried wood, including the interstices; at
Brezowa, 7.39 cubic feet of green wood and 6.39 cubic feet
of dried wood; at Nadrag, from 5 to 6 cubic feet; at Unter-
kochen, 15 or 16 Wurtembergian cubic feet of dried wood,
including the interstices; at Zorge in the Hartz, from 9 to
14 cubic feet. Wood gases are also employed at Villote,
Erzherzogl, Karlshütte, &c.
For the production of 100 lbs. of mill-bars the best finery
process (the Lancashire process, page 763) consumes at least
II cubic feet of charcoal, which is equal to 22 cubic feet of
wood, supposing the yield of coal to be 50 per cent, whilst
the average consumption of wood in the puddling process
amounts to about 8 cubic feet, equal to a saving of 63 per
cent.f
Wood may be considered as the best material for the pro-
duction of gas, as it permits of a ready storing and of
great cleanliness. Turf requires larger store-houses and its
amount of ash and water are inconvenient.
2. Gases Produced from Turf.‡-The amount of ash and
water contained in turf is so large that the production of
gas
from this kind of fuel has been often abandoned (at the
B. u. h. Ztg., 1846, p. 169; 1849, No. 6; 1851, p. 1; 1852, p. 610.
Preuss.
Oesterr. Ztschr., 1853, p. 3; 1854, p. 207; 1855, pp. 219, 251.
Ztschr., 1856, iii., Lief. 4.
+
Berggeist, 1858, p. 147.
B. u. h. Ztg., 1855, P. 93.
+ Bgwkfd., xiii., 39; xvi., 430. B. u. h. Ztg., 1849, p. 81; 1851, p. 1; 1855,
PP. 139, 150. DINGL., Bd. 131, p. 153; Bd. 132, p. 272.
776
IRON.
Upper Hartz and Ilsenburg, for instance), where for this
reason the direct employment of mineral coal is preferred.
The first experiments for the application of turf gases were
made at Lauchhammer. The fire-place of the generator
should be provided with an inclined perforated iron plate
instead of a grate.
Turf gases are used in the following iron works :—At
Buchscheiden,* consuming from 10 to 12 cubic feet of air-
dried turf per 100 lbs. of mill bars; at Freudenbergt (12 cubic
feet; at Kessen‡ (12.71 cubic feet); at Ebenau|| (17 cubic
feet); at Mandelholz,§ in the Hartz, 304 cubic feet of turf
were formerly consumed per 100 lbs. of puddling iron when
using the generator without blast, and 29'6 cubic feet when
using blast. At present 2'75 cubic feet of mineral coal are
consumed in producing the same quantity of iron, and 2:56
cubic feet (102.25 lbs.) of coal are consumed at Königshütte
in the Hartz.
Turf which has been well dried in kilns and does not con-
tain above 6 per cent of ash when air-dried, produces tem-
peratures as high as those produced by wood, and the con-
sumption of fuel is about equal. At Lippitzbach, for
instance, 33 cubic feet of wood (compact mass), equal to
115 lbs. when air-dried, were consumed per 100 lbs. of mill
bars; whilst 12 cubic feet of turf, including the interstices,
weighing 125 lbs., were required to produce the same quan-
tity of iron.
The ash contained in the turf necessitates more frequent
stoking and cleaning of the fire, therefore, in comparison to
wood, from 8 to 10 per cent more may be consumed in pro-
ducing a certain quantity of iron, and the production of iron
may be 10 per cent less.
c. Brown Coal Gases** are used less often than turf gases,
as the brown coals are frequently rich in ash and water, and
* LEOB, Jahrb., vi., 148.
+ B. u. h. Ztg., 1857, p. 106; 1859, p. 287; 1860, p. 208.
Zerrenner's Gasfeuerung, i., p. 194.
Ibid., p. 203. LEOB., Jahrb., xi., 21.
HARTM., Fortschr., iii., 223.
LEOB., Jahrb., 1857, vi., 129.
Bgwkfd., vii., 26; xi., 249, 309; xvii., 661.
1845, p. 521; 1849, P. 437; 1851, p. 1.
B. u. h. Ztg., 1844, p. 89;
TUNNER'S Jahrb., 1842, P. 257.
PUDDLING MATERIALS.
777
fall into too small pieces when dried. It is then advisable
with an inclined perforated
Most of the iron works we
to burn them in a fire-place
iron plate instead of a grate.
mentioned under the heading Brown Coal on page 772 also
make use of brown coal gases.
d. Gases Produced from Coal.- Non-caking coal is
best adapted for this purpose. The erection and repairs
of the generators are more expensive than those for the
production of the former gases. Coal gases are employed at
Kirchhunden in Siegen for puddling steel, as they allow a
better regulation of the temperature, thus saving from 35 to
40 per cent of coal, whilst the yield is increased from
5 to 10 per cent, and the process is performed in a shorter
time.
Gases have also been produced from charcoal and coke.
Krause and Taylor recommend the application of hydrogen
gas; they produce it by conducting steam over red-hot
iron, and Levick suggests the application of an admixture
of carbonic oxide and hydrogen, producing it by the decom-
position of steam by glowing coals.
3. In the re-heating of the puddled iron.
When re-heating the iron in hearths, coal, coke, or a mix-
ture of both is most frequently used; charcoal only on rare
occasions.
At Rothehütte in the Hartz 1°54 cubic feet (61.6 lbs.) of
coal and o˚2 or 0.3 of cinders are consumed per 100 lbs. of
wrought-iron; at Königshütte (Hartz), o'93 cubic foot, and
at Zorge (Hartz), from 50 to 60 lbs. per 100 lbs. of wrought-
iron.
When re-heating in reverberatory furnaces the same kinds.
of fuel are used as in the puddling process, whilst the pyro-
metric effect of the fuel is increased as much as possible by
a suitable construction of the apparatus (Siemens's furnaces,
the application of hot blast under the grate, &c.)
a. Coal is most generally used; from 40 to 70 lbs. are
required per 100 lbs. of wrought-iron, and 4 or 5 cwts. of
coal are consumed hourly in one reverberatory furnace. The
Berggeist, 1859, No. S5. B. u. h. Ztg., 1860, p. 235.
778
IRON.
application of blast under the grate saves one-third of the
coal, the process is accelerated, small coal rich in ash may
be used, and the time required for cleaning the grate is con-
siderably shortened. At Neustadt (Hanover) the blast under
the grate caused a saving of 42 per cent of coal, the produc-
tion of steam by the escaping flame was increased by 25 per
cent, and the production of iron by from 16 to 20 per cent,
whilst the waste or loss of iron was lessened by several per
cents.
Similar results have been obtained in Westphalia* (Hörde,
&c.), where, besides the decreased waste of grate bars, the
welding of the iron was more perfect, as less air is admitted
through the working opening when the blast is applied under
the grate.
Surprisingly large productions have been obtained in
Southern France by applying the blast below the grate.
b. Brown Coal is used at Maximilianshütte (page 772);
Prävali (page 773), consuming from 126 to 129 lbs. of brown
coal per 100 lbs. of blooms; at Krems (page 773), consuming
170 lbs. of coal per 100 lbs. of finished iron; at Leoben (page
773), Lilienberg, Eibiswald, Meutern, Donawitz, and Neu-
berg in Styria. Sometimes the consumption of coal decreases
to 100 lbs. or even 70 lbs.
c. Turf, used at Maximilianshütte, &c. (page 773).
d. Wood is a good material when kiln-dried; at Feistritz
in Carinthia 12 cubic feet of wood (compact mass) are con-
sumed per 100 lbs. of rolled bars produced. Wood is also
used at Neuberg in Styria, Wolfsberg and Lippitzbach in
Carinthia, Lesjöfors in Sweden, Kallich and Reichenau in
Bohemia, &c.
Fuel in the form of Gas produces the highest effect in
Siemens's regenerator furnaces; compared with common
reverberatory furnaces, about 66 per cent of fuel is saved.
The fuel employed may be of inferior quality, and, according
to Scheerer, the temperature may rise to 4000° C.
Waste gases of iron blast furnaces have been experiment-
**
Berggeist, 1860, No. 69. HARTM., Fortschr., v. 209.
PUDDLING MATERIALS.
779
ally employed at Ludwigshütte in Hesse, but without any
good results.
Wood Gases give good results; they were first employed
in 1842 at Audincourt and Bourguignon in France, after-
wards at Hammerau, and on a large scale at Lesjöfors and
Surahammer in Sweden in Eckmann's reverberatory furnace.
In this furnace 4 or 5 cubic feet of wood were consumed at
Lesjöfors per 100 lbs. of wrought-iron, and from 5 to 7 cubic
feet at Rhonitz.
nace.
Charcoal Gases* are also employed in Eckmann's fur-
In Swedish iron works 3 cubic feet of charcoal were
consumed per 100 lbs. of finished iron; 6.52 cubic feet at
Hirschwang near Reichenau, and 6 or 7 cubic feet at Rhonitz.
The furnaces are more easily conducted, and may be kept
cleaner if heated with kiln-dried wood than if charcoal is
used. Wood which has not been kiln-dried does not produce
the required high temperature.
Turf Gases are also often employed, at Andervillierst in
Switzerland for example, where 0138 cubic metres or
85 kilos. of turf are consumed per 100 kilos. of finished iron.
They are also used in Sweden for heating Eckmann's fur-
nace, and at Buchscheiden (page 775), consuming 144 cubic
feet of turf per 100 lbs. of hammered blooms.
C. Fluxes.
A. In the Refining of Pig-Iron.-When performing this
operation in hearths, refinery slags, fluor-spar, or lime are
sometimes added, and lime alone if refining in reverberatory
furnaces. Calcareous fluxes facilitate the separation of sul-
phur and phosphorus, but prolong the process, as they
produce a thickly liquid slag, which impedes the influence of
the blast; but the extra expense thus incurred is balanced
by the superior quality of the refined metal. An addition of
slag is advisable when treating pig-iron containing only a
small amount of silicon and not producing a sufficient
quantity of slag by itself, for an insufficient amount of slag
*
↑
B. u. h. Ztg., 1863, p. 104, No. 12. TUNNER, Jahrb., 1852, p. 233.
Polyt. Centr., Bd. 143, p. 254. B. u. h. Ztg., 1857, p. 26.
780
IRON.
impedes the process and causes the production of an inferior
and non-uniform iron.
B. In the Puddling Process.-a. Slags rich in oxygen,
chiefly those of the re-heating furnaces, finery hearths, and
rich puddling slags; it is best to use these slags or cinders
in a roasted state, when they will react as in the finery
process (page 736). Hammer slag, rich and pure iron ores, &c.,
also facilitate the puddling process and increase the yield.
In English and Scotch iron works, for instance, the red iron.
ore from Lancashire and Cumberland is generally employed.
Wrought-Iron Scraps are sometimes added to the melted
iron after having used the first paddle or ringer; they
assist the iron in coming to nature, and, owing to their
metallic iron, they are to be preferred to hammer slag, &c.
Whilst more oxidising fluxes are given in addition at the
production of fibrous iron, fewer basic slags are added when
aiming at the production of fine-grained iron.*
b. Schafhäutl's Powder is used principally when treating
pig-iron containing sulphur, and oftener when producing.
fine-grained iron.t The powder is added in several portions.
enveloped in paper cornets; the first portion is usually added
immediately after having melted down the iron. Some-
times, instead of a mixture of brown stone and common salt,
each component is employed by itself.
Couailhac uses the following compositions at different
periods of the process :-50 parts of iron ore, 50 of potters'
clay, 20 of lime, and 10 of common salt when melting down;
immediately after 10 parts of slag obtained from hammering
the iron, 10 of rich slag from the rolls, 2 of common salt,
3 of lime; next 12 parts of ore, 3 of slag from the hammer;
and after the iron comes to nature 2 parts of slags from
the rolls, I of lime, and of common salt.
c. According to Richter, litharge facilitates the oxida-
tion of sulphur better than brown stone, as owing to its
greater fusibility, it comes more intimately into contact with
HARTM., Fortschr., v., 199.
↑ Ibid., 204.
Berggeist., 1861, p. 76.
|| LEOB., Jahrb., 1860, x., 505.
PUDDLING APPARATUS.
781
the iron; sometimes, also, brown stone contains a little
copper. At Wolfsberg favourable results were obtained
from the application of litharge, but at Zeltweg* the re-
sults were less satisfactory.
d. Steam in its usual state has a cooling reaction, but
when superheated it is advantageously employed for the
extraction of sulphur, at Dowlaist for instance; but the
apparatus required is exceedingly expensive.
Fleury suggests that an electric current employed in the
puddling process has a purifying action upon the iron.
C. In the Re-heating Process.-When re-heating in
hearths and in reverberatory furnaces, sand is thrown on to
the heated iron, thus dissolving the oxidised iron and pre-
venting a further oxidation; the iron will then weld
thoroughly together. Nasmyth suggests the possibility of
expelling the oxidised iron by the influence of blows of the
hammer, and obtaining a more intimate combination by
making the ends of the iron to be welded together of a convex
shape.
It is said that better results are obtained by brushing the
iron to be welded with a concentrated solution of water
glass than by using sand.
Puddling Apparatus.
Puddling works should contain the following plant :—Ap-
paratus for refining the pig-iron (refinery hearths or reverbera-
tory furnaces); puddling furnaces for converting the crude
iron into wrought-iron; hammers, rolls, and squeezers of
different construction for blooming the puddled balls; re-
heating hearths or reverberatory furnaces, hammers and
rolls for drawing out the blooms into bars; besides tools of
different kind.
The apparatus for refining the pig-iron have been de-
scribed on pages 721 and 729, and though they are more often
* LEOB., Jahrb., 1861, xi., 300.
† TRURAN, Manufacture of Iron.
B. u. h. Ztg., 1861, p. 439.
|| Ibid., 1862, p. 343.
782
IRON.
employed for the puddling process than for the finery pro-
cess, their use has been limited to a great extent since the
introduction of the boiling process.
Puddling Furnaces.-Their construction depends chiefly
on the nature of the fuel (furnaces with direct firing, or
heated with gas), on the quality of the iron to be produced
(whether the iron is to be fibrous or fine-grained), on the
nature of the pig-iron, on the size of the production (single
and doubles furnaces), on the consumption of fuel, on the
utilisation of the waste heat, &c. Under all circumstances
the furnaces must be constructed so as to produce a tem-
perature as high as the welding heat of iron* (about 1600° C.).
Plagget states that from 50 to 80 per cent of the tempera-
ture produced in a puddling furnace escapes through the
flue.
A puddling furnace has the following principal parts :—
1. Fire Place.-Its construction and size vary chiefly
with the nature of the fuel; this will be explained in the
subsequent descriptions.
Firing with Coal.-The following grates are employed.
according to the tendency the coal has to cake :-
Plane Grates, more or less inclined. The grate is often
inclined upwards towards the fire bridge, in order to produce
a more uniform fire. On perfectly level grates the fire is
less lively on the back part, that at the side of the fire bridge ;
when the grate is raised at the back a thinner layer of fuel
will cover it, allowing a more perfect permeation of the atmo-
spheric air. This arrangement also promotes a more inti-
mate contact of the flame with hot air before entering the
furnace hearth, thus more perfectly consuming the smoke.
When the grate inclines the other way, the interstices
between the grate rods are better kept open, but the fuel
accumulates on the back part of the grate, and is more
charred than burnt. This construction is sometimes adopted
when large flaming coal is burned.
* SCHEUCHENSTUEL, Entwickelung des theoret. Windbedarfs und der erreich-
baren Hitzgrade bei den Braunkohlengas-Puddelöfen in Zerrenner's Gas-
feuerung, 1856, p. 39.
B. u. h. Ztg., 1861, p. 235.
PUDDLING APPARATUS.
783
In puddling furnaces burning from 70 to 90 kilos. of coal
per hour, the grate measures about o'7 or o‘9 square metre,
the hearth three times as much, and the chimney is 12 or 14
metres high, its section measuring o'2 metre, or a quarter the
size of the grate. The grate bars lie from 15 to 20 milli-
metres apart, and the effective part of the grate is of the
whole grate when burning good coal. The fire place of re-
heating furnaces is somewhat larger.
The stoking hole, which must be kept closed, is placed from
30 to 35 centimetres above the grate, and measures about
0'04 square metre.
Clinker grates are occasionally employed, and sometimes
two plate grates side by side.
A puddling furnace used at Königshütte in the Hartz is re-
presented by Figs. 233 to 238. A is the fire-place; B, the
FIG. 233.

}
h
D
d
C
N
d
B-
A
k
L
5
10F!
4
hearth; c, the flue; D, the chimney. a are 17 wrought-iron
grate rods 1 inches square with interstices inch wide
between them; b are cross bars supporting the grate bars;
c is a cast-iron plate supporting the brickwork above the
grate; d is an opening through which the fuel is thrown on
the grate; e, an iron case fixed before that opening; f, furnace
lining of fire-bricks; g, h, walling of common bricks; i, cast-
iron plates surrounding the furnace walling provided with
grappling irons; k, opening admitting air below the grate;
7, fire-bridge formed of fire-bricks; m, the furnace roof, also
784
IRON.
formed of fire-bricks; ", cast-iron bottom plate consisting of
two parts; o is a cast-iron hollow box surrounding the fur-
nace hearth; it is closed at the ends, p, and covered with
fire-bricks; q is a tube introducing cold water into the
FIG. 234.

C
b
b
T
S
0
I'
点
​D
70
พ
p
10 F*
d
C
channel of the cast-iron sides; after circulating in the sides
the water flows out at r into the trough, s, used for cooling
the tools; t is a pipe connected with the cast-iron sides for
FIG. 235.

5LLE.
b
J
y
p
q
نا
i
Vay
i
5
10F
evolving the steam formed; u is the working opening of the
furnace, I foot 4 inches high and 1 foot 6 inches broad; it
is furnished with the door, v, I foot 6 inches high, I foot
7 inches broad, and 4 inches thick. This door may be
PUDDLING APPARATUS.
785
drawn up, and contains an aperture 4 inches square, which
may be closed by means of a plate (Fig. 238). w is an iron
plate I foot 10 inches long, 6 inches broad, and 2 inches
thick, fixed before the working opening; x, the tapping-hole
FIG. 236.



2
(Scale also for Figs. 237 and 238.)
3Ft
for the slag in the level of the bottom plate, n; y are standards
supporting the chimney; z is a damper movable by the
damper rod, a'; b' is a vault to diminish the amount of brick-
work and to cool the lower part of the chimney; c' is a cast-
FIG. 238.
FIG. 237.
CC
iron plate provided with a gutter inch deep for removing
slag from the chimney; d', an opening conducting the slag
into c'; e' is a funnel kept filled with burning coal to prevent
a chilling of the flue slag; g' are iron plates covering the
foundation; h' are iron plates in the chimney walling.
FIG. 239.
Section E, F, G.

C.
A
D
C
b
VOL. II.
3 E
D
B
B
10 Ft
786
IRON.
Larger grates with blast under them are sometimes used
to produce higher temperatures in the furnaces.
Step-Grates combined with plane grates are employed
when burning dry coal; for instance, at Alvenslebenhütte*
FIG. 240.

y
y
h
==
C
מ
o
.
30
ас
у
$
10Ft
in Upper Silesia, where they are used in puddling furnaces
both for fibrous and for fine-grained iron. A furnace for the
production of fine-grained iron has a deeper hearth, and the
process is carried on with the addition of poorer cinder,
i
FIG. 241.
Section A, B.

y
40ft
D
E
C/
LELA
C
7.
T
B
A
ی
10 ს
1350
* Schles. Wochenschr., 1859, 398. HARTM., Fortschr., v., 200.
Ztg., 1862, p. 185.
с
B. u. h.
PUDDLING APPARATUS.
787
whilst richer slag is added when producing fibrous iron.
Owing to the corroding nature of the poor slag the fire-bridge
and the flue-bridge must be cooled either with air or water,
whilst this is required less in the furnaces for dry puddling,
in which the lower part of the side walls is bricked up with
fire-bricks, which concentrate the heat better than cooled
iron walls. The greater height of the fire-bridge in puddling
furnaces for fine-grained iron causes the flame to act less
upon the metal.
The puddling furnace at Alvenslebenhütte for the produc-
tion of fine-grained iron is shown in Figs. 239, 240, and 241.
The cast-iron hearth lining, o, consists of six pieces; to facili-
tate changing the two side pieces are cast hollow, jets of water
are injected into them, and the pieces are provided with feet
(Fig. 241), thus forming a channel for the circulation of
air. Two of the pieces slightly curved form the back wall,
two smaller three-cornered pieces form the combination
D
TT
2'
FIG. 242.
Section D, E.
1

C
B
R
77)
78
6 Et
A
with the working door. The cast-iron bottom plate consists.
of three pieces, and is provided with projections on its sur-
face indicating the position of the six cast-iron side pieces.
The other letters of the figures have the same signification
as those used in Figs. 233 to 235.
3 E 2
788
IRON.
A puddling furnace at Alvenslebenhütte for the common
puddling process is shown in Figs. 242 and 243. The side
walls, o, of the furnace hearth are formed of fire-bricks, and
rest upon the bottom plate, n, which consists of three plates,
FIG. 243.
Section B, C.

D
C
d
1
C'
f
77.
и
B
7
d
น
5
G'
e
20 Ft
E
曰​:
and is supported by a cast-iron frame. On the side of the
working opening, u, the furnace is provided with a second
opening, u', which may be closed with a door. The flue-
bridge, f, has a hole, m', for conducting the slag to the
opening, d', by means of the gutter, c'. This hole is kept
closed with sand, coal, and slag, and is opened according to
FIG. 244.

α
a
7
h
k
5
10 Ft
requirement. The other letters of the figures are the same
as in Figs. 233 to 235. At Alvensleben the cost of erecting
a furnace of this kind is 1330 thalers, or about £200.
Puddling furnaces with Brown Coal Firing are of the same
construction as the furnaces for mineral coal, and vary
according to the quality of the brown coal.
SIEMENS'S REGENERATIVE GAS FUrnace.
789
Furnaces with plane grates, and with or without the appli-
cation of blast, are used. To heat the blast it is made to
circulate in the fire-bridge and the flue-bridge, and also in a
worm tube below the grate heated by the cinder falling from
the grate. The blast is introduced into the furnace above
the fire-bridge by means of seven blast pipes, each of an
inch in diameter.
3
Step grates with and without the application of blast are
likewise in use. The furnace represented by Fig. 244 is used
at Maximilianshütte for the employment of turf as fuel. a is
the fire-hole; b, the grate; d, the fire-bridge into which cold
blast is conducted by means of the tubes, c; the blast then
enters the wind-box or reservoir, f, the upper part of which
is provided with a slit 26 inches long and 3 of an inch wide,
evolving the blast against the roof of the furnace, thus pro-
tecting the finished balls from oxidation. The reservoir
is cast outside with spikes to consolidate its coating of loam.
h is a space below the flue used for a preparatory warming of
the iron to be puddled.
The part of the flue is covered with iron, thus conducting
heat into the chamber, k, in which the turf is dried; I is the
chimney. At the commencement of the operation the pres-
sure of the blast is equal to 2 inches of water, during smelt-
ing 6 inches, and when forming the balls from 2 to 3 inches.
When using wood as fuel the grate is formed either of
bricks or a perforated iron plate; sometimes the air is intro-
duced from above.
The construction of furnaces for the use of gaseous fuel
varies in the different establishments, chiefly according to the
material employed for the production of the gas.
Amongst these furnaces Siemens's regenerative gas furnace.
is of the greatest importance. As the principle of Siemens's
furnace is suited to various operations requiring high tem-
perature, such as melting steel and glass, re-heating iron,
&c., it may be justly considered as one of the greatest and
most ingenious inventions on the utilisation of fuel. We
therefore propose to describe this furnace in extenso from
communications which have been kindly made to us by
Mr. C. W. Siemens, F.R.S.:
790
IRON.
The advantages of the regenerative gas furnaces are:-
1. Saving of fuel, amounting to 40 or 50 per cent in quantity;
also the most inferior qualities of fuel, such as slack, coke
dust, lignite, and peat, may be employed, producing a money
saving in many instances amounting to 75 per cent.
2. Increase of work done per day in a furnace of given
dimensions, amounting to 30 per cent or more, owing to the
unlimited command of heat, with low chimney draught.
3. Great purity of flame, which greatly diminishes the
oxidation or deterioration of the metal heated in the furnace.
4. Increased durability of the furnace, owing to the absence
of ashes and a perfect uniformity of heat throughout the
furnace.
5. Saving of space within the works and great cleanliness
of operation, the fuel being converted into gas outside the
works.
6. Complete command of the intensity of the heat and of
the chemical nature of the flame, which may be arrested or
changed from a reducing to an oxidising flame, or the reverse,
at any moment, thus tending to facilitate and improve all
metallurgical operations.
7. Complete absence of smoke from the stack, which pre-
vents this furnace being injurious in large towns.
The late Professor Faraday, in his last Friday evening
lecture at the Royal Institution, on the 20th June, 1862,
described these furnaces in the following terms :—
"The gaseous fuel is obtained by the mutual action of
coal, air, and water, at a moderate red heat. A brick chamber,
perhaps 6 feet by 12 feet, and about 10 feet high, has one of
its end walls converted into a fire-grate, i.e., about half way
down it is a solid plate, and for the rest of the distance con-
sists of strong horizontal plate bars where air enters, the
whole being at an inclination such as that which the side of
a heap of coals would naturally take. Coals are poured
through openings above upon this combination of wall and
grate, and being fired at the under surface, they burn at the
place where the air enters; but as the layer of coal is from 2 to
3 feet thick, various operations go on in those parts of the
fuel which cannot burn for want of air. Thus the upper and
SIEMENS'S REGENERATIVE GAS FURNACE.
791
cooler part of the coal produces a large body of hydrocarbons.
The cinders or coke which are not volatilised approach in
descending towards the grate. That part which is nearest
the grate burns with the entering air into carbonic acid, and
the heat evolved ignites the mass above it. The carbonic
acid passing slowly through the ignited carbon becomes
converted into carbonic oxide, and mingles in the upper part
of the chamber (or gas producer) with the hydrocarbons.
The water, which is purposely introduced at the bottom of
the arrangement is first vaporised by the heat, and then
decomposed by the ignited fuel, and re-arranged as hydrogen
and carbonic oxide, and only the ashes of the coal are re-
moved as solid matter from the chamber at the bottom of
the fire-bars.
These mixed gases form the gaseous fuel. The nitrogen
which entered with the air at the grate is mingled with them,
constituting about a third of the whole volume. The gas
rises up a large vertical tube for 12 or 15 feet, after which it
proceeds horizontally for any required distance and then
descends to the heat regenerator, through which it passes
before it cnters the furnaces. A regenerator is a chamber
packed with fire-bricks, separated so as to allow of the free
passage of air or gas between them. There are four placed
under a furnace. The gas ascends through one of these
chambers whilst air ascends through the neighbouring cham-
ber, and both are conducted through passage outlets at one
end of the furnace, where, mingling, they burn, producing
the heat due to their chemical action. Passing onwards to
the other end of the furnace, they (i.e., the combined gases)
find precisely similar outlets down which they pass; and
traversing the two remaining regenerators from above down-
wards, heat them intensely, especially the upper part, and
so travel on in their cooled state to the shaft or chimney.
Now the passages between the four regenerators and the
gas and air are supplied with valves and deflecting plates,
which are like four-way cocks in their action, so that by the
use of a lever those regenerators and air-ways which were
carrying off the expended fuel can in a moment be used for
conducting air and gas into the furnace; and those which
792
IRON.
just before had served to carry air and gas into the furnace now
take the burnt fuel away to the stack. It is to be observed
that the intensely heated flame which leaves the furnace for the
stack always proceeds downwards through the regenerators,
so that the upper part of them is most intensely heated,
keeping back as it does the intense heat; and so effectual are
they in this action that the gases which enter the stack to
be cast into the air are not usually above 300° Fahr. of heat.
On the other hand, the entering gas and air always pass up-
wards through the regenerators, so that they attain a tem-
perature equal to a white heat before they meet in the furnace,
and there add to the carried heat that due to their mutual
chemical action. It is considered that when the furnace is
in full order, the heat carried forward to be evolved by the
chemical action is about 4000°, whilst that carried back by
the regenerator is about 3000°, making an intensity of power
which, unless moderated on purpose, would fuse furnace.
and all exposed to its action.
"Thus the regenerators are alternately heated and cooled
by the outgoing and entering gas and air, and the time for
alteration is from half an hour to an hour, as observation
may indicate. The motive power on the gas is of two kinds;
a slight excess of pressure within is kept up from the gas
producer to the bottom of the regenerator to prevent air
entering and mingling with the fuel before it is burnt, but
from the furnace downward through the regenerators the
advance of the heated medium is governed mainly by the
draught in the tall stack or chimney.
"Great facility is afforded in the management of these
furnaces. If (applying the furnaces, for instance, to the
manufacture of glass), whilst glass is in the course of manu-
facture an intense heat is required, an abundant supply of
gas and air is given; when the glass is made and the con-
dition has to be reduced to working temperature the quantity
of fuel and air is reduced. If the combustion in the furnace
is required to be gradual from end to end the inlets of air
and gas are placed more or less apart the one from the other.
The gas is lighter than the air, and if a rapid evolution of
heat is required, as in a short puddling furnace, the mouth
SIEMENS'S REGENERATIVE GAS FURNace.
793
of the gas inlet is placed below that of the air inlet; if the
reverse is required, as in the long tube welding furnace, the
contrary arrangement is used. Sometimes, as in the enamel-
ler's furnace, which is a long muffle, it is requisite that the
heat be greater at the door end, because the goods being put
in and taken out at the same end, those which enter last and
are withdrawn first, remain, of course, for a shorter time in
the heat at that end; and though the fuel and air enters first
at one end and then at the other alternately, still the necessary
difference of temperature is preserved by the adjustment of
the apertures at the ends.
"Not merely can the supply of gas and air to the furnace
be governed by valves in the passages, but the very pro-
duction of the gas fuel itself can be diminished, or even
stopped, by cutting off the supply of air to the grate of the
gas producer, and this is important, inasmuch as there is no
gasometer to receive and preserve the aëriform fuel, for it
proceeds at once to the furnaces.
"Some of the furnaces have their contents open to the
fuel and combustion, as in the puddling and metal melting
arrangements; others are enclosed, as in the muffle furnaces
and flint-glass furnaces.
"The economy in the fuel is estimated practically as one-
half, even when the same kind of coal is used either directly
for the furnace or for the gas producer, but, as in the latter
case, the most worthless kind can be employed, such as
slack, &c., which can be converted into a clean gaseous fuel
at a distance from the place of the furnace, so many ad-
vantages seem to present themselves in this part of the
arrangement."
Professor Faraday concludes his lecture with the following
calculations:
"Carbon burnt perfectly into carbonic acid in a gas pro-
ducer would evolve about 4000° of heat; but if burnt into
carbonic oxide it would only evolve 1200°. The carbonic
oxide in its fuel form carries on with it the 2800 in chemical
force, which it evolves when burning in the real furnace with
a sufficient supply of air. The remaining 1200° are employed
in the gas producer in distilling hydro-carbons, decomposing
794
IRON.
water, &c. The whole mixed gaseous fuel can evolve about
4000° in the furnace, to which the regenerator can return
about 3000 more."
Figs. 245, 246, 247 and 248 show the general arrangement
of the furnace. It consists of two essential parts :—
The gas producer, in which the coal or other fuel is con-
verted into combustible gas; and
The furnace with its "regenerators" or chambers for storing
the waste heat of the flame, and giving it up to the in-coming
air or gas.
FIG. 245.

O
A
I
E
C
H
G
Ꭻ
Section of the Gas Producer.
The puddling furnace is of very nearly the ordinary form.
The gas producer is shown in Fig. 245; it is a rectangular
fire-brick chamber. The side, B, is inclined at an angle of
SIEMENS'S REGENERATIVE GAS FURNACE.
795
from 45° to 60°, and has the grate, c, at its foot. The fuel
of any description, coal, coke, lignite, peat, or even saw-dust,
is filled in through a hopper, A, at the top of the incline, and
falls in a thick bed upon the grate. Air is admitted at the
grate, and on burning its oxygen unites with the carbon of
the fuel, forming carbonic acid gas, which rises slowly
through the ignited mass, taking up an additional equivalent
of carbon, and thus forming carbonic oxide. The heat thus
produced distils off carburetted hydrogen and other gases
and vapours from the fuel as it descends gradually towards
the grate, and the carbonic oxide diluted by the inert nitro-
gen of the air and any small quantity of unreduced carbonic
acid, and mixed with these gases and vapours distilled from
the raw fuel, is finally led off by the gas flue to the furnace.
The ashes and clinkers accumulated in the grate are re-
moved at intervals of one or two days.
E is a pipe supplying a little water to the ash-pit, to be
decomposed as it evaporates and comes in contact with the
incandescent fuel, thus forming hydrogen and carbonic oxide,
which serve to enrich the gas; G is a small plug-hole by
which the state of the fire may be inspected, and the fuel
moved by a bar if necessary, and D is a sliding damper by
which the gas producer may be at any time shut off from the
flue.
A slight outward pressure must be maintained through
the whole length of the gas flue leading to the furnace, in
order to prevent the burning of the gas in the flue through
the in-draught of air through crevices in the brickwork.
Where the furnace stands much higher than the gas
producers, the required pressure is at once obtained; but
the furnaces and gas producers are more often placed nearly
on the same level, and some special arrangement is neces-
sary to maintain the pressure in the flue. The most simple
contrivance for this purpose is the "elevated cooling tube."
The hot gas is carried up by a brick stack, H, to a height of
8 or 10 feet above the top of the gas producer, and is led
through a horizontal sheet-iron cooling tube, J (Fig. 245),
from which it passes down either directly to the furnace or
into an underground brick flue.
796
IRON.
The gas rising from the producer, at a temperature of about
1000° Fahr., is cooled as it passes along the tube above, and
the descending column is consequently denser and heavier
than the ascending column of the same length, and con-
tinually overbalances it. In fact, the system forms a
syphon in which the two limbs are of equal length, but one
is filled with a heavier gaseous fluid than the other.
In erecting a number of gas producers and furnaces it is
generally preferable to group the producers together, leading
the gas into one main flue, from which the several furnaces
are supplied.
The puddling furnace proper is shown in Figs. 246, 247,
and 248.
FIG. 246.

Front Elevation of Puddling Furnace.
Fig. 246 is a front elevation of the furnace, showing the
gas reversing valve and flues in section.
Fig. 247 is a longitudinal section at A, B, C, D (Fig. 248).
SIEMENS'S REGENERATIVE GAS FURNACE.
797
C
I
FIG. 247.

E
·E,
洗​慈
​「豬​藏珍​殊​孩​饭
​麻​豬​海
​斯​烤
​86 187 39 10 226, 226,200 108 107 A
海鮮
​滋味
​H
M
您​溪​滾​源​开
​滋​源
​魚​鹽​洗​吸
​Fig. 248 is a sectional plan at L, M (Fig. 247).
FIG.248.
F
K
C
A
'B
F
C
F
K
級數​:

D
The peculiarity of the regenerative gas furnace, as ap-
plied to puddling, or to any process in which a high heat is
required, consists in the utilisation in the furnace of nearly
the whole of the heat of combustion of the fuel, by heating
the gas and air as they enter by means of the waste heat
of the products of combustion after they have left the furnace
and are of no further use for the operation. The waste
798
IRON.
heat is, so to speak, intercepted on its passage to the
chimney by means of masses of fire bricks stacked in an
open or loose manner in certain chambers, called "regenera-
tor chambers,” C, E, E', c' (Fig 247).
On first lighting the furnace the gas passes in through
the gas regulating valve, B (Fig. 246), and the gas reversing
valve, B', and is led into the flue, M, and thence into the
bottom of the regenerator chamber, c (Fig. 247); while the
air enters through a corresponding "air reversing valve"
behind the valve, B', and passes thence through the flue, N,
into the regenerator chamber, E. The currents of gas and
air, both quite cold, rise separately through the regenerator
chambers, c and E, and pass up through the flues, G, G, and
F, F, F (Fig. 248) respectively, into the furnace above,
where they meet and are lighted, and burn, producing a
moderate heat. The products of combustion pass away
through a similar set of flues at the other end of the furnace
into the regenerator chambers, c', E', and thence through
the flues, M', N', and through the gas and air reversing
valves, into the chimney flue, o. The waste heat is thus
deposited in the upper courses of open fire-brick work filling the
chambers, c', E', so heating them while the lower portion
and the chimney flue are still quite cool; then in about an
hour the reversing valves, B' (through which the gas and
air are admitted to the furnace) are reversed by the levers, P,
and the air and gas enter through the regenerator cham-
bers, E', c', which have just been heated by the waste pro-
ducts of combustion, and in passing up through the open
brickwork they are heated, and then on meeting and entering
into combustion in the furnace, D, D, they produce a very
high temperature, probably 500° Fahr. higher than if ad-
mitted cold; the waste heat from this higher temperature of
combustion heats the previously cold regenerator chambers, C, E,
to a correspondingly higher heat. After about an hour's work,
the reversing valves are again reversed, and the air and
gas enter the first pair of regenerator chambers, C, E, which
are now very hot, and therefore the air and gas become very
hot, and on entering the furnace in this state they meet
and enter into combustion, thus producing a still higher
SIEMENS'S REGENERATIVE GAS FURNACE.
799
temperature, probably 500° higher still, and again heating the
second pair of regenerator chambers, c', E', so much higher,
which enables them again to heat the air and gas to a still
higher degree, when the valves, B', are reversed. Thus
we obtain an accumulation of heat and an accession of
temperature step by step, so to speak, until the furnace
reaches the required heat; for unless cold materials are put
in to be heated, and thus abstract heat, the temperature
rises as long as the furnace holds together, and the supply
of air and gas is continued. The heat is at the same time
so thoroughly abstracted from the products of combustion
by the regenerators that the chimney flue is always quite
cool. The command of the temperature of the furnace and
of the quality of the flame is made complete by means of
the gas and air regulating valves shown at B, in Fig. 246,
and by the chimney damper. These are opened to any re-
quired extent by the notched rods, Q, R, and s (Fig. 246)
respectively, so that having the power of producing as high
a temperature as can be desired, there is also the power of
varying it according to the requirements of each individual
case.
The bed of the furnace, D, D, is of the ordinary construc-
tion, formed on iron plates, and is provided with water-
bridges at the ends, as shown, to protect the "fettling" (or
oxide of iron used for lining the furnace) from being melted
away. The overflow of one of the water bridges is led into
a sheet iron tank below the bed and then away. The
evaporation from this tank keeps the bottom plates cool
and preserves their cinder covering from melting off, and the
steam is carried away by a draught of air entering through
two holes, I, I (Fig. 246), below the tap hole, and passing
off by small ventilating shafts, K, K (Fig. 248), at the back
of the furnace.
A heating chamber, H, is arranged at each end of the
furnace, in which the charge of pig-iron may be heated to
redness before it is introduced into the puddling chamber, D, D.
The advantages of this furnace for puddling are that the
heat can be raised to an almost unlimited degree, that the
flame can be made at will, oxidising, neutral, or reducing,
ง
800
IRON.
without interfering with the temperature, that the in-draughts
of air and cutting flames are avoided, and that the gas fuel
is free from ashes, dust, and other impurities, which are
carried into an ordinary puddling furnace from the grate.
In this last respect the new furnace presents the same ad-
vantages as puddling with wood.
2. Ash-Pit. The ash-pit is usually kept open, and some-
times the upper part is closed, thus forming below the grate
a space for the production of hot air. When employing
blast under the grate the ash-pit must be perfectly closed;
also it must be closed when conducting atmospheric air
under the grate from outside the building by means of a
channel or pipe. About o°8 or 1 metre below the grate the
ash-pit may be advantageously provided with a sole of
water.
3. The Furnace Hearth (Figs. 233 to 238) consists of
the following parts:-
a. Cast-iron bottom plate, 1, 3 or 4 inches thick. In order
to facilitate the replacing of this plate it is frequently formed
of two or three parts, which are joined by grooves and luted.
On the sides the plate rests either on brick-work or in an
iron frame, and the middle is supported by cast-iron bearers
or bricked-up pillars, whilst air for cooling is admitted below.
The dimensions of the hearth vary with the size of the
charges (from 2 to 10 cwts.; the hearth usually comprises an
area of about 20 cubic feet, that is to say, it is 4 or 5 feet long.
and broad, and of a shape allowing throughout a free manipu-
lation with the tools. In rare cases the hearths are movable.
b. The Side Walls of the Hearth.-They are usually
made either wholly or partially of fire-bricks, and sometimes
of massive pieces of cast-iron if they come into contact with
rich basic slags only. But hollow cast-iron boshes are
usually inserted all around the hearth, which is kept cool by
a circulation of air, blast, or water. The hole in the boshes.
is 2 or 2 inches wide, and is square, round, or oval. Some-
The
times water is only injected in the hollow boshes.
cast-iron sides or boshes are cast in several pieces, which
facilitates their replacing. The steam formed in the boshes
is emitted by a safety-tube (Fig. 233, t,) and when cooling
PUDDLING APPARATUS.
801
the boshes with air they have a pipe sometimes 20 feet high,
serving to increase the draught. Cooling with air is simpler
but it is less effective when puddling grey thinly liquid iron.
To increase the effect of the air the boshes are connected
with a chimney, or the air is introduced in the form of
blast.
When employing cast-iron sides the dimensions of the
hearth are not changed by the puddling process, as these
sides only waste slightly; neither do they contribute to the
scorification of iron as is effected by sides of fire-bricks.
The cast-iron sides are usually lined inside with "bull dog,"
iron-ore, limestone, or some other material.* The back of
the furnace-hearth is sometimes formed of pieces of wrought-
iron or iron lumps.
The height of the hearth sides is from 7 to 16 inches, and
is varied according to whether the production of fibrous or
fine-grained iron is intended. The furnaces for the fine-
grained iron require a stronger draught for the production.
of very high temperatures, and well made dampers for the
regulation of the temperature.
4. The Fire-Bridge, 1 (Figs. 233 to 238), is occasionally
constructed of fire-bricks, either massive or hollow, for cooling
by a circulation of air; but it is almost everywhere now
a hollow piece of cast-iron like the other hearth sides, of
which the fire-bridge forms a part. This piece of iron
forming the fire-bridge is lined on the top with fire-bricks,
and on the side nearest to the grate. The bridge is 12 or
15 inches broad, and its height above the bottom plate
varies from 12 to 20 inches, according to whether the pro-
duction of fibrous or fine-grained iron is intended, and how
much the iron is to be protected from the influence of the
flame.
The aperture between the fire-bridge and roof is 10 or
12 inches high, and in section it is usually half as large as
the grate; the grate is generally placed in the level of the
hearth-plate or somewhat deeper, but never higher.
HARTM., Fortschr., iii., 232; iv., 200. B. u. h. Ztg., 1861, p. 372; 1863,
p. 272.
VOL. II.
3 F
802
IRON.
5. The Flue-Bridge, f' (Figs. 233 to 238), is of the same
construction as the fire-bridge, and usually some inches
lower than the fire-bridge, thus allowing the cinder to run
out on the sole of the flue by the aperture, d', which is sur-
rounded with ignited coals. Sometimes the cinder is tapped
off by a gutter (m' in Fig. 243) in the flue-bridge.
6. The Furnace Roof, m (Figs. 233 to 244).—The roof
is either gradually inclined from the fire-bridge towards the
bridge of the flue, its distance from the hearth or bottom
plate being 20 or 30 inches; or the roof is horizontal till
it comes near the flue-bridge. The thickness of the roof
seldom exceeds 6 or 8 inches, as thicker roofs are liable to
become fused.
The single furnaces have on one of their long sides the
working opening, u, 16 or 18 inches square and 8 or 12 inches
above the bottom plate. Double furnaces have one of
these openings on each of their long sides. The openings
are closed with a sliding door, v.
The double furnaces have larger hearths and grates, and
may be worked by two men at the same time through the
working doors on both sides of the furnaces; they admit of
larger charges (up to 10 cwts.) and a larger production,
whilst they are economical in fuel and in the erection.
On the other hand, the resulting wrought-iron is not
likely to be uniform in quality, and a greater loss of iron
takes place than in single furnaces on account of the in-
creased admission of air and the greater length of the
operation. Double furnaces may, therefore, be advantageous*
when working good pig-iron with expensive fuel (Carinthia).
A double furnace treating pig-iron which has been previously
warmed consumes about half as much fuel as a single furnace
working cold pig-iron.
Overmannt makes the following statements concerning
the use of these furnaces in America :-
"Throughout the States, on the eastern slope of the
B. u. h. Ztg., 1857, pp. 176, 311, 432; 1859, p. 5. Preuss. Ztschr., iv., 236.
LEOB., Jahrb., 1857, vi., 242; 1860, ix., 350. ZERRENNER'S Gasfeuerung, 1856,
p. 176.
+ OVERMANN, Treatise on Metallury, New York, 1865.
PUDDLING APPARATUS.
803
Alleghanies, we do not find many single puddling furnaces
in operation; most of them are double furnaces with opposite
doors, so that two men may be at work at the same time.
The general dimensions of these furnaces are not much
larger than those of a single furnace, and if any advantage
exists in the single furnace over the double one, which is
frequently asserted, it is in the comparatively larger dimen-
sions of the hearth, which admits of the presence of more
cinder in proportion to the iron worked at a time, and also
the absence of the second door, in consequence of which
the hearth is not so much cooled by the entrance of cold
air. In general, the advantages of the double furnace so
far predominate as to cause the gradual disappearance of
the single one. The surface of the grate is here not often
less than 15 square feet, the area of the hearth about
40 square feet, and the area of a section of the flue 2 square
feet. The total length of the furnace, exclusive of stack,
12 or 13 feet, and the width outside 6 feet and from that to
7 feet. Behind the flue, in the prolongation of it, there is
a stove for warming pigs. In some furnaces we find simply
square cast-iron pipes of about 6 inches in width laid across
the furnace, forming the fire and flue-bridges; these pipes
project through the inclosure. By these means the caking
of cinder through the bridges is prevented. In this instance,
the hearth of the furnace is lined with soap-stone or mag-
netic iron ore, which forms the boshes. In most of these
furnaces hollow cast-iron boshes are inserted all around the
hearth, which are kept cool by a circulation of air or blast,
or, which is not often the case, by a circulation of water.
Most of these furnaces are supplied with fresh air under the
grate by means of fan blast. One fan with a diameter of
3 feet and performing from 700 to 800 revolutions supplies two
furnaces. In this case the ash-pit is provided with iron
doors, and the blast conducted in canals underground."
Puddling and re-heating furnaces (mostly the puddling
furnaces with a sand hearth) are sometimes provided with
a working opening near the fire-bridge, and on the side near
*
Polyt. Centr., 1858, p. 794.
3 F 2
804
IRON.
the flue they have a second smaller opening from 10 to 13
inches wide, and 12 or 13 inches above the bottom plate.
The charges are worked alternately through the two openings,
and now and then the smaller opening is used only for
charging the pig-iron. The puddling furnaces at Montataire*
are provided with two openings on each side, allowing four
men to work at once. These quadruple furnaces save fuel
considerably, but the resulting iron is apt to be not uniform.
To prevent cooling of the furnace by an excessive admis-
sion of air through the working opening, the working door is
provided with a smaller aperture, which is wider inside than
outside, so as to afford sufficient room for the tools. It may
be closed by a plate, a (Fig. 237). Sometimes the roof is
constructed with an inclination from the working side towards
the opposite door, thus repelling the entering air. Furnaces
with the blast under the grate act advantageously with regard
to the admission of air. The working door, v (Fig. 236), with
the working aperture is a cast-iron frame lined with fire-
brick. The iron frame is cast with some ears, upon which
may be placed a cross-bar fixed to the enclosing plates of the
furnace, thus firmly fastening the door and preventing it from
being lifted when forming the balls in the furnace. The
same result is obtained by driving in an iron wedge between
one side of the door and the door frame.
The fore plate, w, is about 10 or 12 inches above the bottom
plate and 15 inches below the roof, and serves especially as
the fulcrum for the tool. A tap-hole, x, for running off the
cinder is placed below the fore plate, and kept closed with
clay during the manipulation. The fore plate is 1 or 2 inches
deeper than the flue-bridge, and below it, inside the furnace,
the hearth is lined with brick-work without any circulation
either of air or water. The side walls of the furnace are
enclosed with iron plates, which are firmly fastened with grap-
pling irons. Openings in the plates about I square foot in
area allow the circulation of the air below the bottom plate;
this circulation is sometimes promoted by a special channel,
k (Fig. 233), for the admission of air.
*
Polyt. Centr., 1860, p. 191.
PUDDLING APPARATUS.
805
7. The Flue, c (Figs. 223 to 244), is almost invariably
inclined downwards, thus forcing the flame to the bottom of
the flue, and keeping the slag which passes over the flue-
bridge in a liquid state. The wider the flue the stronger is
the draught, but also the greater the loss of iron. In Eng-
land, when using caking coal, 32 inches square area is given.
to the flue per 1 square foot of the grate, and 21 square
inches when using half anthracitic coal. As in America, the
flue is sometimes prolonged, the prolongation forming a stove
for warming the pigs, chiefly when treating white iron; this
arrangement saves from 10 to 25 per cent of fuel. The
escaping waste heat is also sometimes used for heating
boilers, for drying fuel, and for heating blast when higher
chimneys (above 100 feet) are required.
The previous warming of the pig-iron is the more advan-
tageous the longer the melting down lasts in proportion to
the whole puddling operation, and therefore its advantages
are greater for white pig than for grey.
Warming stoves behind the flue-bridge are advisable if the
puddling process is carried on with but little slag. Some-
times the warming stove on the side of the flue (Low Moor*)
or the lower part of the chimney is used for the purpose.
The large production of the puddling furnaces in Southern
Francet must be partly attributed to the warming stoves.
8. The Chimney, D (Figs. 233 to 244).—Each puddling
furnace is provided either with a special chimney from 40 to
60 feet high, and about a quarter the section of the grate, or
one chimney up to 180 feet high serves for several furnaces at
the same time. On the top of each chimney is a damper
(≈, in Fig. 235) to regulate the draught; the damper may be
opened and shut to any extent by means of the lever and
chain. When using the waste flame a higher chimney (up
to 120 feet) is required; the draught is less influential when
employing blast gas producers or blast under the grate, but
chimney is still necessary to direct the flame. Caddik||
*
HARTM., Fortschr., i., 273.
B. u. h. Ztg., 1863, p. 16.
HARTM., Fortschr., v., 214.
|| Ibid., i., 266.
806
IRON.
cools the chimney with steam. The section of the chimneys.
of some English puddling works* is larger at the top than
at the bottom.
When erecting a puddling furnace the foundation for the
chimney, about 10 feet deep and 6 feet square, is first formed
with large quarry stones and with as little mortar as possible;
the top of this foundation must be level with the ground.
The iron plate, g' (Fig. 233), is placed upon this foundation
and joined to it with grappling irons. This iron plate
bears the four iron pillars, y, which again support the iron.
plate, h'. Both plates are then connected by brick-work,
which forms the lower part of the chimney, and the chimney
is raised above the upper plate and provided with grappling
irons. The further foundation is then formed of a layer of
quarry stones, and of a brick course laid on edge resting upon
the quarry stones, and upon this the fire-place is erected.
The cast-iron plates arming the fire-place together with their
grappling irons are now put in their places and fastened, and
the hearth or bottom plate, n, is placed upon two brick walls
about 7 inches thick, upon which the cast-iron boshes
are arranged in their proper places. After having placed
and fastened the remaining iron plates for arming the fur-
nace, the fire-bridge, l, and the flue, c, are finished. The
roof, m, above the boshes is now formed of fire-bricks, and
the space between the lining iron plates, i, and the boshes
filled up with pieces of bricks.
Dimensions of Puddling Furnaces.-In some English
puddling furnaces for the boiling process the iron plates binding
the furnaces are about 12 feet long, 5 feet broad, and 6 feet
high; 9 or 10 inches of their height lies under ground. The
grate has an area of 8 or 12 square feet according to the
quality of the coal. The roof on the fire-bridge is 26 or 27
inches above the sole of the furnace and 9 or 10 inches on
the flue-bridge. The fire-bridge is 14 or 15 inches broad and
10 or 12 inches below the roof; the flue-bridge is as much
below the roof and 9 inches thick. The furnace hearth is
* HARTM., Fortschr., v., 188. Schles. Wochenschr., 1861, No. 32. DINGL.,
Bd. 161, p. 431.
APPARATUS FOR RE-HEATING THE IRON.
807
about 6 feet long and 3 feet broad; its middle lies 2 feet.
below the roof. The working door is 16 inches square; the
height of the working plate from 8 to 11 inches above the
bottom plate. The stoking hole is 10 inches wide, and the
tap or floss-hole for the slag in the flue is from 4 to 6 inches
wide; the height of the stack is 36 feet, and its width 2 feet
square.
The furnaces for the dry puddling process have similar
dimensions, except the fire-bridge, which is seldom higher
than 6 inches, and the hearth bottom which lies only 8 inches
below the working or fore plate.
The tools used in the manipulation of the molten pig-iron
are paddles and rabbles. The paddle is a strong iron
bar with a round knob on one end. The handle is round or
octagonal, and the part which enters the furnace is square
and terminates in a chisel edge. The rabble or hook is
simply the same tool bent at the chisel end. A rabble for
puddling iron weighs from 40 to 60 lbs., and is from 5 to
7 feet long according to the width of the furnace, whilst a
smaller and lighter rabble is used for collecting the half-fused
masses, for cleaning the hearth, turning the balls, &c. The
waggons for the cinder are made of wrought-iron, and have
a hopper-like shape; the cinder is tapped into them, and
wheeled away. Different kinds of tongs, small and pointed
iron bars for opening the tap-hole, and other minor tools are
also necessary.
Apparatus for Re-heating the Iron.
The compressed balls and the piles are sometimes re-
heated in hearths constructed like finery hearths, or in
hollow fires (page 748), but this manner of re-heating only
gives a small production. Therefore
Re-heating Furnaces are most frequently employed in
connection with puddling works. These furnaces produce
higher temperatures than the puddling furnaces, from which
they differ chiefly in the 'following points:-The grate is
larger in proportion to the hearth, the flue is wider, the roof
less high above the hearth, which is formed of sand and
rests upon massive brick-work instead of an iron plate, thus
808
IRON.
causing less cooling; the fore plate and working door are in the
level of the hearth. The working door is best constructed of
native quartz sand, and has an inclination both towards the
back wall and the flue, thus conducting the cinder towards
the hole for flowing out of the furnace; the furnace has no
flue-bridge. The hearth is about 6 feet long, 4 feet broad,
and the height of the roof is 12 or 14 inches; but the size
of the hearth varies with the size of the charges (from 10 to
40 cwts. and more).
The re-heating furnaces are heated like the puddling
furnaces, either with compact or gaseous fuel. When using
compact fuel, blast under the grate is frequently employed
and has led to a considerable saving in fuel.
With the gaseous fuel Siemens's regenerative principle
(page 789) has given the best results, and proved to be of the
greatest advantage.
Hopfgärtner has considerably lessened the consumption
of fuel by combining puddling and re-heating furnaces in the
following manner :-Very hot combustible gases are con-
ducted from the gas producer downwards into the re-heating
furnace, on the side where the fire bridge is situated in the
common furnaces; hot air for combustion is here admitted at
the same time. Both the gases and air then pass along the
hearth of the re-heating furnace, and on to the hearth
of the puddling furnace, whilst a fresh supply of air is given.
The waste gases escaping are still hot enough to heat a
warming stove, a hot blast apparatus, and a boiler.
When larger pieces of iron are to be welded, or heated up
to welding heat, the re-heating furnaces are sometimes pro-
vided with a second hearth for previously warming the iron
and thus lessening the oxidation as much as possible;
this second smaller hearth is situated between the chimney
and the chief hearth, from which it is separated by a high
bridge and a deep gutter for collecting the cinder and con-
ducting it out of the furnace.
The waste heat of the re-heating furnaces is frequently
employed for other purposes, chiefly for the production of
* Oesterr. Ztschr., 1862, No. 10. HARTM., Fortschr., vi., 212.
RE-HEATING FURNACES.
809
*
steam; at Creuzot, for instance, the re-heating furnaces.
produce 20 kilos. of steam per hour and 1 square metre of
heating surface, whilst the puddling furnaces produce only
10 kilos., owing to the cold air admitted by the working
door.
Figs. 249, 250, and 251 show a re-heating furnace, heated
FIG. 249.

b
e
72
with coal, and used at Alvenslebenhütte in Upper Silesia.
a is the grate; b, the stoke-hole; c, the fire-bridge; d, the
FIG. 250.

A_
B
9
h
α
-D
hearth formed of sand; e, the working opening; f, the flue;
and g, the opening for running out the cinder. h is the
chimney or stack.
*
HARTM., Fortschr., vi., 218. Phönix-Walzwerk in: Bauliche Anlagen,
&c., iii. Jahrg., Lief. i.
810
IRON.
E
I
f
U
་་་
-
FIG. 251.
G

5
d
|
U
a
C
"
10 FI
FT
F
Figs. 252, 253, and 254 represent a re-heating furnace
used at Creuzot for welding large pieces of iron, p, which
FIG. 252.

с
a
d
f
1.
0
3°F
are placed across the furnace hearth through the two oppo-
site working openings; these openings are then walled up
FIG. 253.

市
​C
d
a
12
.10
10 Ft
f
h
RE-HEATING FURNACES.
SII
with bricks. The letters in the drawings have the following
signification -a, the grate; b, the stoke-hole; c, the fire-
FIG. 254.

&
d
h
--་
10 Ft
bridge; d, the hearth formed of sand; e, the working
openings; f, the flue; g, the aperture for emitting the
cinder; h, the chimney.
Figs. 255 and 256 show a re-heating furnace provided with
step-grate and warming hearth, used at Dernoe in Hungary.
FIG. 255.

e
d
И
f
万
​5
10 Ft
a is the grate; b, the stoke hole; c, the fire bridge; d, the
hearth of sand; c, the working opening; f, the flue; g, the
aperture for emitting the slag; h, the chimney; i, the
warming stove for previously heating the iron.
The engravings, Figs. 257 and 258, represent a gas re-
heating furnace* constructed by T. Groebe. This furnace
* Civil Ingenieur, Bd. ix., tab. 23, p. 339.
812
IRON.
་
1.9
FIG. 256.

e
ם
མ་
α
d
0
5
10 Ft
allows the application of air-dried fuel, and yields very hot
gas, owing to the position of the blast pipes, p and q, which
provide air under the grate, and as the 4 feet layer of coal
FIG. 257.

Boiler
K
A.
S
10 P
Fig. 258.
e
ག
e
q
B
JoFt
10 Ft
מן

MACHINES FOR FORGING IRON.
813
perfectly transforms the carbonic acid into carbonic oxide,
and also decomposes the steam. The furnace can produce
temperatures of at least 3000° C. When putting off the
lowest blast, the ash may be easily removed by r; this
allows the application of fuel rich in ash. The pressure of
the blast amounts to 4 or 6 inches water, and about 200
cubic feet are consumed per minute, and heated in m to
about 300° C. The escaping flame serves to heat a boiler
placed at K. These gas producers are also well adapted
for the melting of cast-steel, the smelting of pig-iron in
reverberatory furnaces, &c.
Machines for Forging and Condensing Iron.
Hammers are the most effective machines for condensing
the iron and for pressing out the intermixed slag. The
heavy tilt hammers are at the same time a test of the quality
of the iron, as balls which are not thoroughly converted into
wrought-iron will crumble into pieces when hammered.
Several kinds of hammers used for the purpose we have
already described (pages 750 to 753), but besides these there
are several others in use; first comes the
Steam Hammer.*-These hammers are extensively
employed in forging and shingling iron, and have almost
superseded the tilt and other lever hammers, as they possess
one great advantage, that is the power of regulating the
force of the blow according to the requirement of the work.
The hammer may be stopped at any period of its stroke by
checking the exit of the exhaust steam.
Steam hammers were first introduced by Nasmyth, in
1842, and they consist essentially of an inverted vertical
cylinder, supported by a frame formed of two standards.
The piston rod passes through the lower cylinder cover, and
is directly connected with a heavy hammer block, which
moves vertically between two guides attached to the inner
surface of the iron standards. Single acting steam hammers
use the steam only for lifting the hammer block, whilst
Systeme der Dampfhämmer. Zeitschr. des Ver. deutcher Ingenieure,
Bd. 4, pp. 6, 40. Revue Universelle, iii., 112. Geschichte der Dampfhämmer.
Mittheil. des Hannov. Gew.-Ver., 1863, p. 236. Hämmer auf. der Londoner
Ausstellung. Polyt. Centr., 1863, Lief. ii.
814
IRON.
double acting hammers allow the steam to act on the upper
surface of the piston, thus increasing the speed of the
descent and the force of the blow.
The hammers invented by Nasmyth and Condie are those
most frequently employed in this country.
Condie's hammer is constructed with a fixed piston and a
movable cylinder, the cylinder being cast in one piece with
the hammer block. The piston is suspended by a rod con-
nected with a ball-and-socket joint to the top cross bar of
the framing.
Nasmyth's hammer has also been modified in the most
various ways by the inventor and by others, and to enumerate
these modifications would exceed our limits.*
Fig. 259.

Ъ
2/96
HARTM., Fortschr., ii., 291; v., 240, 216.
Polyt. Centr., 1855, Lief. 2;
1861, p. 860. B. u. h. Ztg., 1849, p. 161; 1852, p. 452; 1857, P. 153.
Bgwkfd., vi., 343; ix., 395. DINGL., Bd. 134, p. 199.
Oesterr. Ztschr., 1856,
No. 33.
APPARATUS FOR FORGING THE IRON.
815
A steam hammer constructed by Daelen,* which is fre-
quently used in Germany, is shown by Fig. 259. The steam
enters under the piston, c, which is connected with the
hammer block, b, by means of the piston rod, a.
The following hammers are rarely employed :-
a. The Stamp Hammer (stempel-hammer), which is con-
structed and lifted by cams like the pounders in stamping
mills. When lifted it is thrown against a spring of caout-
chouc, which repels it and augments the weight of its blow.
These hammers weigh from 40 to 60 cwts. When they are
not continually in use, they may be preferable to steam
hammers, which require a large store of steam. The
hammer is constructed by Schmerber,† and hammers of
similar construction have been tried by Froming, Water-
house, Winton, and Von Schwind.
b. Hydraulic Hammers.-The piston of the hammer is
raised by some liquid (oil, for instance), which is pressed
under it and allowed to run out (Guillemin,§ Minary, Fair-
bairn). T
c. Pneumatic Hammers.-Air is compressed either under
or above the piston, thus retarding or accelerating its descent
(Cowan's hammer).
d. Friction Hammers.-By means of a driving shaft a
hammer is raised between a friction rod and two movable.
pulleys, as in a rolling mill by the friction between the
pulleys; upon relaxing the pulleys the hammer descends
(hammers of Manhardt, Kitson, Eassie).
Brown's Patent Bloom Squeezert+ or Shingling
Machine is represented by Fig. 260. The heated ball of
puddled iron, k, thrown on the top is gradually pressed be-
tween the revolving rollers as it descends, and at last comes
out at the bottom, where it is thrown on to a movable
*
ARMENGAUD., Publ. Industr., xi., livr. 1-6.
† Spamer, Kalender f. d. Bergm., 1857, p. 74. Ann. d. Min., 4 ser., xvii., 87.
HARTM., Fortschr., iii., 193.
+
|| B. u. h. Ztg., 1863, p. 95. RITTINGER'S Erfahr., 1861.
§ B. u. h. Ztg., 1856, No. 50.
¶ Polyt. Centr., 1862, p. 647.
HARTM., Fortschr., vi., 239.
+ URE's Dictionary of Arts and Mines, vol. ii., p. 743.
816
IRON.
66
'Jacob's ladder," by which it is raised to the rolls. This
machine causes a considerable saving of time, will do the work
of twelve or fourteen furnaces, and may be kept constantly
FIG. 260.

(K)
A
going as a feeder to one or two pairs of rolls. There are two
distinct forms of this machine. In that shown in the figure
the bloom receives only two compressions; in the other,
which is much the more effective, it is squeezed four times.
before it leaves the rolls and falls upon the Jacob's ladder.
According to Percy, the machine is liable to frequent break-
age, and has, therefore, fallen into disuse.
FIG. 261.

Ъ
O
d
α
APPARATUS FOR FORGING IRON.
817
The Squeezers* most commonly used are those represented
by Figs. 261 and 262; they are either single or double. A
FIG. 262.

single lever squeezer of the simplest construction is shown
in Fig. 261; the bed plate, a, is cast in one piece; it is 6 feet
long, 15 inches wide, and 12 inches high. The whole is
screwed down upon a solid foundation of stone, brick, or
timber; b is the movable part, which makes from 80 to
90 movements per minute. The motion is imparted by the
crank, c, which in turn is driven by means of a strap and
pulley by the initial power; the diameter of the fly-wheel is
3 or 4 feet. The anvil, d, is about 2 feet long, and from 12 to
14 inches wide; it is a movable plate at least 3 inches thick,
and can be replaced when injured. The face of the working
part of the lever exactly fits the anvil, and consists of plates
attached by means of screws. It is desirable to have all
these face plates in small parts 8 or 10 inches in width, so
that breaking by expansion and contraction is secured against.
The whole machine, including the crank, is made of cast-
iron, and weighs 4 or 5 tons. According to Overmann, this
machine is both cheap and durable, and will squeeze 100 tons
of iron per week.
Fig. 262 represents the double squeezer employed at many
English iron works. The drawing is of a machine at the
Dowlais Iron Works figured in Mr. Truran's work; many
other forms are in use.
* URE'S Dictionary of Arts and Mines, vol. ii., p. 742.
VOL. II.
3 G
818
IRON.
The Rotary Squeezer* is shown in Fig. 263; a table, A, A,
with a ledge rising up from it to a height of about 2 feet, so
as to form an open box, is firmly embedded in masonry;
within this is a revolving box, c, similar in character, but
FIG. 263.

B
D
B
A
Ε
much smaller, and placed eccentrically in regard to it. The
ball or bloom, D, is placed between the innermost revolving
box, c, and the outer case, A, A, where the space between
them is greatest, and is carried round till it emerges at E
compressed and fit for the rolls.
Rolls or Cylinders.t-This plan of compression between
cylinders now effects in a few seconds the condensation and
distribution of the fibres which forty years ago was only ob-
tained after many heatings in the furnace and many blows
of the hammer. The cylinders are of two kinds :—1, those
which serve to draw out the ball, called puddling rolls or
roughing rolls, and which are in fact reducing cylinders; and
2, the cylinders of extension, called rollers, for drawing into
bars the massive iron after it has received a welding to make
it more malleable. This second cylinder is divided into
several varieties, according to the required patterns of bar-
iron. These may vary between 2 square inches and less than
1-6th of an inch.
Beneath the cylinders there is usually an oblong fosse into
which, when the iron is compressed, the scoria and the scales
fall. The sides of this fosse, constructed of stone, are
founded on a body of solid masonry capable of supporting
* URE's Dictionary of Arts and Mines, vol. ii., p. 743.
+ Ibid., p. 744.
ROLLS OR CYLINDERS.
819
the enormous load of the cylinders. Beams of wood help to
form the sides of this pit, and the frames may be secured to
them with screws and bolts; but massive bars of cast-iron
are found to answer better, not only because the uprights
and bearers may be solidly fixed to them, but because the
basement of heavy metal is more difficult to shatter or dis-
place, an accident which happens frequently to the wooden
beams. A stream of water is supplied by a pipe to each pair
of cylinders to prevent their getting hot, and to prevent the
hot iron from adhering to the cylinder, by cooling its surface,
and perhaps slightly oxidising it.
The shafts are I foot in diameter for the roughing rolls,
and 6 inches where they communicate motion to the cylinders
for drawing the iron into bars.
The roughing rolls are employed either to work out the
lump or ball immediately after it leaves the puddling furnace,
as in the Welsh forges, or only to draw out the piece after
it has been shaped under the hammer, as is practised in most
of the Staffordshire establishments. These roughing cylin-
ders are generally 7 feet long, including the trunnions, or
5 feet between the bearers, and 18 inches in diameter, and
weigh altogether from 4 to 4½ tons. They contain from five
to seven grooves, usually elliptical in form, each decreasing
gradually in size, as is shown in Fig. 264. The small axis.
of each ellipse, as formed by the union of the upper and under
grooves, is always placed vertically, and is equal to the great
axis or horizontal axis of the succeeding groove; so that in
transferring the bar from one groove to another it must re-
ceive a quarter of a revolution, whereby the iron gets elon-
gated in every direction. Sometimes the roughing rolls serve
as preparatory cylinders, in which case they bear rectangular
grooves towards one extremity, as the figure shows. Several
of these large grooves are bestudded with small asperities,
analogous to the teeth of files, for biting the lump of iron and
preventing its sliding. On a level with the under side of the
grooves of the lower cylinder there is a plate of cast-iron,
with notches in its edge adapted to the grooves. This
piece, called the apron, rests on iron rods, and serves to sup-
port the balls and bars exposed to the action of the rollers,
3 G 2
820
IRON.
and to receive the fragments of ill-welded metal which fall
off during the drawing. The housing frames in which the
rollers are supported and revolve are of great strength; they
are 5 feet high; the side perpendicular to the axis of the
cylinder is I foot thick, and the other 10 inches. The
upper ends of each pair of bearers are connected by two
iron rods, on which the workmen rest their tongs or pincers
for passing the lump or bar from one side of the cylinder to
the other.
The bushes are each composed of two pieces; one
of hard brass, with a cylindrical notch, is framed into
the other, which is made of cast-iron, as is seen in Fig.
264.
The iron bar delivered from the square grooves is cut by
the shears into short lengths, which are collected in a bundle
(pile) to be welded together. When this bundle of bars
has become hot enough in the furnace it is conveyed to the
rollers, which differ in their arrangement according as they are
meant to draw iron from a large or small piece. The first
(Fig. 264), possesses both elliptical and rectangular grooves;
the cylinders are I foot in diameter, and 3 feet long between
the bearers. The bar is not finished under these cylinders,
but is transferred to another pair with grooves of the dimen-
sions proper for the bar, with a round, triangular, rectangular,
or fillet form. The profile of a triangular groove used for
square iron is the shape of an isosceles triangle slightly
obtuse, so that the space left by the two grooves together
may be a rhomb, differing little from a square, and the
smaller diagonal is vertical. When the bar is to be passed
successively through several of these grooves the larger or
horizontal diagonal of each groove is made equal to the
smaller or upright of the preceding, so that the iron must be
turned one-fourth round at each successive draught, and
thus receive pressure in opposite directions. Indeed, the
bar is often turned in succession through the triangular and
rectangular grooves that its fibres may be more accurately
blended together. The decrease in the capacity of the
grooves follows the proportion of 15 to 11.
When the iron is to be reduced to a small rod the diameter
ROLLS.
821
of the cylinders is such that three may be set in the same
housing frame. The lower and middle cylinders are em-
ployed as roughing rollers, while the upper and middle ones.
are made to draw out the rod. When a rod or bar is to be
drawn with a channel or gutter in its face, the grooves of
the rollers are suitably formed.
2
The velocity of the cylinders varies with their dimensions.
In one work, cylinders for drawing out iron from to of
an inch thick, make 140 revolutions per minute, whilst those
for iron of from of an inch to 3 inches make only 65. In
another establishment the cylinders for two-inch iron make
95 revolutions per minute; those for iron from of an inch.
to 1 inch make 128, and those for bars from to of an
inch, 150. The roughing rollers move with only one-third the
velocity of the drawing cylinders.
Fig. 264 represents the shingling and plate-rolling mill.
The shingling mill for converting the blooms from the balling
or re-heating furnace into bars consists of two sets of grooved
cylinders, the first called puddling or roughing rolls; the
second are for reducing or drawing the iron into mill-bars,
and are called simply rolls.
a, a, a, a are the powerful uprights or standards, called
housing frames, of cast-iron, in which the gudgeons of the
rolls revolve; b, b, b, b are bolt rods for binding these frames
together at top and bottom; c are the roughing rolls, each
with a series of triangular grooves, so that between those
of the upper and under cylinder rectangular concavities are
formed in the circumference with slightly sloping sides. The
end groove to the side of c should be channelled like a rough
file in order to take better hold of the blooms-to bite the
metal as the workmen say, and give it the preparatory elon-
gation for entering into and passing through the remaining
grooves till it comes to the square grooves, where it becomes
a mill-bar.
d, d are the smooth cylinders hardened upon the surface,
or chilled by being cast into iron moulds, for rolling iron into
plates. e, e, c, e are strong screws with rectangular threads
which work by means of a wrench or key into the nuts,
c', c', c', c', fixed in the standards; they serve to regulate
822
IRON.
the height of the plumber blocks or bearers of the gud-
geons, and thereby the distance between the upper and under
cylinders. ƒ is a junction shaft; g, g, g are solid coupling
boxes which embrace the two separate ends of the shafts
FIG. 264.

m.
b
R
m
9
9
d
d
m
m
m
ROLLS.
823
and make them turn together. h, h are junction pinions
whereby motion is communicated from the driving shaft, f,
through the under pinion to the upper one, and thus to both
upper and under rolls at once. i, i are the pinion standards
in which their shafts run; they are smaller than the uprights
of the rolls. k, k are screws for fastening the head pieces, l,
to the top of the pinion standards. All the standards are
provided with sole plates, m, whereby they are screwed to
the foundation beams, 1, of wood, or preferably iron, as
shown by the dotted lines; o, o are the binding screw bolts.
Each pair of rolls at work is kept cool by a small stream of
water impinging upon it from a pipe and stop-cock.
In the cylinder drawing the workman who holds the ball
in tongs passes it into the first of the elliptical grooves, and
a second workman on the other side of the cylinders receives
this lump and hands it over the top of the rolls to the first,
who re-passes it between the rollers, after bringing them
somewhat closer to each other by giving a turn to the ad-
justing pressure screws. After the lump has passed five or
six times through the same groove it has an elliptical form,
and is called in England a bloom. It is next passed through
a second smaller groove which draws it out. In this state
it is treated by a second pair of cylinders, by which the iron
is drawn into flat bars 4 inches broad and an inch thick.
Fragments of the ball or bloom fall round about the cylin-
ders, which are afterwards added to the puddling charge.
In a minute and a half the rude lump is transformed into a
bar. A steam engine of 30-horse power will in a week rough
down 200 tons of coarse iron.
Different means are employed for the purpose of rolling
at greater speed, as for example, two or three pairs of rolls
are placed one in advance of another, or the blooms are
passed alternately through the grooves of two mills moving
in opposite directions; the bar is received on a carriage,
which is rapidly driven from one to the other by steam
power. Frequently a combination is made of three rolls
placed one above another in the same housing; this is called
the three-high train, which is driven from the middle; the
central roll runs forward with the lower and backward
824
IRON.
with the upper one, so that the bar is passed backwards and
forwards by entering alternately between the grooves of the
middle and upper and middle and lower rolls.
Very heavy rolls are frequently reversed at each passage
of the pile. For some purposes,* rolling armour plates, &c.,
this is almost a necessity, because of the great weight of the
masses to be dealt with, and the inconvenience which would
therefore attend the lifting of the slabs over the rolls at
each pass.
In other cases the reversing principle saves
time and allows more passes to be made at one heat, whilst
it also does away with a good deal of lifting. In the re-
versing plate mill at the London and North Western Steel
Works at Crewe, the reversal of the motion of the rolls is
effected by reversing the pair of engines by which the rolls
are driven, and the plan is found to answer well. The
engines in question have 28-inch cylinders with 4 feet stroke,
and make 3 revolutions to one revolution of the rolls; they
have balanced steam valves, and are reversed by hydraulic
power without shutting off the steam. The arrangement
works so freely that the engines have been reversed seventy-
three times in one minute. The rolls of this train are 6 feet
10 inches long by 24 inches in diameter, and the train will
bring down a slab from the full thickness of the pile to that
of a locomotive frame plate in eleven passes, whilst the
lowering of the top rolls at each pass, like the reversal of the
engines is effected by hydraulic power.
In the case of the plate mill at Messrs. W. H. and
G. Dawes's works at Elsecar, near Barnsley, as well as in
that of the armour plate mills of Sir John Brown and
Co., at Sheffield, and in some other instances, the plan of
reversing the motion of the rolls by means of a clutch has.
been adopted.
Compound or universal rolling mills consist of a combina-
tion of a vertical with an ordinary horizontal pair of plain
rolls, and are sometimes used for the production of bars of
plain rectangular sections of the most various dimensions,
by simple adjustments of the rolls instead of a special
Engineering, Jan. 4, 1867.
THE UNIVERSAL ROLLING MILL.
825
pair of grooves for each size, as is commonly the case.
Figs. 265 and 266 represent one of these universal rolling
FIG. 265.

C
1/18
mills. By properly adjusting the horizontal rolls, a, and
the vertical rolls, b, rectangular bars of any size may be
FIG. 266.

JO)
826
IRON.
produced. In the newer construction of this mill only one
of the vertical rolls is made movable. This compound mill,*
known as While's mill, has been advatageously adopted in
South Wales for blooming rail piles. The horizontal rolls
are driven in the usual way from below, the vertical pair
being connected with them by an intermediate shaft, carrying
a mitre-wheel gearing into a wheel upon one of the vertical
rolls. The machine is driven at a very low speed, making
only five revolutions per minute, so that the pile is subjected
to a powerful and long continued pressure compared with
the usual system of blooming in the first grooves of the
roughing pair in an ordinary train, making from So to
100 revolutions per minute.
Iron bars of an irregular section, rails, &c., are rolled with
suitably grooved rolls. Fig. 267 represents a pair of finishing
rolls as constructed for rolling Birkenshaw's rails.
FIG. 267.
The

open spaces along the middle of the figure, and which owe
their shape to the moulding and turning on the periphery of
the rollers, indicate the form assumed by the iron rail as
it is passed successively from the larger to the smaller
tures until in the last it is finished.
aper-
A peculiar kind of roll is used for slitting sheets of metal
into light rods either for the use of wire drawers or of nail
makers. The rolls are represented in Fig. 268. The cutters
are steel discs fixed on spindles a certain distance apart by
stops. The discs on one spindle interlock with those on the
other, thus forming what may be called revolving shears, for
*BAUERMAN, Treatise on Metallurgy of Iron, p. 306.
URE'S Dictionary of Arts and Mines, &c., vol. iii., 576.
LEVER SHEAR.
827
the principle is that of clipping; so that a sheet of metal on
being passed through this machinery is separated into square
FIG. 268.

or rectangular slips agreeing in size with the divisions of the
rollers.
The puddled bars for piling are usually cut up with a
lever shear, shown in Fig. 269. This shear has one fixed
FIG. 269.

1/40
and one vibrating jaw, the latter forming one arm of a straight
or bent lever, which is moved by a crank or eccentric.
Guillotine shears, with a diagonal edged knife which moves
vertically between parallel guides, are often used in plate
works when a cut of considerable length is required.
The rough or crop ends of rails or heavy bars are sawn off
while still hot with circular, saws 3 or 4 feet in diameter.
They are driven either by belts or by direct acting steam
turbines on the same shaft, and make from 900 to 1300 revolu-
tions per minute.
828
IRON.
II. THE PRODUCTION OF WROUGHT-IRON IN THE
PUDDLING PROCESS.
As the finery process is carried out by different methods
according to the quality of the pig-iron and that of the
wrought-iron to be produced, so the puddling process may
be classified, but only into two methods, namely, dry puddling,
which is carried out on hearths formed of sand whilst em-
ploying refined metal, and wet puddling (boiling process), for
which crude iron or a mixture of crude iron with refined metal
is used.
The refining of the pig-iron in hearths (page 715) and in
reverberatory furnaces (page 726) is more frequently used
for iron intended for the puddling process than for that
which is fined in open hearths, and the operation of refining
is mostly performed in the refinery fires.
The process of refining originated in England, and is fre-
quently employed in Wales, chiefly for the production of the
superior kinds of wrought-iron. The refinery fires are about
4
feet square, 15 or 18 inches deep, and are provided with
two or three tuyeres on each side. In some of these fires
the pig-iron is charged when cold, whilst in others it is run in
a liquid state direct from the blast furnaces; this method ad-
mits of a larger production at a decreased consumption of
fuel and time. The bottom of the furnace is formed of sand-
stone or of fire-bricks, upon which a layer of small pieces of
sandstone is placed.
When commencing an operation some ignited fuel is put in
the middle of the hearth with coke upon it, and the pig-iron is
charged and covered, when the blast machine is put in
motion. The blast has a pressure of from 1 to 2 lbs. ac-
cording to the density of the coke. A charge of 2 tons of
pig-iron begins to melt in about 1 hour, and is completely
fused in 2 or 2 hours. Fresh coke is then added and the
blast kept in action until the pig-iron is sufficiently de-
carbonised; the metal is now tapped off, together with
the formed slag, into cast-iron moulds 3 feet long and 6 or
8 inches thick, which are cooled with water from below. On
throwing water upon the metal it becomes chilled and the
THE PUDDLING PROCESS.
829
separation of the slag is facilitated. Sometimes the fine metal
is made to run from the refinery fires direct into the puddling
furnaces; in this case a previous tapping off of the slag is
required, as this slag contaminates the puddling process.
I
The lower part of the bright and light metal is white and
compact, and its surface cellular to a depth of from to
1 inches, according to whether the metal was produced from
grey or white pig-iron; this circumstance also regulates the
time required for the refinery process, which accordingly
takes 3 or 4 hours. When running the liquid pig-iron into
the refineries 5 cwts. of coke are consumed per 2 tons of
pig, and 22.3 cwts. of common forge iron, or 22'1 cwts. of
good grey pig-iron per ton of fine metal; about two-tenths
more of both coke and pig are consumed when charging
the refineries with cold pig-iron. The resulting slags, about
3 cwts., contain from 56 to 60 per cent of iron.
When refining hot blast pig-iron, about 36 lbs. more are
required than of cold blast pig for producing 1 ton of fine
metal. The treatment of pig-iron produced from blackband.
iron ore is difficult owing to its easy fusibility, which requires
that the iron should be refined at lower temperature for
longer time with the application of a greater quantity of
blast, thus increasing the loss of iron; 24 cwts. of this kind
of pig-iron are required for the production of 1 ton of fine
metal. An addition of slag (page 779) increases the loss of
iron, whilst an addition of limestone, 15 or 18 lbs. per ton,
(page 779) improves the quality of the fine metal. The
blast pipes are inclined about 38°, and they are 1 or
13 inches wide; the quantity of blast introduced is about
94,000 cubic feet 3 tons per ton of fine metal when the pig
is run in from the blast furnaces; but when melted in the
hearth, 136,000 cubic feet are required with white, and
153,000 cubic feet with grey pig-iron.
The weekly pro-
duction of a refinery is from 150 to 160 tons in the former,
and from 80 to 100 tons in the latter case. Hot blast acts
disadvantageously upon the decarbonisation and the loss of
iron, without saving fuel.
The consumption of fuel depends on the nature of the
coke and pig-iron, and on the construction of the refinery
830
IRON.
hearth.
When running the liquid pig-iron into the re-
fineries, the consumption amounts to 4 cwts. of coke per ton
of fine metal with forge pig, and to 5 cwts. with grey pig-
iron; and it amounts to from 6 to 8 cwts. when charging
the refineries with cold pig.
About 16-horse power is required to furnish the blast for
a weekly production of 100 tons of fine metal.
At Dowlais 1 or 2 tons of pig-iron are refined with coke
in 2 or 3 hours, with an addition of slag of the previous
operation; the hearths employed are 1 foot 4 inches deep,
4 feet long, and 3 feet broad; the blast is introduced by
four or six tuyeres, which are inclined at an angle of 20°.
At Aberdare, an addition of fluor-spar is sometimes used,
and 1 tons of pig-iron are refined in 3 hours at a loss of
15 per cent of iron.
In Yorkshire the tuyeres usually alternate with each other
on opposite sides, two being placed on one side and three
on the other.
In French and Belgian iron works, 0.303 or 0.313 kilo.
of coke are consumed for refining 1 kilo. of pig-iron. A
hearth with six tuyeres produces 130 tons, and one with
four tuyeres only 90 tons of fine metal per week.
Dry Puddling in Furnaces with a Sole of Sand or of
Iron Plates.-This method was invented in 1787 by Henry
Cort; fine metal in admixture with white and mottled pig-
iron is usually treated by this method. The kinds of white
and mottled pig employed are not purer than the fine metal
which is produced with coke in the refineries and has be-
come poorer in silicon, manganese, and phosphorus, but
richer in sulphur, owing to the sulphurous coke.
The puddling furnaces employed have a hearth of sand
resting upon a layer of slag or other materials, and covered
very thinly with some slag; the side walls of the furnace
are made of fire-bricks, and the fire-bridge is low.
At this method of puddling charges of from 400 to 480 lbs.
of pig-iron are converted into a pasty state a strong fire
being given; the softened masses are then divided with a
B. u. h. Ztg., 1862, p. 414.
THE PUDDLING PROCESS.
831
rabble and stirred at a lower temperature; the pasty mix-
ture of iron and a small amount of slag is stirred and broken
up several times when it has come sufficiently to nature, that
the iron may be formed into balls, and before they are
brought under the hammer or squeezer they are strongly
heated for a few minutes with the doors of the puddling fur-
nace closed. One charge is finished in 1 or 1 hour, at a
consumption of 70 or 80 per cent of coal (from 0.165 to
0°2 ton per 100 lbs. of mill-bars) and at a loss of 6 or 8, and
sometimes even 10 per cent of iron.
No ebullition like that in the boiling process takes place
at this puddling method; the charges are put into the fur-
nace through a small door on the side of the chimney about
15 minutes previous to the removal of the balls, thus
shortening the operations.
During the melting down of the pig-iron a coating of
magnetic oxide is formed on the metal which, together with
the admitted air, forms the decarbonising agent; the de-
carbonisation is facilitated when all the carbon in the metal
is present in a chemically combined state. Phosphorus and
sulphur are but imperfectly extracted as the iron comes to
nature so quickly, and owing to its pasty state is in less
intimate contact with the oxide of iron. Besides, according
to Berthier, oxide of iron and sulphide of iron do not decom-
pose each other, but combine in any proportion.
This method of puddling admits of a large production at
a small loss of iron and a small consumption of fuel, but it
always yields red-short wrought-iron if a very pure fine
metal is not employed, chiefly metal produced from pure
ores with cold blast.
This dry puddling process is at present seldom used, and is
almost superseded by a similar process with an addition of a
little ferruginous slag or hammer slag; this method is carried
on in furnaces with a sole of iron plates, and produces fibrous
wrought-iron. The sand bottoms increase the loss of iron.
and the consumption of fuel, and the resulting wrought-
iron is richer in silicon.
Truran states that at the dry puddling process in England
I ton of puddle iron is produced from an average of 21 cwts.
832
IRON.
I qr. 20 lbs. of fine metal, and from 21 cwts. 3 qrs. at the boiling
process; the consumption of coal in the latter amounts from
14 to 22 cwts. per ton of mill-bars. The weekly production
of the dry puddling with fine metal amounts to 23 tons, and
that of the boiling process to from 18 to 21 tons.
The Boiling or Wet Puddling Process is the method
generally employed, and admits of the treatment of the most
different kinds of pig-iron, both of pig-iron rich in graphite
and of the impure sorts which are sometimes previously
refined (Low Moor, Alvenslebenhütte). The iron is melted
down at a high temperature, thus converting the graphite
into chemically combined carbon as in the refining process;
and as the perfectly liquid iron is kept in contact by stirring
sufficiently long with a large quantity of liquid slag, capable
of giving up oxygen, the iron is more thoroughly purified
than by the previous method; the purification may be facili-
tated by using Schafhäutl's powder.
When employing slags more or less rich in oxygen, a
higher or lower temperature, an oxidising or a reducing
flame, and a suitable construction of the furnace, either
fibrous iron or fine-grained iron is produced; in this way a
superior kind of wrought-iron will result, though, on the
other hand, the waste of fuel, time, cost, labour, and metal
is greater than in the dry puddling process.
As the boiling process necessitates a more powerful
stirring of the fused mass, and the resulting cinder has a
corroding action, the side walls of the hearths in the
puddling furnaces employed in this process are formed of
hollow castings of iron instead of brick walls; the hollow
side walls are cooled by means of circulating water or air.
When less acid slag is formed and when it is quickly
removed from the hearth (for instance, at the production of
fibrous iron from good partially refined pig-iron with an
addition of not too much rich cinder) the side walls of the
furnace hearth are walled-up with fire-bricks, thus saving
fuel.
The chemical and physical reactions which take place in
the puddling process have been explained chiefly in the
THE PUDDLING PROCESS.
833
investigations of Calvert and Johnson, Becker, Schaf-
häutl, Lan, List,§ Zobel, and Gurlt,** and lately by
Drassdo.tt They are founded on the composition of the iron
and cinder at the different periods of the puddling process.
The following analyses by Calvert and Johnson show the
gradual modifications which takes place during its trans-
formation into wrought-iron :-
Time at which
the Sample
was taken.
Raw material
40 minutes after
commencing
бо
65
80
95
105
IIO
Blooms
Fe.
C.
Si.
S.
P.
Appearance and
Properties of the Observations.
Sample.
94'052 2*275 2·720 0·311 0·645 Grey pig-iron.
2.726 0·915
2.925 0.197
2'444 0'194
2*303 0*182
1.647 0.185
0.693 0.163
0'772 0.168
99.338 0.269 0120 0134 0*139
Finished bar iron 99.490 o'III 0·088 o'094 0'117
White.
White and
somewhat
ductile.
Consisting of
single grains.
Similar as the
last sample.
Malleable
grains.
Puddle iron.
The same.
Very brittle.
Maximum of
the amount of
carbon.
Commence-
ment of the
boiling.
Finishing of
the boiling.
End of the
operation.
Different theories have been suggested by Grundmann,‡‡
Cailletet, Minary, and Resal.§§
The Puddling Process Aiming at the Production of
Fibrous Iron.
All different kinds of pig-iron are submitted to this pro-
cess, sometimes also fine metal in admixture with pig-iron;
and the pig-iron is more or less quickly converted into
wrought-iron, according to the impurities it contains; for
instance, by employing more or less rich cinder as flux, by
using a higher or lower temperature, &c.
VOL. II.
* DINGL., Bd. 146, p. 121; Bd. 153, p. 156.
↑ B. u. h. Ztg., 1858, p. 5.
‡
HARTM., Fortschr., iv., 186.
B. u. h. Ztg., 1860, pp. 258, 435.
§ Ibid., 1860, pp. 52, 472; 1862, p. 191.
¶DINGL., Bd. 154, p. III.
**
Berggeist, 1860, p. 523.
+ Preuss. Ztschr., xi., 170.
+
B. u. h. Ztg., 1855, p. 337.
DINGL., Bd. 154, p. 111.
§§ Ibid., Bd. 163, p. 352.
3 H
834
IRON.
3
The puddling process comprises the following operations:
1. The Formation of the Hearth.-After thoroughly
drying a new furnace by means of a strong fire, and coating
the iron side walls, and sometimes also the interior furnace
walling, with a refractory mixture of I part of clay and 2 of
sand of an inch thick, rich cinders of the puddling or re-
heating furnaces in a roasted state (called bull-dog, page.
345), in pieces the size of the fist, are placed round the edge
of the hearth in such quantity as to cover the bottom plate
about 3 inches thick when in a fused state; at the same time
rich slag deposits from the hearths of puddling furnaces and
slags which have stuck to the tools are charged on the iron.
sole plate, and fused together with the other cinders. The
slag hole is now bricked up, some slag heaped up before the
working door, and the whole hearth is covered 1 or 2 inches
high with slags from the tools; the working door is closed
and luted, and a gradually increasing fire is applied till all
the cinder is fused, which will be in six or eight hours. The
mass is then broken up with a pointed iron bar, and the
melting repeated till the slag has attained a uniform consis-
tence. Next, layers of slag are repeatedly melted down till
the slag hearth is about 4 inches thick; a hearth will last
eight weeks or longer.
The very refractory rich slags only attain a pasty consis-
tence; they protect the iron bottom plate from being wasted,
and the fused iron from chilling, and they form a compact but
plastic support, allowing a ready removal of the half-fined iron.
When commencing an operation in an old furnace, it must
be heated for six or eight hours previous to the charging,
until the walls have attained a uniform red heat, and the
slags of the hearth have become thickly liquid.
2. Charging the Puddling Furnace.-Some rich cinder
from the previous charge is generally left in the furnace, but
it is not usually in sufficient quantity, as part of the poorer slag
has run out of the furnace; therefore it is necessary to add
some fluxes more or less rich in oxide of iron, in proportion
to the ease with which the pig-iron to be treated comes to
If the hearth should contain too much cinder it
must be either ladled out or tapped off; the hearth should
nature.
THE PUDDLING PROCESS.
835
be each time freed from deposits. When puddling a thinly
liquid iron which with difficulty comes to nature, rich slags,
hammer slag, &c., are thrown chiefly on both bridges, and
the back wall, after having finished one operation; and a
larger quantity of these fluxes is added, if the temperature of
the furnace is low and the hearth contains poor cinder.
The mass is then well mixed together by means of a
rabble, and somewhat cooled by throwing water upon it,
thus allowing the hearth to be made even and preserved;
this is also the best time for thoroughly cleaning the fire-
grate. The pigs, each being I or 1 inches thick, and from
10 to 25 lbs. in weight, are now uniformly distributed on the
hearth by means of a flat iron bar, but care is taken to place
the thicker pieces near the fire-bridge in order to facilitate
their melting; smaller pieces are sometimes charged later
on.
A charge for single furnaces is usually 3 or 4 cwts.,
and that for double furnaces from 6 to 10 cwts., but
usually 7 or 8 cwts. When employing stoves for previously
warming the iron, half an hour less is required for the
melting down.
I
In Hoerde (page 770) some advantages are obtained by
melting the iron in cupola furnaces, and running it into
the puddling furnaces in a liquid state; and also where the
locality allows it, as in a few cases in England, the liquid iron
is run into the puddling furnaces direct from the blast furnaces.
3. Melting Down the Pig-Iron.-The iron is melted
down in from 25 to 45 minutes with the working door closed
and luted, and the damper opened; during this time, the fire
is well stirred three or four times. To prevent the admis-
sion of cold air the lower joints of the working door are
closed with small cinders, and a large piece of coal is placed
inside before the door. The half-fused masses on the furnace
walls are then shifted towards the middle of the hearth, in-
troducing the tool for performing this manipulation through
the opening in the working door. A strong melting heat is
again given for four or five minutes. A pointed iron bar is
now systematically moved to and fro on the bottom of the
hearth, to remove parts which have deposited, and the masses
are then again shifted towards the middle of the hearth.
3 H 2
836
IRON.
When treating thinly liquid iron on a sufficiently hot
hearth the pieces of iron are not moved or shifted during the
melting down; but if the hearth is too cool, the red-hot.
pieces on the back wall and the bridges are raised up so that
the flame may play on the hearth and raise its temperature
sufficiently high. The raised iron partly melts in that posi-
tion, and after the hearth has become hot enough, the pieces
of iron are shifted into the hearth, broken into small pieces
which are covered by the liquid cinder. The pieces of iron.
are placed in such a position that one piece partly covers
the other, so as to allow the flame to be conducted all round
them.
At the treatment of white pig-iron, which is less apt to
corrode the hearth and attains a pasty state on fusion, a
quick melting at a high temperature is required to lessen the
loss of iron. This kind of iron is therefore not charged im-
mediately after the finishing of the previous charge, the
cleaning of the grate, and the preparation of the hearth, but
the hearth is strongly fired for a quarter or half an hour
without any charge. This does not waste any fuel or time,
as the process is afterwards carried on the more quickly. If
the working of white iron increases the thickness of the
hearth too much, some charges of grey iron are puddled, thus
reducing the hearth.
The following chemical reactions take place at the
melting down of the pig-iron-As soon as the pig-iron is
red-hot it becomes coated with magnetic oxide, which is
dissolved off when the fused iron is immersed in the
fused cinder. The pig-iron thus loses its impurities, which
become oxidised by the oxygen of the peroxide of iron
contained in the magnetic oxide, whilst the peroxide is trans-
formed into protoxide and even to metallic iron. Silicon
oxidises first, forming silica, and combining with the prot-
oxide of the magnetic oxide; manganese and phosphorus are
next oxidised, and lastly sulphur. The originally rich
cinder (page 737), 8(3 FeO,SiO3)+ FeO, Fe₂O,, now becomes
more acid until the foreign admixtures are removed,-
9(3FeO,SiO3) + 3(3FeO₂SiO3) + FeOFe¸0¸.
THE PUDDLING PROCESS.
837
After the most vehement reaction ceases the cinder again
becomes more basic, but still remains more acid,—
10(3 FeO,SiO3) + 3(3FeO₂SiO3) + FeOFе₂03,
than the original cinder. The carbon contained in the pig-
iron does not oxidise at this period (as in the refining of pig-
iron), but its proportion may even be increased owing to the
oxidation of iron, silicon, &c. On the other hand, the graphite
is converted into chemically combined carbon, being aided
by the high temperature and the separation of silicon (page
716). When rapidly melting down at a very high tempera-
ture pig-iron rich in graphite, the graphite sometimes finds
no opportunity to dissolve perfectly, and is then, Osann*
states, separated on the surface of the puddling cinder.
Mr. C. W. Siemens instituted the following experiments
respecting the behaviour of silicon and carbon, in fluid cast-
iron, when contact with the atmosphere or the flame of the
furnace is strictly prevented, and he thus describes them:
10 cwts. of Acadian pig metal and I cwt. of broken glass
were charged upon the bed of a regenerative gas furnace
(usually employed for melting steel upon the open hearth).
"The bed of this furnace was formed of pure siliceous
sand, and one object in view was to ascertain whether any
reaction takes place between silica and fluid cast metal, it
being generally supposed that metallic silicon is produced
under such circumstances by the reducing action of the
carbon in the metal upon the silica or silicates present; the
cast metal employed in this experiment was Acadian pig,
containing-
Silicon
Carbon.
15 per cent.
4.0
"In the course of an hour the metal and glass were com-
pletely melted; a sample was taken out containing—
Silicon . 108 per cent.
Carbon.
2.90
*
(0.6 per cent combined carbon.
•12.3
B. u. h. Ztg., 1861, p. 328.
graphite.
838
IRON.
"At the end of the second hour another sample was taken
out and tested, the result being-
Silicon
Carbon
.
o'96 per cent.
2.40
combined.
"The physical condition of the metal had now undergone a
decided change; the carbon having wholly combined with the
iron rendered it extremely hard.
"The amount of silicon having steadily diminished, these
results prove that no silicon is taken up by fluid cast metal in
contact with silica or silicates.
The reduction of the amount of silicon in the metal might
be accounted for by the presence of minute quantities of
oxides of iron produced in melting the pig metal, which
oxides were now increased by the addition of hæmatite ore
in small doses.
"At the end of the third hour another sample was taken
containing-
Silicon
Carbon
0'76 per cent.
2.40
combined.
,,
the metal being extremely hard as before.
Additional doses of red ore were added gradually without
agitating the bath, and the effect upon the fluid metal was
observed from time to time.
"At the end of the fifth hour the samples taken from the
fluid bath assumed a decidedly mild temper; when the addi-
tion of ore was stopped, exactly six hours after the charging,
the metal was tapped and run into ingots; it then contained
o'046 per cent.
0*250 ""
Silicon
Carbon.
Thus both the silicon and the carbon had been almost
entirely removed from the pig metal by mere contact with
metallic oxide under a protecting glass covering.
"The quantity of red ore added to the bath amounted to
2 cwts., and the weight of the metal tapped to 10 cwts. 5 lbs.,
being slightly in excess of the weight of pig metal charged.
"But the pig metal had contained-
Silicon .
Carbon.
1'5 per cent.
4'0
">
Total.
5'5
وو
THE PUDDLING PROCESS.
839
whereas the fined metal contained collectively only o°296 of
silicon and carbon, showing a gain of metal of
5'5-0*296=5'204 per cent;
or, including the 5 lbs. of increased weight, a total gain of
5'7 per cent of metallic iron.
Supported by these observations, I venture to assert that
the removal of the silicon and carbon from the pig-iron in
the ordinary puddling or "boiling" process is due entirely
to the action of the fluid oxide of iron present, and that an
equivalent amount of metallic iron is reduced and added to
the bath, which gain, however, is generally and unnecessarily
lost again in the subsequent stages of the process. The
relative quantity of metal thus produced from the fluid cinder
admits of being accurately determined.
"The cinder may be taken to consist of Fe,O, (this being
the fusible combination of peroxide and protoxide), together
with more or less tribasic silicate (3FeO,SiO,) which may
be regarded as a neutral admixture not affecting the argu-
ment, and silica is represented by SiO3, from which it follows
that for every 4 atoms of silicon leaving the metal 9 atoms
of metallic iron are set free, and taking the atomic weight
of iron 28, and that of silicon = 225, it follows that for
every
4 × 22°5
90'0
grains of silicon abstracted from the metal
9 × 28 = 252
grains of metallic iron are liberated from the cinder.
Carbonic oxide again being represented by CO, and the
cinder by Fe,O,, it follows that for every 4 atoms of carbon
removed from the metal 3 atoms of iron are liberated; and
taking into account the atomic weight of carbon =
of iron 28, it follows that for every
grains of carbon oxidised.
6 × 4
24
28 x 3
84
grains of metallic iron are added to the bath.
==
6, and
"Assuming ordinary forge pig, after being re-melted in the
puddling furnace, to contain about 3 per cent of carbon and
840
IRON.
2 per cent of silicon, it follows from the foregoing that in
removing this silicon
252
90
X 2 =
5'6
per cent, and in removing the carbon
84 × 3 10'5
24
per cent of metallic iron is added to the bath, making a total
increase of
5.6 + 10°5 - 5 II'I
per cent; or a charge of 420 lbs. of forge pig metal ought to
yield 466 lbs. of wrought metal, whereas from an ordinary
puddling furnace the actual yield would generally amount to
only 370 lbs. (or 12 per cent less than the charge), showing
a difference of 96 lbs. between the theoretical and actual
yield in each charge.
"This difference, amounting to fully 20 per cent, is due to
the enormous waste by oxidation to which the iron is exposed
after it it has been brought to nature, when it is in the
form of a granular or spongy metallic mass, and during the
process of forming it into balls.
So great a waste of metal
by oxidation seems at first sight almost incredible, but con-
sidering the extent of surface exposed in the finely divided
puddled mass, it is not at all exceptional, and is, in fact,
almost unavoidable in a furnace of the ordinary construction,
maintained like a puddling furnace at a welding heat.
Many attempts have been made (for example, by Chenot,
Clay, Renton, and others), to produce iron directly from purer
ores, by reducing the ores in the first instance to a metallic
sponge, and balling up this sponge, which is a loose porous
mass, somewhat similar to spongy puddled iron, on the bed
of a furnace, but all these attempts have failed, simply on
account of the great waste of iron, a waste amounting to
from 25 to 50 per cent in balling up the sponge. Indeed,
the loss in an ordinary puddling furnace would probably
be greater than 20 per cent if the metal were not partly pro-
tected from the flame by the bath of cinder in which it lies;
for in one instance in which the cinder accidently run out of
a puddling furnace during the balling up of the charge, leaving
the iron exposed to the flame, I found the yield reduced from
THE PUDDLING PROCESS.
841
the average of 413 lbs. down to 370 lbs., showing an in-
creased waste of 43 lbs., or over 10 per cent, due to the more
complete exposure of the metal to the oxidising action of
the flame.
"In order to realise the theoretical result, a sufficient
amount of oxides must have been supplied to effect the oxi-
dation of the silicon and carbon of the pig-iron, and to form
a tribasic silicate of iron (3FeO,SiO3) with the silicic acid
produced.
The amount of oxide may be readily ascertained.
In taking the expression Fe,O,, the atomic weight of
which is-
3 × 28 + 4 × 8
=
116,
while that of the three atoms of iron alone is—
3 × 28
84,
it follows that—
116
84
× 46
63.5 lbs.
of cinder or oxide of iron are requisite to produce the 46 lbs.
of reduced iron, which were added to the bath. There
must, however, remain a sufficient quantity of fluid cinder
in the bath to form with the silicon (extracted from the iron)
a tribasic silicate of iron, or about 60 lbs., making in all
124 lbs. of fettling which would have to be added for each
charge, a quantity which is generally exceeded in practice
notwithstanding the inferior results universally obtained."
Mr. Siemens adds the following concerning the removal of
sulphur and phosphorus :-
The sulphur and phosphorus being generally contained
in English forge pig in the proportion of from o'2 to
0.6 per cent each, can hardly affect the foregoing quantita-
tive results, although they are of great importance as
affecting the quality of the metal produced.
It has been suggested by Percy that the separation of
these ingredients may be due to liquation.
The following was the result of an analysis by Mr. A.
Willis of an inferior English pig-iron before and after being
puddled :-
842
IRON.
Pig Metal.
Puddled Bar.
Sulphur.
0*08
Sulphur
•
0'017
Phosphorus.
1.16
Phosphorus.
0°237
Silicon
I'97
Silicon
0*200
showing the extent to which foreign matters are actually
removed by the process of puddling.
4. Boiling and Stirring, or Puddling the Fused Mass.
After having melted down the pig-iron, it is in the state of
fine metal and covered by the cinder bath, and both sub-
stances, iron and cinder, do not sensibly react upon each
other. This reaction, which is principally decarbonising,
must be induced by converting the fused masses into a pasty
state; this is effected by cooling them; a more intimate
contact of the substances thus takes place, and the affinity
of the carbon with iron is lessened. The cooling may be
effected by checking the draught and even by throwing
water upon the fused mass whilst stirring it, or there
are incorporated into the mass refractory basic slags,
which have at once a cooling effect and also facilitate the
subsequent decarbonisation.
During this stiffening of the
mass iron is now and then brought to the surface in contact
with atmospheric air, when magnetic oxide is formed, which,
by its transformation into protoxide and metallic iron,
liberates oxygen; this combines with the impurities, silicon,
manganese, phosphorus, and sulphur still contained in the
iron, at the same time decarbonisation commences, and its
beginning is indicated by blue flames of carbonic oxide that
appear in the furrows of the metal bath, and which are caused
by moving the rabble through it. As, however, the mixture
cannot yet be intimate, and also owing to the lowered tem-
perature, no energetic reactions take place at this period. On
account of the decomposition of the magnetic oxide the cinder
becomes somewhat more basic, forming a compound of—
3FeOSiO3 + 3(3FeO₂SiO3) + FeO‚Fe₂O3.
When stirring with the rabble, the puddler searches or
sweeps every portion of the bed by moving the point of the
tool in curved lines from the centre outwards towards the
bridges, commencing at the front. The sides are reached
by a kind of scooping action, the rabbles being worked
against the door-frame as a fulcrum. The tool must be
THE PUDDLING PROCESS.
843
changed every 5 or 8 minutes, as it would soften and adhere
to the iron in the furnace. When taken out it is cooled by
being plunged into a cistern to detach the adhering cinder;
the point is afterwards dressed into shape by forging with a
light hammer. The two workmen attending the puddling
furnace are changed with every tool.
In order to lessen the great amount of labour involved
in stirring the charges,* various mechanical appliances have
been proposed as substitutes for manual puddling. Tooth,
Menelaus, Bessemer, &c., have suggested the use of rotating
or oscillating hearths. The molten iron is fined by exposure
to oxidising influences in a cylinder lined with clay or other
refractory material, occupying the position of the hearth in
an ordinary puddling furnace, which receives a slow move-
ment of rotation about its long axis. The charge is also
traversed from end to end of the cylinder by inclining the
lining from the fire-place to the flue, and in the reverse
direction at opposite points of the circumference. The ball
is withdrawn from the furnace by removing the puddling
chamber and tilting it up at one end. These furnaces have
not as yet been successful, owing to the difficulty of
getting linings to stand the scouring action of the metal.
Menelaus found the best material to be titaniferous iron ore,
which was used in solid blocks. The ordinary fettling
materials, such as bull dog, were quite useless. Bessemer's
furnace has an egg-shaped puddling chamber, mounted at
the top of a rocking frame. The flame of the fuel is intro-
duced through a trunnion and passes out through the
opposite one.
Other means are proposed of imitating the motions of
hand-stirring, by moving the tool through a curved path by
a combination of reciprocating rotatory mechanism. One of
the best mechanical stirrers is that invented by Eastwood.
At the commencement of this period the iron may be
clearly distinguished from the slag with the eye, and it may
be felt with the rabble, but these indications disappear after
using the second tool. When using the third and fourth
* BAUERMAN'S Metallurgy of Iron, p. 284.
8++
IRON.
rabbles the reaction is violent, much carbonic oxide gas is
evolved, and the fused mass rises in blisters, as the particles
of iron which come into contact with the air oxidise, forming
magnetic oxide, which again oxidises the carbon and becomes
transformed into protoxide and metallic iron. As the oxida-
tion of the carbon takes place in a more rapid proportion
than the replacing of the peroxide of iron used for the oxida-
tion, the quantity of oxide and cinder decreases in this
period, whilst the amount of iron is increased, for the reason
that most of the peroxide of iron is transformed into metallic
iron. The resulting cinder is more acid, containing-
22(3FeO,SiO3) + 4(3FeO₂SiO3) + FeOFe₂03.
During the ebullition, part of the slag flows across the flue-
bridge and through the working door. As long as the cinder
contains magnetic oxide, peroxide of iron only is deoxidised,
and afterwards protoxide. After the most violent reaction
ceases the decarbonisation proceeds more slowly, the
amount of peroxide of iron in the cinder is again increased,
the cinder becomes more basic, corresponding to the
formula-
10(3FeO,SiO3) + 3(3FeO₂SiO3) + FeO,Fe2O3.
The particles of iron which first appear on the surface of
the cinder are light and glowing, and in the form of points;
they then adhere together, forming single lumps of larger
size and of crisped surface, first reddish, and afterwards
glowing white. After having used the fourth rabble the mass
settles, the cinder sinks to the bottom, and the glowing
white particles of iron project from it in a sponge-like
agglomeration. The puddler then finds it difficult to stir
the mass any more, and he is warned of the end by
the degree of resistance which he finds in stirring the mass;
three or four rabbles are usually sufficient. When aiming
at the production of a very soft fibrous iron, more rabbles
are used, and fewer when old wrought-iron is charged in
addition.
The iron is now in the state of steel, containing from 12
to 2 per cent of carbon. This period takes or an hour.
5. Consolidation of the Iron into Masses or Balls fit
for Hammering, and further Decarbonisation.-The iron
THE PUDDLING PROCESS.
845
projecting from the cinder, chiefly when it is in larger lumps
or balls, is oxidised, owing to the free admission of air
(when using gaseous fuel an excess of blast is now intro-
duced), the cinder becomes richer in protoxide of iron, and
the decarbonisation of the iron proceeds until fibrous
wrought-iron is formed, containing less than o˚5 per cent of
carbon. The cinder increases again in quantity, becomes
more basic, and approaches its original composition, of the
formula 8(3 FeO,SiO3) + FeO,Fe₂03.
During the whole process of puddling the atmospheric air
appears to act less directly upon the oxidation of the carbon,
silicon, &c., than indirectly, as its oxygen first oxidises the
iron and is thence carried to the other substances.
This part of the process is performed by the first puddler,
and comprises the following manipulations :-
a. In order to protect the balls from intermixed ash, the
fire is stirred very strongly at the commencement of this
period, so that it requires no further stirring, or the ash
would fly off along with the flame and contaminate the balls
which are in formation. The deposits on the hearth are now
broken off with a rabble, and the iron is moved to and fro
whilst air is admitted. When the mass has become very
pasty, the particles of iron lying near the flue are shifted to
the hottest place, thus separating the cinder as perfectly as
possible and facilitating the welding of the iron.
b. Still the mass of iron is not uniformly decarbonised,
and in order to expose it more to the atmospheric air, at
suitable portion is separated by means of the rabble, the
lowest part is raised, and in this position the separated
iron is placed before the fire-bridge. The whole mass of
iron is thus separated into pieces and transferred towards
the fire-bridge, together with the small isolated pieces
of iron deposited on the flue-bridge. Next the whole
of the iron is transferred towards the flue-bridge in
the same manner, and this manipulation is once or twice
repeated, when the mass will have become compact, welding
strongly together, and white and glowing in appearance. This
operation keeps the hearth free from deposits, exposes the
iron to the highest temperature, and it takes 10 or 15 minutes.
846
IRON.
c. For the formation of the balls, which requires great
skill and strength, the mass of iron is divided into pieces of
a certain size, usually weighing from 60 to 80 lbs. Each
piece is moved from the fire-bridge towards the flue-bridge, and
vice versa, whilst it is turned over; the single pieces or balls
must not weld together. When a large ball for heavy
articles is required it must be formed in good time, so as to
allow the ball to become compact. The balls must be
smaller the softer the iron is desired, for the manufacture of
plates, &c.
We have before observed that no stirring of the fire is
allowed during this period for fear of contaminating the balls
with ash; the required high temperature is now produced
by the position of the damper and by moving a flat iron bar
across the grate, thus lifting the slag of the ash and admitting
air. When using gaseous fuel, more gases are introduced,
together with a corresponding quantity of hot air.
d. Pressing and turning the balls. By means of a strong
hooked rabble one ball after another is moved before the
working door and pressed so as to flatten it; they are
then placed on their small edges and pressed together as
much as possible whilst supporting the hook against the
upper part of the working door, which is fixed by wedges;
the ball is then pressed in the same manner, but in a differ-
ent position, until it assumes a roundish form. It is then
moved towards the flue, the lowest side turned upwards.
At the formation of the balls the object aimed at is putting
the upper part of the iron mass which is most decarbonised
into the middle part of the ball, thus rendering the iron
more uniformly fibrous. In some iron works in Styria the
balls are sometimes formed by forcibly pushing them against
the side walls of the furnace. After forming all the balls
and placing them before the flue-bridge, they are turned
again and distributed against the back wall, so that the
least decarbonised balls, which have not the characteristic
wet appearance, are placed near the flue to expose them to
a higher temperature and to a stronger draught of atmo-
spheric air. The working door is then closed for some
minutes and the temperature of the furnace is raised to a
THE PUDDLING PROCESS.
847
high heat.
The formation of the balls is finished in about.
a quarter of an hour.
e. The balls are withdrawn from the furnace thus; one
workman opens the working door, a second one seizes the
ball with tongs, lifts it up to the working plate and thence
to an iron car, which a third workman carries to the
blooming apparatus. During the shingling of this ball the
other balls are continually turned, whilst the damper is kept
half closed; the balls are thus prevented from being decar-
bonised on one side more than on the other, isolated small
particles of iron are welded to them, and the balls are pre-
vented from getting a coating of cinder which, upon
shingling, would impede the reaction of the atmospheric
air.
6. Shingling the Balls.-In order to condense the iron and
to press out the contained cinder the balls are treated by
suitable machinery (page 813) of which the hammer is
best adapted; the ball is first placed upright on its
longest axis and afterwards uniformly pressed, so as to attain
a prismatic or a hexagonal form. Well puddled balls are
light in colour; they contain a tolerable quantity of cinder
and offer little resistance to the blows of the hammer. The
hammered balls or blooms when cold show a smooth uniform
surface. Raw and impure balls break under the hammer,
and if not too raw show, when cooled, a cracked slaty ap-
pearance, but good blooms may have the same appearance
if hammered too long. Blooms are seldom free from cracks,
which may be removed by welding. The blooms are usually
passed on to the rolling mills.
I
The last manipulation is finished in or an hour,
according to the size of the charge; the whole operation,
therefore, takes 1 or 2 hours, and consequently six or eight
charges can be worked in 12 hours. The consumption of fuel
in the puddling process has been already stated on page 771;
the loss of iron is from 6 to 16 per cent, and the weekly
production in a single furnace is above 400 cwts., and that of
a double furnace more than 700 cwts. In the south of France
from 23.6 to 34°49 cwts. of iron are produced in 12 hours,
the pig-iron being previously heated.
848
IRON.
The normal puddling process is liable to various irregu-
larities, thus:-
a. When treating a very thinly liquid pig-iron, it agglomer-
ates with difficulty and appears dark and dull, owing to its
slow oxidation. The process may then be facilitated by an
addition of rich cinder or by cooling the fused mass before
stirring it, but not without a great loss of iron; it is there-
fore advisable to employ a better mixture of pig-iron at the
next charge.
b. When the resulting cinder is too liquid, this may be
remedied by stiffening it by throwing water upon it, but this
is apt to damage the hearth and bottom plate, and it is
better to give an addition of hammer slag moistened with
water.
c. When the dimensions of the furnace, chiefly those of
flue and chimney, are enlarged, and when the arched roof has
been burnt out; in the latter case the draught in the furnace
passes along the roof and comes little in contact with the
iron, which therefore comes less easily to nature.
Another irregularity takes place when treating good thickly
liquid iron with an excess of rich cinder, as then the tem-
perature of the furnace rises too high owing to the energetic
oxidation; the iron then comes to nature too rapidly, causing
great loss of iron, and yields an inferior product if the pig-
iron under treatment is not very pure. The best remedies
for this irregularity are acid slags and lowering the tem-
perature of the furnace by almost closing the damper.
During the puddling operation the fire is usually stoked
at times when a cooling of the furnace is no disadvantage,
for instance, at the commencement of the process after
having cleaned the grate, in the first period of the process
when melting down the pig-iron, also when stirring the iron,
and at last after the formation of the balls. Skilful work-
men, however, will also regulate the temperature by a little
careful stoking at other parts of the operation.
The following are the results of different puddling works:-
a. Puddling works using mineral coal as fuel (page 771).
In Staffordshire two workmen (a puddler and a labourer)
in a turn of 12 hours will work off from five to seven charges of
RESULTS OF DIFFERENT PUDDLING WORKS.
849
4 or 4 cwts. The smaller number refers to grey pig, and
the larger to mixtures containing from to by weight of
fine metal. The loss of weight between the pig-iron charged
and the puddled blooms or bars produced is from 7 to 10 per
cent; and the consumption of coal amounts to from I to
to I'I ton per ton of puddled bars. The fettling materials
required during 12 hours to keep the hearth in proper order
are from 6 to 7 cwts. of bull-dog, and 2 or 3 cwts. of puddler's
mine, in addition to the mill scale added to the charge. One
hammer will shingle the balls from II or 12 puddling fur-
naces. Whilst the weekly average production of a single
furnace amounts to 18 tons, and that of a double furnace to
36 tons (the highest production of a double furnace is 46 tons),
the puddling furnaces in Staffordshire yield less, the double
furnaces about 28 tons, principally owing to the shorter
time of working.
In Scotland, where dark grey pig-iron rich in silicon is used.
without being previously refined, only four or five charges of
4 cwts. are made in the same time. The loss of weight is
from 15 to 18 per cent in converting the pig-iron into puddled
bars, and the consumption of coal per ton of puddled bars
is 25 or 26 cwts. When admixtures of fine metal and grey
forge pig, partly Scotch and partly hæmatite, are used, the
operations are accelerated and facilitated, though the con-
sumption of coal is not much less than when puddling raw
metal only, and the results are generally similar to those
obtained in Staffordshire.
In Cleveland and Durham, six or seven charges of 4 or 4
cwts. each of a mixture of grey forge iron and fine metal, are
worked off in 12 hours at a consumption of from 24 to
27 cwts. of coal per ton of puddled iron, and with a loss of
iron of 8 or 10 per cent.
The above quantities are in long cwts. of 120 lbs. each.
In Belgium the average weight of the charge is 230 kilos.
According to the quality of metal employed, the time re-
quired for each charge is from 1 to 24 hours, namely, 2 to
24 hours with grey pig, 13 to 2 hours with white pig, and
1 to 2 hours with fine metal. The loss in weight is from
3 I
VOL. II.
I
850
IRON.
7 to 10 per cent, and the coal consumed is equal in weight
to that of the puddled bars produced.
At Phoenixhütte in Ruhrort, a charge is 450 lbs., yielding from
385 to 390 lbs. of puddled bars, and from 120 to 140 lbs. of
coal are consumed for the production of 100 lbs. of puddled
bars. The number of charges in 12 hours amounts to from
five to ten, according to the quality of the iron employed; when
using a mixture of grey iron and fine metal, seven or
eight charges are usually made.
At Königshütte in the Hartz, a charge consists of 24 cwts. of
grey iron and 1 cwts. of white pig, and twelve or thirteen
charges are made in 24 hours, six or eight balls being formed in
each charge. The production per 12 hours is 20 or 21 cwts., the
yield of puddled iron being 87.7 per cent; the consumption
of coal is 832 lbs. per 100 lbs. of pig-iron, and 96 lbs. per
100 lbs. of puddled iron. The time used for the different mani-
pulations is 4 minutes for charging and repairing the hearth,
32 minutes for melting down the pig-iron, 33 minutes for
stirring the fused iron, 22 minutes for the formation of the balls,
10 minutes for pressing and heating the balls, and 7 minutes for
blooming the balls under a lift-hammer 6 or 7 cwts. in weight;
therefore one charge is finished in 1 hour and 48 minutes.
b. Puddling works using brown coal as fuel.
At Leoben, the charge is 4 cwts. of white partly cellular
pig-iron, which before being placed in the puddling furnace
is brought to an orange-red heat in a second hearth; in
5 minutes it is turned for the first time, and a second time
5 or 10 minutes later, when the iron will fall into pieces;
the workman then divides it with the rabble and stirs it, and
as soon as it has assumed a pasty state some rich cinder is
added according to the fusibility of the iron. Whilst the
pasty mass is continually moved the cinder quickly rises,
the iron comes rapidly to nature and projects out of the
cinder after having used the third rabble. The iron is now
turned once, divided, and formed into balls, which after being
heated for a short time are shingled under a hammer 12 cwts.
in weight and rolled out to mill-bars. These mill-bars are
cut up whilst still red-hot, formed into piles, and rolled out
to finished iron. A charge is worked off in 1 hour, usually
RESULTS OF DIFFERENT PUDDLING WORKS.
851
producing four balls at a loss in weight of 9'0 or 9'4 per cent,
and at a consumption of from 110 to 120 lbs. of coal per
100 lbs. of blooms. When re-heating the piles 9 cwts.
are charged after being previously warmed, and they lose
in weight from 15 to 20 per cent, according to the dimensions
of the finished iron to which they are worked up. The
consumption of coal amounts to from 100 to 110 lbs. per
100 lbs. of finished iron.
At Franzenshütte, the charge is likewise 4 cwts., and the
pig-iron treated is the radiated and cellular kind in plates
1 inch thick, each weighing from 80 to 100 lbs. The
melting down is finished in 25 minutes; the iron is then stirred
and worked with the rabble for 15 or 20 minutes whilst
mixing from 40 to 50 lbs. of cinder with the iron; the iron
has then come to nature and is shifted; the shifting takes
6 minutes, and 10 minutes are used for forming the balls,
usually four in number. The iron is then worked up as in
Leoben. A hearth lasts from 6 to 10 weeks; the slag hearth
is melted down in from 9 to 12 hours at a consumption of
7 cwts. of small coal and 3'5 cwts of coal in pieces,
and a cold furnace requires 20 cwts. of small coal to raise it
to the proper temperature. During operation a furnace con-
sumes 37°5 cwts. of coal in 12 hours. The loss of iron
amounts to 6.6 per cent.
c. The puddling process when turf is used as fuel
(page 773).
At Maximilianshütte, the weekly production of a puddling
furnace is 350 cwts., and 14'5 cubic feet of turf are consumed
per 100 lbs. of blooms.
d. The puddling process when wood is used as fuel
(page 773).
At Surahammer and Nyby in Sweden, double furnaces are
used, working off six charges of 600 or 700 lbs. of mottled
iron each in 12 hours. The loss of iron amounts to 6 or 7
per cent, and from 10 to 14 cubic feet of wood are consumed
per 100 lbs. of blooms.
e. The puddling process with gases produced with brown
coal (page 786).
At Prävali, from twelve to fourteen charges are worked off
3 1 2
852
IRON.
in double furnaces in 24 hours. A charge consists of 8 org cwts.
of white forge pig. The loss of iron is 9 or 10 per cent, and
from 122 to 140 lbs. of brown coal are consumed per 100 lbs.
of blooms. Upon re-heating the puddled iron 126 or 129 lbs.
of coal are consumed per 100 lbs. of iron, at a loss of from
11.6 to 13 per cent.
At Krems, a double furnace produces 83.86 cwts. of rolled
bars in 24 hours at a loss of 63 per cent of iron and at a
consumption of 116 lbs. of coal. A re-heating furnace works
off 66.87 cwts. of iron in the same time, causing a loss of
15.84 per cent of the latter and consuming 170 lbs. of coal
per 100 lbs. of iron to be heated.
f. The puddling process with gases produced with turf
(page 775).
At Nothburgahütte, near Freudenberg, a double furnace is
charged with 8 cwts. of previously warmed white (partly
cellular) forge pig, and a charge is worked off in 1 hour and
15 minutes, namely, 5 minutes are used for charging the
furnace, 25 minutes for melting down the iron, 15 minutes
for stirring it, 15 minutes for working it into balls, and
15 minutes for shingling the balls. 10 or 12 cubic feet of
kiln-dried turf and I cubic foot of wood are consumed per
100 lbs. of blooms, and the yield of iron 94 per cent. (At
Mandelholz in the Hartz 30 cubic feet of turf were consumed,
and the yield of iron was 90 per cent.)
At Buchscheiden, near Klagenfurt, 8 cwts. of previously
warmed white and mottled forge pig are charged before all
the balls of the previous charge are withdrawn; they melt
down in about 30 minutes, and are mixed with cinder when
they have scarcely attained the pasty state; the blast of
combustion is now removed and re-applied when using the
third rabble; the mass is turned twice and formed into
twelve balls, eight of which are rolled out to mill-bars im-
mediately after they have been shingled under a hammer
weighing 12 cwts., and the other four blooms are returned to
the furnace. A charge is finished in 13 hours, and yields from
92 to 94 per cent of mill-bars, consuming from 14 to 18 cubic
The workmen work
feet of turf per 100 lbs. of puddled iron.
in shifts of 5 or 5 hours, in which they finish three charges.
RESULTS OF DIFFERENT PUDDLING WORKS.
853
The re-heating furnaces are heated with kiln-dried turf
and wood, and charged each with three piles weighing about
1500 lbs., which are rolled once when they have been heated
for an hour, and of an hour later they are drawn out to
rails. Ten charges are made in 24 hours; the total loss of
iron is 15 per cent, and the consumption of turf from 12 to
18 cubic feet per 100 lbs. of iron, and that of wood 164 cubic
feet for the ten charges.
g. The puddling process with gases produced with wood
(page 775).
At Zorge in the Hartz, grey charcoal pig-iron is puddled
in charges of 4 cwts. each, yielding 83 per cent of mill-bars,
at a consumption of 10 or 13 cubic feet of wood per 100 lbs.
The balls are shingled under a lift-hammer weighing 53 cwts,,
making 70 blows per minute, and being lifted 20 inches
high.
I
4
At Lippitzbach, double furnaces are used producing about.
455 kilos. of mill-bars in 1 or 2 hours from 476 kilos. of
pig-iron; 1*047 tons of pig-iron and I'OLI tons of kiln-dried
wood are consumed for the production of I ton of mill-bars.
They are re-heated in reverberatory furnaces with the same.
kind of gaseous fuel, consuming 13 tons of wood per ton of
finished iron, which results from 1187 tons of mill-bars.
8 cwts. of white and grey pig-iron are charged, yielding
95 or 96 per cent of blooms in 1 hours, at a consumption of
4 cubic feet of wood per 100 lbs. of blooms.
The weekly
production is 400 or 420 cwts.
At Neuberg, 8 cwts. of white or mottled iron yield in
2 hours 94 or 95 per cent of blooms, consuming 5 cubic feet.
of raw wood per cwt. of blooms. The weekly production
varies between 290 and 400 cwts.
h. The puddling process in Siemens's regenerative gas
furnace.
Mr. Siemens has kindly furnished us with the following
details:-
"Tables 1 and 2 give the working results which were ob-
tained from Siemens's furnace, compared with the results
obtained at the same time in an ordinary furnace from
the same pig (theordinary forge mixture).
854
IRON.
Nos. of
Heats.
REGENERATIVE GAS FURNACE.
Time
Charged.
Table No. I.
First
Ball Out.
First Shift..
Metal
Charged.
Yield.
h. m.
h. m.
lbs.
lbs.
I.
5.25
6.32
410
392
2.
6.45
7.50
433
396
3.
8.8
9.9
430
410
4.
9.15
10.7
425
426
5.
I0.20
II.22
426
430
6.
II.40
12.46
412
412
Second Shift.
I.
1.48
2.47
428
410
2.
2.50
3.47
420
414
3.
3.56
4.53
426
418
4.
5.00
6.3
432
417
5.
6.5
7.12
425
407
6.
7.20.
8.15
420
422
Third Shift.
I.
9.10
10.15
423
414
2.
10.25
11.30
422
412
3.
II.35
12.40
420
420
orer A
4.
12.45
2.00
430
410
5.
2.10
3.10
424
418
6.
3.16
4.20
420
400
First Shift.
I.
5.38
6.45
423
402
2.
6.50
8.00
422
400
3.
8.6
9.8
430
390
4.
9.15
10.25
426
407
5.
10.35
II.45
426
420
6.
II.55
1.8
430
416
Second Shift.
I.
2.00
3.I
422
422
2.
3.6
4.00
424
415
3.
4.5
5.18
423
424
4.
5.23
6.27
423
415
5.
6.33
7.46
427
420
6.
7.49
8.50
420
406
RESULTS OF DIFFERENT PUDDLING WORKS.
855
h. m.
h. m.
Table No. 1 (continued).
Nos. of
Heats.
Time
Charged.
First
Ball Out.
Third Shift.
Metal
Charged.
Yield.
lbs.
lbs.
I.
10.00
II.20
420
424
2.
II.25
12.33
420
410
3.
12.40
I.45
423
412
4.
1.50
2.58
425
420
5.
3.13
4.20
430
418
6.
4.30
5.35
422
426
cwts. qrs. lbs.
Total charge.
136 I
2
yield
""
132
3
7
20
2
2
Being at the rate of
of pig-iron per ton of puddled bar.
Time.
ORDINARY FURNACE.
Table No. 2.
Weight of Metal Charged.
Weight of Puddled
Bars Produced.
These times were not taken for each
charge, but six heats were produced
every twelve hours.
Mean, 484 lbs.
lbs.
424
425
405
430
430
438
416
456
410
432
426
420
422
422
425
430
450
410
Ibs.
Mean charge.
yield
,,
484
426
or 22 cwts. 2 qrs. 20 lbs. of pig-iron per ton of puddled bar.
856
IRON.
"It will be observed that the ordinary furnace received
charges of 484 lbs. each, and yielded on an average 426 lbs.,
representing a loss of 12 per cent, whereas the gas furnace
received charges averaging 424 lbs., and yielded 413 lbs.,
representing a loss of less than 2.6 per cent.
"It is important to observe, moreover, that the gas furnace
turned out 18 heats in three shifts per 24 hours instead of
only 12 heats per 24 hours, which was the limit of production
in the ordinary furnace.
"This rate of working was attained without the employment
of any arrangement for heating the pig-iron before charging
it into the furnace, the heating chambers at the ends not
having been used. The adoption of the plan of heating the
metal beforehand effects a further saving of ten or fifteen
minutes in the time required for working each charge, as well
as a considerable economy in fuel.
"The quality of the iron produced from the gas furnace was
proved decidedly superior to that from the ordinary furnace,
being what is technically called 'best best' in the one and
'best' in the other case from the same pig.
"The economy of fuel was also greatly in favour of the gas
furnace, but could not be accurately ascertained, because
some mill furnaces were worked from the same set of pro-
ducers. Still, judging from the experience of several years
in the working of regenerative gas furnaces as re-heating or
mill furnaces, and as glass furnaces, the saving of fuel in
puddling cannot be less than 40 or 50 per cent in quantity,
while a much cheaper quality may be used.
"The consumption of 'fettling' was, however, greater in
the gas furnace, and the superior yield was naturally attri-
buted by the forge managers to that cause, although the
writer held a different opinion.
"The gas furnace, however, had not been provided with
water bridges; these were subsequently added, and the fur-
nace put to work again in February, 1868, since which time
it has been worked continually.
"The result of the water bridges has been that the amount
of "fettling" required is reduced to an ordinary proportion,
the average quantity of red ore used being 92.6 lbs. per
RESULTS OF DIFFERENT PUDDLING WORKS.
857
charge, besides the usual allowance of bull-dog, while the
yield per charge of 483°3 lbs. of grey forge pig has been in-
creased to 485 lbs. of puddled bar, as shown by the following
return of a series of 80 consecutive charges in June last :—
Total Charges and Yields.
Average per
Heat.
Nos. of
Heats.
80. Pig-iron charged.
Puddled bar returned
Red ore for fettling
lbs. cwts. qrs. lbs.
lbs.
38,668=345
I O
483'3
38,808=346
I 25
485'0
7,406 66 0 14
92.6
•
proving that the yield of puddled bar slightly exceeds the
charge of pig metal (representing a saving of fully 12 per
cent over the ordinary furnace), while the superiority of
quality in favour of the gas furnace is fully maintained. .
"It is also worthy of remark that these results are obtained
regularly by the ordinary puddlers of the works, and that no
repairs have been necessary to the gas puddling furnace for
more than six months, the roof of the furnace keeping also
in excellent condition.
"In these investigations I have confined myself to the
puddling of ordinary English forge pig in order to avoid con-
fusion, but it is is self-evident that the same reasoning also
applies in a modified degree to white pig metal or refined
metal, the use of which I should not, however, advocate.
"Regarding the water bridges, I was desirous to ascertain
the expenditure of heat at which the saving of "fettling"
and greater ease of working was effected. The water passing
through the bridges was accordingly measured by Mr. W.
Hackney, and found to amount to 25 lbs per minute, heated
40° F. This represents 60,000 units of heat per hour, or a
consumption not exceeding 8 lbs. to 10 lbs. of solid fuel per
hour, an expenditure very much exceeded by the advantages
obtained where water or cooling cisterns are available.
"The labour of the puddler and of his underhand being very
much shortened and facilitated by means of the furnace, I
should strongly recommend the introduction of three working
shifts of 8 hours each per 24 hours, each shift representing
the usual number of heats, by which arrangement both the
employer and the employed would be materially benefited. The
858
IRON.
labour of the puddler may be further reduced with advantage by
the introduction of the mechanical rabble, which has already
made progress on the Continent.
66
By working in this manner, a regenerative gas puddling
furnace of ordinary dimensions would produce an annual
yield of about 940 tons of bar iron of superior quality from
the same weight of grey pig metal and the ordinary propor-
tion of fettling.
"A considerable number of these puddling furnaces have
been erected by me abroad, and in this country they are also
taken up by the Monkbridge Iron Company, Leeds, and a
few other firms.
"The construction of these furnaces has been still further
improved lately by the application of horizontal regenerators
to save deep excavations, and by other arrangements whereby
the first cost is diminished and the working of the furnace
facilitated."
2. The Puddling Process for the Production of
Fine-grained Iron.
This process differs from the preceding process for the
production of fibrous iron in the following points :-
a. As the fine-grained iron which excels in homogeneous-
ness, compactness, and tenacity, contains a larger proportion
of carbon than fibrous iron, the process for its production is
shorter and it requires a superior pig-iron (white radiated,
specular, or grey pure charcoal iron), as sufficient time is not
given to separate the noxious substances. Refined cold
blast coke pig-iron is also used for the production of fine-
grained wrought-iron, but seldom by itself (Low Moor), and
more frequently in admixture with good grey or mottled
varieties of pig; refined metal by itself can be puddled only
with great difficulty (page 709). When less pure kinds of
pig are to be used the puddling process is artificially
prolonged by the application of poor cinder, a higher
temperature, purifying fluxes (Schafhäutl's powder), &c.
Manganese in the pig-iron causes the formation of a thinly
liquid slag, prolongs the decarbonisation by precluding the
THE PUDDLING PROCESS FOR FINE-GRAINED IRON. 859
iron from contact with the atmospheric air, and promotes
the purification of the resulting wrought-iron. This is
the reason why manganiferous pig is peculiarly fitted for the
production of fine-grained wrought-iron.
b. Smaller charges are sometimes employed, as they insure
a more perfect success; they are chiefly used when treating
refined metal (a charge at Low Moor amounts to only 2 cwts.,
at Vierzon to 180 kilos.), which consumes a considerable
amount of fuel in becoming sufficiently fluid (at Low
Moor 300 lbs. of coal are consumed
are consumed per 200 lbs. of
refined metal). Small charges increase the expenses of the
process.
c. The larger quantity of poor cinder employed prolongs
the process and impedes the contact of the iron with
atmospheric air, consequently less magnetic oxide is formed
and the cinder remains poorer. (Fibrous iron poor in car-
bon is formed when employing a smaller proportion of basic
slag rich in magnetic oxide). The quality and larger quan-
tity of the poor slag employed necessitate a deeper hearth,
the side walls of which must be cooled with water page 787).
To prevent the air from acting too energetically upon the
fused mass the puddling furnace has a somewhat higher
roof, and when employing gas firing the blast is conducted
in a direction nearly parallel with the roof.
d. In order to produce as uniform a product as possible,
the fused metal must be stirred for a longer time (two or
three rabbles more, or of an hour longer); a higher tempera-
ture is, therefore, required at this stage of the process and
also at the melting down of the pig-iron than for the pro-
· duction of fibrous iron, and for the same reason the puddling
furnaces require a considerable draught. When bringing
the iron to nature, a reducing flame is employed; it is pro-
duced by keeping a good layer of coal upon the grate and
by partially closing the damper; when using gas firing the
blast must be reduced. When producing fibrous iron an
oxidising flame is applied at this period of the process.
Before it is formed into balls the metal is usually transferred
only once, and the balls are formed and withdrawn from the
furnace quickly, thus making up for the longer time used
860
IRON.
in stirring the iron. The balls are sometimes withdrawn.
immediately after being formed in this manner, so that one
ball is under the hammer whilst the following one is still
in formation.
A smaller number of larger balls are usually formed, and
this also influences the decarbonisation; the balls are
shingled under a hammer more perfectly and carefully, and
larger grooves are employed upon rolling the blooms. The
re-heating of the mill-bars requires a very high tempera-
ture.
The wrought-iron produced may be harder or softer
according to the fineness of the grain, which depends chiefly
upon the proportion of carbon contained in the iron; on
the other hand, the grain is also influenced by the degree
of temperature at which the iron is treated, and also by the
quickness of the cooling of the iron. The operation of rolling
is a more uniform mechanical treatment than hammering;
rolled iron, therefore, has usually a more uniform grain than
hammered iron.
The fine-grained iron must not be confounded with
another form of granular crystalline iron, which is hard and
rich in phosphorus and frequently used for the central part
of common rails. This kind of iron is produced from im-
pure pig-iron with a larger addition of poor cinder or, in
some cases, of sand or ground bricks, diminishing both
the time of boiling and working at a low temperature.
The following results have been obtained in different iron
works at the production of fine-grained iron :-
1. At Alvenslebenhütte* in Silesia. Charges of 400 lbs.
of a mixture of 250 lbs. of grey pig and 150 lbs. of refined
metal are employed for the production of fibrous wrought-
iron as well as of that of fine-grained wrought-iron; the
former is puddled in the dry puddling furnace (page 787),
and the latter in the furnace represented in Figs. 239,
240, and 241 on page 786, and the following results are
obtained :-
Berggeist, 1861, No. 78. HARTM., Fortschr., v., 195. B. u. h. Ztg., 1860,
p. 399; 1862, p. 186. Schles. Wochenschr., 1859, No. 50.
RESULTS IN PUDDLING FINE-GRAINED IRON.
861
Wet
Dry
Puddling. Puddling.
Charging and melting down.
40
40 minutes.
Stirring with four or five rabbles
30
30
""
Working the metal with a pointed
rabble
15
25
وو
Formation of the balls.
ΙΟ
20
""
Forging the balls
15
15
""
Total length of the operation
ΙΙΟ
•
130
وو
Loss of iron
9
14-16 per cent.
Number of charges in 12 hours
•
6-7
5—6
about 3*15 cubic feet.
Consumption of coal per 100 lbs. of
fine-grained mill-bars
2. Pielahütte in Upper Silesia. Six charges of 400 lbs. each
are worked in 12 hours, at a loss of 9 or 10 per cent of iron.
3. Low Moor. The puddling furnaces are comparatively
small (the hearth is only 3 feet broad) with a very high stack,
in order to command a strong heat. The metal used is re-
fined metal, produced from cold blast coke pig-iron, and
2cwts. (long weight) of refined metal forms one charge,
nine of which are made in 12 hours. Each charge is
previously heated to redness in a second bed with which the
puddling furnaces are provided, and which are heated in the
usual way by the waste flame. Only two or three balls are
formed, which, after shingling under a 7-ton helve hammer into
plates or stamps 10 or 12 inches square and 1½ inches thick,
are broken into pieces by blows from a heavy weight falling
from a considerable height. These pieces are sorted ac-
cording to the fracture, and piled into small heaps called.
balls," which are heated in a re-heating furnace and
hammered into blooms under a steam hammer. The
blooms are then re-heated and hammered several times.
(usually three times) again so as to make them perfectly
sound; they are then worked into the required form. The
pieces with the most uniform crystalline fracture are used
for railway tires and purposes requiring hard iron, whilst
blooms which are less uniform or show fibre are better fitted
for making boiler plates and wire rods. The loss of iron in
puddling amounts to 4 per cent, and the consumption of
coal to 30 cwts. per ton of fine metal treated, or 37 cwts.
per ton of blooms produced.
862
IRON.
The Bowling and Farnley Companies and several other
firms in South Yorkshire work upon the same system, and
produce iron equal in quality to that of Low Moor.
4. Vierzon.* Four or five charges, each of 180 kilos. of
grey manganiferous pig-iron, are worked in 12 hours at a
loss of 5 per cent of iron, and at a consumption of from
1120 to 1360 kilos. of coal per 750 or 800 kilos. of fine-
grained wrought-iron. When working fibrous wrought-iron
seven charges of 180 kilos. each are made in 12 hours; the
loss of iron is 8 or 8 per cent, and the consumption of coal
is the same as in the production of fine-grained iron.
The composition of puddle cinder is shown by the
following analyses :-
I.
II.
III.
IV.
Sio,
16:53 28.04 36.8
46.1
FeO
66*23 70'45
61.0
45'0
MnO
4'90
0'44
I'9
CaO
0'70
0*28
ΙΟ
MgO
0°30
2'0
A1203
I'04
0°24 I'5
3°0
Fe₂O₁
FeS
6.80
PO5
3.80 P. 0.16
S
0'09
No. I is cinder remaining in the furnace after the re-
moval of the balls, analysed by Calvert and Johnson ;
formula, 6(3 FeO,SiO) + 5(6FeO,SiO3); its proportion of
oxygen is as 9*08: 15'59. No. 2. Cinder from Sore in Styria.
No. 3. Cinder from Dowlais; compact, greyish black,
analysed by Berthier; formula, 2FeO,SiO3. No. 4. Cinder
from Skebo in Sweden, analysed by Sefström; formula,
3 FeO2SiO3.
Re-heating the Blooms, &c.-The blooms vary in size
according to the purpose for which they are intended; they
are not usually sound or uniform, having cracks and
containing an admixture of cinder. To render them
perfectly homogeneous and compact they must be repeatedly
heated at a strong white heat, and treated each time
mechanically, which is most effective by applying hammers;
* B. u. h. Ztg., 1860, p. 170.
RE-HEATING THE BLOOMS, ETC.
863
the tenacity of the iron is also increased. The operation of
welding shows faults which the blooms may contain, such as
particles of raw iron, an intermixture of cinder, &c., and
allows their removal as well as the extraction of smaller
proportions of silicon, phosphorus, manganese, and sulphur
contained in the blooms, owing to the oxidising action of the
intermixed rich cinder. To prevent excessive loss of iron by
oxidation in the welding heat and at the mechanical treat-
ment, some fine sand is thrown upon it, to combine with the
oxidised iron and form a protecting coating of slag.
Iron bars of various size and form, which are required to
be perfectly homogeneous, are merely drawn out from a solid
piece of iron, as the employment of piled up iron makes it
likely that some of the joints formed by the heaped up bars
will not be perfectly welded; but the use of piles is in most
cases necessary for economical reasons. The commoner
kinds of bar iron are made of puddled or mill-bars (iron No. 1),
which are cut into lengths, piled up, and rolled when brought
to a welding heat; the piles are sometimes first worked into
a bloom under a hammer. In most cases, however, that
produced from the second rolling (iron No. 2), is employed
as the top and bottom plates of the piles, thus insuring a
smoother surface of the resulting bars, which are called
No. 3, or best iron. Upon piling and welding this iron it is
distinguished as best best, which then may be improved to
treble best by welding and rolling it again, and so on.
The object to be rolled always fixes the size of the pile
and the arrangement of the different qualities of iron; but
this composition is frequently modified by other than scientific
circumstances, such as the convenience of the iron works,
the cheapness of the article to be produced, &c. For in-
stance, rails, which perhaps form the chief article of the
rolling mills, are necessarily made of inferior iron and with
as little work as possible if they are contracted for a low
price. Sometimes even the composition of the piles is pre-
scribed by the railway companies. For those reasons the
piles for the common cheap rails are usually composed of
about 1-5th of iron No. 2, and 4-5ths of iron No. 1; the iron
No. 2, consisting of bars an inch thick, is so arranged in
864
IRON.
the piles as to cover the top and part of the sides. The iron
No. 1 is usually 1 inch thick and 3 inches broad. The
piles themselves are in some cases 7 inches square and about
4 feet long, and in others 8 inches broad, 10 inches high,
and nearly 3 feet long. These latter piles are rolled out in
two heats, thus insuring a better welding; but in many
cases the thin top pieces of the pile are so much cooled on the
way from the furnace to the rolls as to lose their proper
welding heat; thus, on rolling, they merely adhere to the
rails by virtue of the compressing power, but when using
those rails on the lines the top plate is apt to come off,
rendering the rail useless. The different nature of the two
kinds of iron also make a proper welding a difficult opera-
tion. In many cases part of the iron No. 1 in the piles is
replaced by crop-ends of rails, and old scraps of wrought-
iron.
Many railway companies have profited by experience, and
now procure their rails in the proper way. They leave the
manufacture entirely in the hands of the iron-masters, and
pay the price for a good article, which they secure by a
guarantee from the iron-master that all rails becoming faulty
in the first, second, third, and even fifth years of their use,
must be replaced gratis by the iron-master. This guarantee
has in every case answered its purpose, as it ensures a
proper welding of the rails and the use of a good material.
On the Continent the top of the piles for the warranted
rails is usually formed of a single iron slab of fine-grained
iron 2 or 3 inches thick, and the piles are worked off in two
heats, hammering them under a steam hammer after the
first heat, and rolling them into rails after the second heat.
Beattie's system of piling adopted for railway axles has
proved advantageous. It consists of a ring built up of
several segments, arranged round a central bar.
To increase the strength of iron articles, chiefly boiler
plates, the bars are placed crosswise over each other in the
pile, so that the fibres in one layer may be at right angles
with the fibres of the next layer.
When requiring piles I ton or more in weight, for the
manufacture of armour plates, cranks, &c., a number of
RE-HEATING IN HOLLOW FIRES.
865
blooms, or lighter piles of suitable composition, are
hammered or rolled and successively piled one upon the
other.
The loss in re-heating the blooms (from 10 to 20 per cent)
depends chiefly on how perfectly the slag has been pressed
out of the blooms when shingling the balls.
The slag may
be retained in the blooms to a certain degree, when shingling
them under squeezers or by careful blows of the steam
hammer, and the yield of the puddling process varies ac-
cordingly. When rolling the blooms into mill-bars, the
removal of the slag is of less importance, and the welding
of the piles formed of the mill-bars offers a further oppor-
tunity for that removal; a certain amount of slag in mill-bars,
which are still partly raw, may be even advantageous, as it
helps the raw parts to come to nature.
A. Re-Heating in Hollow Fires.
This method of re-heating is but little used. The opera-
tion is carried on in the iron works of the Upper Hartz in
the following manner :-The hearth is filled with a moistened
mixture of washed coal and cinders, blast of low pressure
is introduced; the bloom fixed in tongs is placed 2 or
3 inches above the current of blast, and covered with coal,
which then soon cokes and forms a vaulted roof. The blast
passes below the bloom, is repelled from the opposite side of
the hearth, thus playing round the bloom, and escapes by an
aperture left in the arched roof of coke into the warming
stove. As soon as the bloom has become red-hot it is turned
and somewhat lowered whilst increasing the blast, and it is
turned again when its lower part approaches a white heat;
it is then further lowered, and the blast is employed at its
greatest pressure. Sand is now thrown on the bloom, and
it is kept for some time in the highest heat of the hearth
it is then sufficiently hot for the hammer, under which it is
drawn out, leaving a small lump on one end of the bar.
After each turning of the bloom the loosened coal is pressed.
round it and fresh coal charged.
A good, well-fined, bloom glows uniformly, and shows a
smooth surface, and a bloom which is still partially raw
VOL. II.
3 K
866
IRON.
scintillates and glows unequally on its different parts. In
such cases blast of lower pressure is employed to prevent the
raw particles from melting out, and to facilitate their coming
to nature by the decreased temperature. Upon employing
too strong a blast, a melting of the raw particles would take
place, and the fined iron would be too much decarbonised,
and become burnt and friable. When re-heating raw blooms.
a larger quantity of sand is employed than is used for well-
fined blooms in order to impede the oxidation.
Whilst drawing out the bloom, the cinder is tapped off
from the hearth and deposits are removed, the hearth is
filled with fresh coal, and another bloom, previously warmed,
is introduced together with the lump attached to the drawn
out bar; the vaulted roof is likewise repaired. When the
second lump has become white hot, the bloom will be red,
and will rapidly rise to a welding heat; the lump and bloom
are drawn out under the hammer.
One part of the hammer face is 12 inches square, and
another 9 inches by 3 inches. The bloom is treated first
with some blows of the large part of the hammer face, thus
causing any slag which may have adhered to the bloom to fall
off, and blunting its sharp edges; it is then drawn out by the
oblong face as long as it remains soft, and it is continually
turned. Next the bar is again hammered with the square
face to remove the impressions made by the other face, and
if a further drawing out is required, the bar is replaced in
the hearth or the warming stove. Before bringing the hot
iron under the hammer it is always turned round in a heap
of sand.
At Königshütte in the Hartz, 100 lbs. of blooms yield 82.45
per cent of finished iron, at a consumption of 45.8 lbs. of coal
per 100 lbs. of finished iron. The production in 24 hours
amounts to 15 or 16 cwts. of bar-iron; 100 lbs. of pig yield
about 72 lbs. of bar-iron.
B. Re-heating in Reverberatory Furnaces.
These furnaces (page 807) are used for re-heating when a
larger production is aimed at, and their charges vary from
a few cwts. to over 10 tons. The piles or blooms are better
RE-HEATING IN REVERBERATORY FURNACES.
867
placed in the furnaces crosswise than lengthwise, and they
must not touch each other. According to how much they
have been previously warmed, they are placed either in the
middle of the hearth or near the fire-bridge; the coolest piles
are placed before the flue-bridge. They are several times.
turned, and sand is thrown upon them before they are with-
drawn, which is done as soon as the edges become rounded.
The hearth is repaired from time to time after several heats;
it is best done with pebbles or large pieces of quartz, as they
fly into small pieces and stick to the hearth under the
influence of the heat.
The loss of iron in the operation of re-heating varies with
the number of heats, and amounts to from 12 to 20, on an
average 15, per cent in the first heat, without taking into
consideration the rough ends; the loss is less in the later
heats. The consumption of fuel has been stated on page
777, and we have given several results of the re-heating pro-
cess on page 848.
The resulting cinder is used in the same way as the pud-
dling cinder. This cinder runs from the inclined hearth into
the flue, the deepest part of which is provided with a tap-
hole, which is always kept hot by a coal fire, thus allowing
the cinder to be tapped off at any time.
The re-heating cinders are richer in iron than the cinder
resulting in puddling furnaces and finery fires; they seldom
contain less than 60 per cent of iron; they are more difficult
to reduce, and are little influenced by roasting or the opera-
tion of decaying.
The following analyses show their composition:-
I.
II.
III.
SiO3 ·
15°15
25'4
42°2
FeO.
76.73
68.8
52°0
MnO
I'51
CaO.
trace
MgO
trace
A1203
I'95
5'0
3°3
Fe₂03
3'I
S.
* 1*36
P.
2.22
No. I is cinder from Witkowitz, analysed by Mayrhofer.
868
IRON.
No. 2, cinder from Lohhütte near Siegen, analysed by
Schnabel; its specific gravity is 4'167. No. 3 is cinder from
Dowlais of the formula 6FeO,SiO,, analysed by Berthier.
The following are the results of some of the re-heating
furnaces:
In England four charges of piles intended for rails are
usually made, each weighing about 5 cwts. One charge is
heated in about an hour, and the re-charging of the furnace,
including the repairs of the hearth, takes 16 minutes; 36 piles.
are heated in 12 hours, or about 83 tons per week in each
furnace. When using double heat, ten furnaces are employed
for the first heat and four for the second, in order to keep the
mill constantly at work. The loss, including the crop ends,
is about 20 per cent of the weight of the pile.
When producing common bar-iron, 16 or 18 piles, about
18 inches long, 3 inches broad, and 2 or 3 inches high, are
heated in about 30 minutes, and 31 tons are heated weekly
in one furnace, but from 15 to 20 tons only of the finer sorts
of iron. The loss of iron varies with the size of the piles,
and amounts for these piles to 80, 130, and 210 lbs. respec-
tively per ton, and the consumption of coal to 7, 10, and 13
cwts. respectively per ton of piles.
The total amount of coal consumed in the manufacture of
iron from the ore to the finished bars of common or No. 2
quality may be taken as five times the latter, with an in-
crease of about 10 cwts. per ton for every additional heat.
At Karolihütte near Dernoe (Hungary), a furnace is charged
with seven piles, each weighing from 240 to 260 lbs. (formed
of eight mill-bars). When up to welding-heat, the piles are
first hammered under a hammer 24 cwts. in weight, which
is lifted 32 inches high, and gives 65 or 70 blows
per minute; the re-heating and the hammering are once
repeated, when the piles are heated for the third and last
time, and rolled into the size required; the loss is 20 per
cent of the weight of the piles. When re-heating blooms,
50, each weighing 28 lbs., are charged, after being first warmed.
The charging is effected in a quarter of an hour; the re-
heating takes half an hour, and the rolling out three quarters
of an hour; 5 or 6 cwts. of quartz are consumed in 12 hours
RE-HEATING IN REVERBERATORY FURNACES.
869
for repairing the hearth; ten or eleven charges are made in
12 hours, causing a loss of 14 or 16 per cent of iron, and
48 lbs. of coal are consumed per 100 lbs. of finished iron.
At other iron works in Dernoe re-heating furnaces are
used with a greatly inclined hearth; from 60 to 90 cwts. of
blooms are re-heated in six charges in 12 hours. The loss
of iron amounts to 18 per cent, the consumption of coal to
50 or 60 cwts. per 12 hours, and that of quartz to 12 or
15 lbs. per 100 lbs. of finished iron. The hearth bottom of
quartz is three or four times renewed in 12 hours, and when re-
heating large piles, after each charge. A furnace lasts from
2 to 6 months, but the fire-bridge and flue require to be
renewed every week. 20 cwts. of coal in pieces are re-
quired to raise a cold furnace to the required temperature,
which takes 6 hours. The piles, usually seven in number,
are placed in one row, and blooms in two or three rows, the
long way of the hearth.
Some Examples of Finishing Wrought-Iron of Different
Kinds. Bars of 1 inch square may be rolled out of piles
18 inches long made up of six bars, 4 inches by 3 of an inch,
the top and bottom plates being iron No. 2, whilst the other
bars are mill-bars. Round bars, 4 inches in diameter and
16 feet long, are rolled from piles 10 inches by 11 inches,
and 6 feet long, the top and bottom plates also con-
sisting of iron No. 2 or No. 3.
Tunner has reported on the manufacture of tyres, T-iron,
wrought-iron tubes, &c.; a more particular account of it
would occupy too much of our space. The manufacture of
rails and of bar iron of different form is described in the
articles and treatises given in our note.†
We have already mentioned that the piles for the heavier
plates are formed of layers of bars, placed alternately across
each other, the top and bottom plates being made each of
one bar of iron No. 2, either rolled or hammered. A boiler
*
LEOB., Jahrb., 1852, ii., 124; 1960, ix., 176; 1863, xii., 83.
Maass. und Gewichtsverhältnisse,' pp. 189, 521, 522.
Maurer,
B. u. h. Ztg., 1843, p. 1105; 1845, p. 320; 1847, P. 778; 1850, p. 439;
1854, PP. 44, 141; 1857, p. 129. HARTM., Fortschr., i., 384; ii., 307; iii., 277;
V., 224. Ansiaux et Masion. &c. (p. 483). Polyt. Centr., 1859, No. 27.
GRUNER et LAN, état présent, &c.
870
IRON.
plate 6 feet long, 3 feet broad, and 3-16ths of an inch thick,
is made of a pile 20 inches long, 6 or 7 inches high, and
12 inches broad.
Plates and sheets* are classified according to their thick-
ness; the former term being restricted to all sizes above
No. 4 of the Birmingham wire gauge, corresponding to a
thickness of o°238 inches. Sheet iron is further classified
into three divisions, as follows:-
Singles, including from No. 4 to No. 20 gauge or o'238 to 0.035 inches thick.
Doubles
Trebles or lattens
""
20
25
""
25
27
""
0'035 :)
0'020
0'020 0'016
Thin sheet iron or black plates require a tough and
soft iron, and this material is at present more frequently
provided by puddling furnaces than by the finery fires.
The iron employed must be as free as possible from inter-
mixed cinder and scalings; a little red-shortness is of less
importance. This iron is piled up and rolled so as to form
bars, for instance 4 inches broad, 1 or 1 inches thick, and
from 2 to 4 feet long, weighing from 20 to 60 lbs. These bars
or plates are re-heated and rolled, first each bar by itself and
then two together. The resulting plates are cut to small sizes,
re-heated, and again rolled, taking more plates (up to fifty)
on the top of the other, coal dust being thrown between the
different plates. The last rolling is effected under hard
rolls.
The re-heating of the plates is mostly effected in a re-
verberatory furnace provided with a proportionally small
grate and high fire-bridge to protect the plates from oxida-
tion, and to facilitate their turning. The hearth has pro-
jections upon which the plates rest; the flue is sometimes
conducted downwards towards the chimney, and sometimes
upwards. A hearth is now and then constructed between
the flue and chimney for the preparatory heating of iron.
The plates are frequently heated by the waste flame of other
furnaces, and two or three hearths are placed one above the
other. In order to prevent the oxidation of very thin plates,
they are heated in well closed cast-iron boxes. A special
* BAUERMAN, Metallurgy of Iron, &c., p. 317.
THE MANUFACTURE OF TIN PLATES.
871
grate is sometimes placed before the working opening of the
furnace, to prevent a cooling of the hearth (Alvensleben-
hütte).* Before the last heating the plates are sometimes
brushed with muriatic acid, and well beaten to remove their
scale.
Owing to the great number of re-heatings and the large
surface of the finished work compared with its weight, the
loss and consumption of materials are very great. For
the production of 1 ton of sheets sheared to the proper size,
25 cwts. of coal are required, and 25 or 26 cwts. of rough
bars. There is a waste of 4 or 5 cwts. on this quantity,
about 4 cwts. of which are accounted for in the shearings
and crop ends produced at different stages of the process.
No great amount of scaling takes place, owing to the com-
paratively low temperature at which the work is done.
Before shearing the plates they are annealed again, and
covered with old plates, to obviate brittleness and tension.
When reduced to the proper dimensions, the plates are
brought to a bright metallic surface by pickling in weak
sulphuric acid. A final polish is given by cold rolling, after
which the plates or sheets are ready for tinning.
The Russian, Belgian, Swedish, and other plates have
a great reputation. The manufacture of sheets for the
manufacture of buttons requires the very best wrought-iron.§
To prevent the oxidation of black plates they are coated
either with tin or zinc; zinc prevents oxidation even if
partially coating the iron, whilst tin would facilitate oxida-
tion.
The Manufacture of Tin Plate.-This process, which
is now carried on in some of the best works, is de-
scribed in Ure's "Dictionary of Arts and Mines," iii., p. 918,
as follows:-"When the sheet iron leaves the plate
mill, and after separating the plates, and sprinkling
B. u. h. Ztg., 1862, p. 206.
+ BAUERMAN's Treatise on Metallurgy of Iron, p. 620.
P. 56.
TUNNER'S Bericht. über die Londoner Industrie-Ausstellung, in 1862,
|| Ibid., p. 52.
§ LEOB., Jahrb., 1852, ii., 124, 171.
872
IRON.
between each plate a little sawdust, the effect of which is to
keep them separate, they are then immersed, or, as techni-
called termed, 'pickled,' in dilute sulphuric acid, and after
this placed in the annealing pot, and left in the furnace
about 24 hours; on coming out, the plates are passed
through the cold rolls. After passing the cold rolls, the
plates seem to have too much the character of steel, and are
not sufficiently ductile; to remedy this they are again
annealed at a low heat, washed in dilute sulphuric acid to
remove any scale of oxide of iron, and scoured with sand
and water. The plates in this state require to be perfectly
clean and bright, and may be left for months immersed in
pure water without rust or injury; but a few minutes ex-
posure to the air rusts them. With great care to have them
perfectly clean, they are taken to the stove, Fig. 270 being
a section through the line, K, K, of the the plan, Fig. 271.
FIG. 270.

K
UN
6
5
4
3
В

B
2
I
FIG. 271.
Taken from the right to the left, I represents the tinman's
pan; 2, the tin pot; 3, the washing or dipping pot; 4, the
grease pot; 5, the cold pot; 6, the list pot. The tinman's
pan is full of melted grease; in this the plates are immersed,
and left there until all aqueous moisture upon them is
evaporated, and they are completely covered with the grease;
from this they are taken to the tin pot, and there plunged
into a bath of melted tin, which is covered with grease; but
K
TINNING OF IRON PLATES.
873
as in this first dipping the alloy is imperfect, and the sur-
face not uniformly covered, the plates are removed to the
dipping or wash pot; this contains a bath of melted tin
covered with grease, and is divided into two compartments.
In the larger compartment the plates are plunged, and left
sufficiently long to make the alloy complete, and to separate
any superfluous tin which may have adhered to the surface;
the workman takes the plate and places it on the table
marked в in the plan, and wipes it on both sides with a brush
of hemp; then to take away the marks of the brush and give
a polish to the surface, he dips it in the second compartment
of the wash pot. This last always contains the purest tin,
and as it becomes alloyed with the iron it is removed on to the
first compartment, and afterwards to the tin pot. The plate
is now removed to the grease pot (No. 4); this is filled with
melted grease, and requires very skilful management as regards
the temperature it is to be kept at. The true object is to
allow any superfluous tin to run off, and to prevent the alloy on
the surface of the iron plate from cooling more quickly than
the iron. If this were neglected the face of the plate would be
cracked. The plate is removed to the cold pot (No. 5);
this is filled with tallow, heated to a comparatively low
temperature. The use of the grease pots, Nos. 4 and 5, is
the process adopted in practice for annealing the alloyed
plates. The list pot (No. 6) is used for the purpose of re-
moving a small wire of tin which adheres to the lower edge
of the plate in all the foregoing processes. It is a small
cast-iron bath, kept at a sufficiently high temperature, and
covered with tin about of an inch deep. In this the edges
of the plates are dipped and left until the wire of tin is
melted, and then detached by a quick blow on the plate with
a stick. The plates are now carefully cleaned with bran to
free them from grease. Lastly they are taken to the sorting
room, where every plate is separately examined and classed,
and packed in boxes for the market.
The tests of quality for tin plates are ductility, strength,
and colour. To obtain these, the iron must be of the best
quality, and the manufacture must be conducted with pro-
portionate skill.”
VOL. II.
3 L
874
IRON.
The plates to be tinned are sometimes made of puddled
steel; 100 lbs. of plate require from 6'5 to 8 lbs. of tin; iron
may be tinned also in the wet way.
The plates are coated with zinc* in the same manner as
cast-iron; they are sometimes also enamelled.
Wire Drawing.
Strong fine-grained iron produced in the finery process is
best adapted for the manufacture of wire, but since the im-
provements in the puddling process puddled iron is likewise
employed. The iron is either rolled to the required dimen-
sion (about two lines in diameter), or it is cut by a slitting
mill (page 826), and wound up on a windlass whilst still hot.
In order to keep the wire rod as soft as possible, it is slowly
cooled in a closed box made of iron plates. The wire rod is
then immersed in dilute sulphuric acid, and afterwards washed
in lime-water and dried in the air; it is then placed on the
reel, B, Figs. 272 and 273. One end of the wire rod is
FIG. 272.

B
D
A
pointed either with the file or the hammer, so as to make it
fit the largest hole of the draw-plate, F, which is movable in
the pillar, D, and fixed to the cylinder, c, by means of the
hook, u. The cylinder, c, is then set in motion by being
lifted somewhat, and the cross-bar, i, connected with the
cross-bar, n, by means of the hook, o, when the spindle, 7,
DINGL., Bd. 109, pp. 478, 480; Bd. 112, p. 121.
WIRE DRAWING.
875
resting upon the support, d, will be put in motion by means
of the bevelled wheels, b and c, and of the movable axis, a,
below the draw-bench, A. When the wire has passed through
one hole and been wound upon the cylinder, the hook, o, will
FIG. 273.

Ֆ
и
URDU IDE
WHEN
C
d
no longer be kept in contact with the cross-bar, n, by strained
wire, and consequently drops down, when the ring of wire
is removed and drawn through the next hole. During the
operation of drawing the wire must be kept in a horizontal
position by means of the movable draw-plate.
The draw plates are frequently filled with hard white
cellular pig-iron, and the holes are made large on the
side where the wire enters, and taper off regularly to
the other side. Plates with jewelled holes are seldom
employed. When drawing the wire a lump of tallow
is placed before the draw-plate; soap is also used for the
same purpose of diminishing friction. The wire offers some
resistance on being drawn again through the same hole.
The operation of drawing renders the wire hard, and there-
fore it must be annealed. The wire is placed in a closed
cylindrical ring of cast-iron, through which flame is con-
ducted. It is then cleaned from the superficial oxide either
876
IRON.
by immersion in dilute sulphuric acid or by scouring me-
chanically; for instance, by fixing the wire-rings on the arm
of a lever, which is jerked against an anvil, whilst a continual
stream of water impinges upon the wire. The wire is some-
times cleaned by being placed together with pebbles in a
perforated rotating cask with an admission of flowing water.
When the production of bright wire is desired, the wire
on drawing is conducted through a mixture of water, yeast,
sulphuric acid, and copper vitriol; a little olive oil is kept on
the surface of the admixture as well as on the draw plate.
Iron coated with copper passes more readily through the
draw plate, and is also protected from rusting for some time.
When wire is diminished one-half, one-third, one-fourth,
&c., in diameter, it is augmented in length respectively four,
nine, sixteen times, &c.
The speed at which wire may with prudence be drawn out
depends on the ductility and tenacity of the metal. Iron
wire of 0.3 inch in diameter bears drawing at the rate of
from 12 to 15 inches per second; but when of o'025 of an
inch, at the rate of 40 to 45 inches at the same time.
The ultimate strength of iron and steel wire almost always
decreases as the diameter increases; this is also the case
with forged and rolled bars, in which the metals are united
in greater bulk. Some very small kinds of charcoal wire
only break with loads of about 100 tons to the square inch,
while the average strength of wire may be taken as double
that of rolled bars.
INDICES.
!
1
CLASSIFIED INDEX
OF
SMELTING WORKS AND MINING DISTRICTS.
A FRICA (Copper), 6, 217, 222.
Cape of Good Hope (Copper), 5.
Algeria (Copper), 6
Agordo, Venetian Alps (Copper), 13,
22, 103, 222, 246, 257, 259
America (Iron), 685
North (Copper), 4, 238; (Iron), 337,
417
United States (Iron), 265, 266,
329, 330, 332, 802
574
Baltimore (Copper), 227
Benton County (Iron),
Boston (Copper), 8, 11,
14, 42, 48, 55, 127, 215
227
Cleaveland (Copper) 227
Connecticut (Copper), 5
Detroit (Copper), 14, 226,
Illinois (Iron), 340
Kentucky (Iron), 340
Lake Superior (Copper),
I, 3, 4, 6, 80, 226
Maryland (Copper), 5;
(Iron), 340.
Minnesota (Copper), 3
New England (Copper), 4
New Haven (Copper), 227
New Jersey (Copper), 4,
6; (Iron), 271, 328, 700
New York (Copper), 5
North Carolina (Copper),
4; (Iron), 325, 340, 344
(Iron), 340
Ohio (Copper), 227;
Pennsylvania (Copper),
4, 227; (Iron), 340, 423, 558
America, North, United States, Pitts-
burg (Copper), 227
Vermont (Iron), 700
-Virginia (Copper), 4, 5;
(Iron), 340
Mexico (Copper), 7
South (Copper), 6, 7, 227, 229
Bolivia (Copper), 7
Chili (Copper), 3, 5, 6, 7, 167,
186, 187, 189
New Granada (Iron), 337
Peru (Copper), 3, 5, 6, 187
Cordillera Mountains (Cop-
pcr), 6
Viquintipa (Copper), 36
Asia Minor (Copper), 217, 222
Australia (Copper), 186
Burraburra (Copper), 80
South (Copper), 5, 6, 55, 66
Kanmantoo (Copper), 7
New Zealand (Iron), 316, 327
Austria (Copper), 263; (Iron), 265, 266
Bohemia (Copper), 7; (Iron), 326,
334, 760, 764, 778
Kallich (Iron), 773
Neudeck (Iron), 372
Ransko (Iron), 420
Carinthia (Iron), 332, 336, 471,
481, 702, 715, 724, 766, 768, 778,
802
Buchscheiden (Iron), 776
Feistritz (Iron), 763
Frantschach (Iron), 770
853
Lavanthal (Iron), 561
Lölling (Iron), 561
Lippitzbach (Iron), 775, 776,
iv
INDEX.
Austria, Carinthia, Nothburgahütte
near Freudenberg (Iron), 852
Wolfsberg (Iron), 773
Carniola, Toplice (Copper), 239
Galicia (Copper), 4; (Iron), 402
Hungary (Copper), 1, 4, 5, 7, 27,
44, 67, 70, 76; (Iron), 315, 764
Altgebirg (Copper), 101
Altwasser (Copper), 14
Dernoe (Iron), 868
Felsöbanya (Copper), 117, 119
Fernezely (Copper), 117
Kapnick (Copper), 117
Kremnitz (Copper), 14, 213
Laposbanya (Copper), 117
Nagybanya (Copper), 13, 117
Neusohl (Copper), 14, 213
Olahlaposbanya (Copper), 117
Schemnitz (Copper), 14, 213, 235
Schmöllnitz (Copper), 14, 15,
48, 51, 54, 71, 101, 211, 231, 232,
234, 248, 250
Sztrimbul (Copper), 117
Tajova (Copper), 14, 68, 213
Zsarnowitz (Copper), 213
Moravia, Blensko (Iron), 591, 764
Gaja (İron), 556
Wengerska Gorka (Iron), 556
Silesia, Wengers-Gorka (Iron), 482
Styria (Iron), 326, 332, 336, 412,
Belgium, Espérance (Iron), 482
Liege (Iron), 633
Ougrée (Iron), 406
Séraing (Iron), 371, 393, 482, 505,
574
FRANCE (Copper), 3, 11; (Iron),
265, 266, 229, 334, 339, 372, 411,
422, 456, 478, 557, 685, 761, 768,
779
South (Copper), 8; (Iron), 805
Auvergne (Iron), 328
Bley (départ. Saône) (Iron), 391
Bouc (Copper), 41, 222
Chanon St. Etienne (Iron), 393
Champigneulles near Nancy (Iron),
556
Chessy near Lyon (Copper), 7, 14,
221
Creuzot (Iron), 279, 311, 557, 623,
809
Decazeville (Iron), 557
Fourchambault (Iron), 557
Marseilles (Copper), 233
Mauberge (Iron), 557
Pisos (Iron), 738
Rhone (Copper), 14, 49, 217
Savoy (St. Helen) (Iron), 391
Vierzon (Iron) 859, 862
ERMANY (Iron), 266, 332, 335,
471, 481, 491, 541, 571, 709, 762, GE
767, 768, 778
Eisenerz (Iron), 277, 470, 548,
561
Hieflau (Iron), 556, 561
Hüttenberg (Iron), 561
Leoben (Iron), 773, 778, 850
Lietzen (Iron), 391
Mariazell (Iron), 372, 555, 573,
575, 610, 616, 622, 633, 724
Miesling (Iron), 555
Neuberg (Iron), 561, 773, 774
Oeblarn (Copper), 13, 44, 124
Rottenmann (Iron), 773
St. Stephan (Iron), 470, 482,
555, 585
Turrach (Iron), 561
Vordernberg (Iron), 336, 368,
404, 561, 567
Transylvania (Copper), 4
Tyrol (Copper), 5
Jenbach (Iron), 561
BANAT, Csiklowa (Copper) 101,
131
Dognorka (Copper), 101
Moldava (Copper), 101
Steierdorf (Iron), 342
Szaszka (Copper), 100
Belgium (Iron), 263, 265, 266, 354,
456, 478, 483, 510, 555, 557, 563,
571, 849
403, 411, 456, 759, 761
Baden (Iron), 334, 354
Bavaria (Iron), 334, 768, 774
Ebenau (Iron), 773, 776
Hammern (Iron), 773
Hof (Iron), 557
Maximilianshütte (Iron), 772,
773, 789, 851
Brunswick (Zorge Rübeland) (Iron),
556,774, 853
Prussia (Copper), 263; (Iron), 265,
418
Berlin (Iron), 582, 593, 620,
623
Hartz (Copper), 6, 67; (Iron),
326, 334, 337, 358, 363, 391, 710,
777, 783
Lower (Copper), 15, 26, 33,
43, 51, 64, 66, 73
Gittelde (Iron), 277, 388,
556, 561, 738
776
130, 235
Ilsenburg (Iron), 555,
Oker (Copper), 33, 124,
Rammelsberg (Copper),
7, 14, 124, 231, 250, 333
Upper (Copper), 13, 33, 108,
130, 131; (Iron), 291, 296, 412,
449, 473, 555, 566, 570, 572, 709,
741, 745, 746, 758, 776
INDEX.
V
Germany, Prussia, Upper Hartz,
Altenau (Copper), 33, 51, 52, 55,
62, 84, 130; (Iron), 324, 360
62, 65, 130
Andreasberg (Copper),
Lautenthal (Copper), 33,
51, 62, 130
Lerbach (Iron), 297, 368,
592, 604, 610
Königshütte (Iron), 281,
283. 318, 319, 360, 535, 548, 588,
609, 850, 866
482
Rothehütte (Iron), 283,
Hanover (Iron), 768
Georg Marienhütte (Iron),
Alexishütte, near Lingen
557
(Iron), 335
Meppen (Iron), 398, 557
Neustadt (Iron), 473, 556
Silbernaal, near Clausthal
(Iron), 737
Sollingerhütte (Iron), 738
Hesse (Iron), 334
Ludwigshütte (Iron), 555
Thalitter (Copper), 116
Mansfeld (Copper), 1, 2, 13, 15,
33, 34, 43, 44, 45, 48, 51, 55, 58,
65, 66, 67, 76, 191, 192, 193, 230,
235
Müsen, Lohe (Copper), 13, 61;
(Iron), 119, 270, 319, 336, 567
Siegen (Copper),189,232; (Iron),
283, 320, 332, 370, 391, 404, 412,
449, 556, 709, 710, 777, 868
Nassau (Iron), 710, 768,
Dillenburg (Copper), 4, 13,
66, 77, 110, 136
Isabellenhütte (Copper), II
Polonia (Iron), 555, 709, 773
Pomerania (Iron), 738
Provinces of the Rhine (Copper),
187; (Iron), 334, 414, 491, 762,
768
Aggerthal (Copper),
189
Commern (Copper),
234, 242
Duisburg (Copper),
13, 189, 204, 208;
557
(Iron), 429,
Linz (Copper), 15,
232, 233, 234, 236, 252, 254
Germany, Prussia, Waldeck, Twiste
(Copper), 230, 234
Silesia (Iron),326, 334, 339, 368,
376, 429, 491, 501, 555, 556, 557,
558, 571, 572, 709, 710, 759, 760,
764, 772
860
Álvenslebenhütte (Iron), 787,
Breslau (Iron), 590
Gablau (Iron), 342
Gleiwitz (Iron), 368, 393, 482,
557, 582, 589, 590, 594, 610, 624,
726, 861
Königshütte (Iron), 346, 368,
393, 401, 406, 408, 447, 463, 482,
495, 556, 726
109
Laurahütte (Iron), 557
Kupferberg (Copper), 13, 54,
Malapane (Iron), 482, 556
Rybnik (Iron), 709
Vorwärtshütte (Iron), 297, 556
Tarnowitz (Iron), 557
Westphalia (Iron), 339, 342, 360,
361, 363, 388, 399, 414, 478, 480,
491, 563, 571, 778
Gravenhorst (Iron), 555
Hasslinghausen (Iron), 283,
512, 557
-Hattingen (Iron), 393, 400, 557
Hörde (Iron), 376, 393, 400,
451, 505, 557, 770, 835
Lünen (Iron), 482
Porta Westphalica (Iron),
400, 557
Stadtbergen (Copper), 15, 58,
233, 236, 252, 254
Saxony (Copper), 67, 263; (Iron),
326, 334
Antonshütte (Copper), 44
Bergieshübel (Iron), 482
Erlahammer (Iron), 709
Freiberg (Copper), 13, 14, 43,
45, 48, 49, 51, 55, 58, 136, 214,
234, 259
Grünthal (Copper), 14, 73, 229,
Königin Marienhütte, near
Zwickau (Iron), 482, 572
Schwarzenberg (Iron), 325
Schleiz (Iron), 482
Thuringia (Iron), 338, 481, 709,
710
Eppichnellen (Copper), 239
Ilmenau (Iron), 2
Katzhütte (Iron), 482, 561, 610
Schmalkalden (Iron), 332
Wurtemberg (Iron), 334, 354, 768
Itzelberg (Iron) 773
Ruhrort (Iron), 850
Saarbrücken (Iron),
339
Siegburg (Iron), 451
Stolberg (Iron), 354
Sterne (Copper), 107
-St. Josephsberg
(Copper), 107
773
VOL. II.
3 M
Königsbronn (Iron), 482
Unterlochen (Iron), 774
Wasseralfingen (Iron), 594, 712,
vi
INDEX.
Great Britain (Iron), 263, 265, 339,
403
England (Copper), 3, 77, 132,
134, 136, 192, 238; (Iron), 266,
267, 326, 334, 355, 414, 429, 455,
473, 478, 480, 496, 555, 557, 563,
571, 670, 868
Cornwall (Copper), 4, 5, 6,
8, 82, 233; (Iron), 326, 328
Perran (Iron), 337
Cumberland (Copper), 9;
(Iron), 295, 328, 329, 330, 333,
403
613
Alston Moor (Copper), 9
Keswick (Copper), 10
Derbyshire (Iron), 331, 428,
Devonshire (Copper), 6;
(Iron), 328
849
332
Exmoor (Iron), 337
Durham (Iron), 539, 570,
Ferryhill (Iron), 404
Weardale (Iron), 337
Forest of Dean (Iron), 331,
Lancashire (Copper), 9, 143;
(Iron), 279, 295, 339, 428, 706
Backbarrow (Iron), 411
Newland (Iron), 411
Uverstone (Iron), 330,
441, 483, 557,
767
Whitehaven (Iron), 330
London, Rotherhithe (Iron),
Woolwich (Iron), 590, 669
Northamptonshire (Iron), 339
Northumberland (Iron), 339
Perthshire (Iron), 328
Shropshire (Iron), 574
Somersetshire (Iron), 331
Brendon Hill (Iron), 337
Staffordshire (Iron), 267,
362, 393, 403, 421, 471, 483, 848
Dudley (Iron), 301, 723
North (Iron), 341
South (Iron), 339, 283
Stourbridge (Iron), 723
Westmoreland (Copper), 9
Worcestershire (Iron), 267
420
Yorkshire (Iron), 403
Bowling (Iron), 509
Cleveland District (Iron),
295, 343, 366, 422, 431, 471, 483,
491, 514, 849
Low Moor (Iron), 509,
557, 771, 805, 858, 859, 861
Middlesbro' (Iron), 471
North (Iron) 339
Ireland (Copper), 6, 9; (Iron),
417, 556
Great Britain, Ireland, Wicklow
(Copper), 7, 14, 27, 232
14
Isle of Anglesea (Copper), 7, 9,
Alderley Edge (Cop-
per), 15, 234, 245, 249, 252, 255
Amlwck (Copper), 252
Holyhead (Copper), 9
Isle of Man (Copper), 9
Scotland (Copper), 9, (Iron),
267, 283, 297, 341, 346, 362, 388,
400, 402, 403, 417, 420, 422, 423,
471, 478, 483, 514, 571
Clyde (Iron), 574
Dundyvan (Iron), 558
Gartsherrie (Iron), 279, 283,
394, 542, 558
Govan (Iron), 394, 558
Lanarkshire (Iron), 341
Calder (Iron), 513, 558
Mockland (Iron), 558
Wales (Copper), 51, 136, 156,
165, 170; (Iron), 267, 328, 333,
403, 422, 423, 470, 471, 473, 483,
563
331
North (Copper), 9; (Iron),
South (Copper), 8, 13, 143,
157, 177; (Iron), 339, 341, 362,
491
Aberdare (Iron), 557
Abersychen (Iron), 536
Blaenavon (Iron), 509
Carmarthenshire (Penbry
Works) (Copper), 140
Dowlais (Iron), 366, 418,
421, 480, 484, 557, 771, 817, 830,
862, 868
718
145
Ebbw Vale (Iron), 347, 450,
Pontypool (Iron), 509, 709
Swansea (Copper), 143,
Yniscedwin (Iron), 423, 558
Ystalifera (Iron), 557, 558,
709, 772
Italy (Iron), 702
Arvellino (Naples) (Iron), 327
Corsica (Copper), 217; (Iron), 702
Elba (Iron), 326, 328
Etna (Iron), 328
Lipari Islands (Iron), 328
Lombardy (Iron), 561
- Traversella (Piedmont) (Iron), 325,
353
Tuscany (Copper), 5, 6, 187;
(Iron), 471
Capannevechia (Copper), 238,
262
Vesuvius (Iron), 328
India (Copper), 7, 13, 64, 116
East (Iron), 702
INDEX.
vii
APAN (Copper), 13, 78, 116
JAPAN
APLAND (Iron), 326, 328
Gilsaa (Copper), 33
LA
N
ORWAY (Copper), 4, 77, 187,
189; (Iron), 266, 326, 335, 375,
555, 763, 764
Egersund (Iron), 327
Foldal (Copper), 246, 259
Hassel (Iron), 353
Kaafjord (Copper), 13
Kongsberg (Copper), 57, 61
Röraas (Copper), 13, 98, 230
PACIFIC Islands (Copper), 227
RUSSIA (Copper), 80; (Iron), 265,
326, 332, 555, 556
Crimea (Kertch) (Iron), 308
Finland (Iron), 309, 335, 375, 555,
702
Nischnetagilsk (Copper), 116
Perm (Copper), 14, 45, 54, 57, 69,
75, 222
Siberia (Copper), 4, 219; (Iron),
326
Soumboul (Iron), 419
St. Anna (Iron), 309
Tourinski (Copper), 4
Ural (Copper), 6, 7, 76; (Iron),
328, 375, 463
Russia, Ural, Nischnetagilsk (Iron),
474
SMALAND, Ekersholm (Iron), 391
Taberge (Iron), 353
Spain (Copper), 6, 217, 222, 263;
(Iron), 266, 326, 424
Basque Province (Iron), 337
Biscay (Iron), 332, 704
Huelva (Copper), 11
Pyrenees (Iron), 332, 337
Riotinto (Copper), 11, 14, 15, 80,
107, 231, 232, 249, 256
Sweden (Copper), 4, 39, 48, 49, 59, 65,
98, 263; (Iron), 265, 266, 324, 326,
328, 335, 358, 375, 444, 456, 470,
555, 633, 749, 759, 760, 763, 764,
767, 768, 774, 778
Atvidaberg (Copper), 13, 53, 54,
60, 84
Boo (Iron), 736
Dannemora (Iron), 315, 377, 391,
509
Fahlun (Copper), 4, 13, 43, 45, 54,
95, 119, 230, 235
Garpenberg (Copper), 97
Näfvequard (Copper), 97
Nyby (Iron), 851
Skebo (Iron), 862
Switzerland (Iron), 266, 334, 761
Underwiller (Iron), 417, 556, 736
INDEX.
ADAPTABILITY of Iron Ores to
the smelting process, 347
pig-iron for castings, 578
Advantages of hot blast, 506
Aitites, 334
Alkali metals in iron, 319
Aluminium in pig-iron, 318
steel, 317, 318
wrought-iron, 318, 692
Aluminous yellow iron ore, 334
Amlwch mine near Holyhead, 9
Analyses of blackband, 342
black copper, 55, 186, 200
blue metal, 180
brown iron ore, 338
calcined copper slate, 195
flowery and cellular white pig-
iron, 277
Cleveland iron ores, 343
concentrated matt, 181
copper bottoms, 181
mica, 62
slags, 33
blast furnace gases, 197
ferriferous bears, 45
grey pig-iron, 283
hæmatite, 330
hearth ends (copper), 115
mottled pig-iron, 283
malleable cast-iron, 70
Mansfeld copper slags, 34
magnetic ore from Cornwall,
326
radiated pig-iron, 277
raw matt (copper), 43
slags (copper), 113, 166
red metal, 180
refined copper, 77
refinery slags (copper), 182
rosette copper, 201
Analyses of slags from charcoal blast
furnaces, 389
coke blast furnaces, 392
resulting at the produc-
duction of white concentrated
matt, 179
some varieties of sand, 627
sparry iron ore, 337
spathic ore, 270
speiss, 44
spiegeleisen, 270
tough copper, 211
white metal, 179, 181
pig-iron produced by the
regular process, 279
Anthracite, 423
blast furnaces, 423
Antimony in wrought-iron, 315, 692
Apparatus for collecting the waste
gases, 447
re-heating the balls or blooms,
747
used in the finery process, 740
Appearance of molten pig-iron, 290,
549
Application of blast furnace slags,
572
grey pig-iron, 294
Argillaceous carbonate of iron, 338
Arseniate of copper, 7
Arsenic in pig-iron, 314
steel, 315
wrought-iron, 315, 692
Atacamite, 7
Austrian slag process, 765
Azurite, 2, 7
BARILLA copper sand, 189
slags, 51
BECCHI and HAUPT's process, 262
INDEX.
ix
Behaviour of white iron at a gradually
rising temperature, 689
BELL'S roasting process, 147
Berlin sands, 625
Bessemer process, 706
Best selected copper, 173, 184
shot copper, 182
BISHOFF'S method, 262
Bismuth in wrought-iron, 315
Blackband iron stone, 341
Black copper, 43, 55, 92
smelting, 91, 99, 106, 120, 123,
182, 183, 184, 185, 189, 212
Blacking, 629
Black oxide of copper, 6
Blast heating apparatus, 506
furnaces, 432
construction of, 433
process, conduct of, 548
furnace slags, the
slags, the nature
565
pipes, 485
regulators, 505
Calcining raw sulphuretted copper
ores, 135
Calcium in pig-iron, 318
steel, 318
wrought-iron, 318, 692
Cannon bronze, 219
Carbonate of iron, 336
Carinthian process, 766
CARLBERG's furnaces, 87
Casting, 625
in metallic moulds, 667
Castings, 578
Cast-iron, 264
Catalan forge, 700
or French process, 697
Cellular white pig-iron, 276, 561
Cementation, 244
Chamoisite, 344
Charcoal, 411
of,
required for cupola furnaces, 603
blast furnaces, 481
Blauöfen, 444
Bloomeries, 697
Blötmalm, 84
Blowing in, 534
machines, 491
for cupola furnaces, 601
out, 547
Blue copper matt, 51
metal, 133
vitriol, 7
Blumige floss, 276
BOCCARD'S furnace, 599
Bog iron ore, 334
Bohemian finery process, 759
Boiling process, 832
BORNEMANN'S graphic table, 500
Boshes, 474
Bournonite, 6
Boxes, 635
Braten, 206
Breaking up the iron ores, 382
process, 758
BREDBERG'S furnaces, 86, 91
Bronze matt, 178
Bronzing castings, 672
Brown coal, 772, 778
gases, 776
iron ores, 331
BROWN'S patent squeezer, 855
Buntkupfererz, 5
Burraburra mine, 6
Burnt iron, 696
CALCINING copper ores, 183, 184,
pyrites, 178
raw matt, 183, 184, 185, 186
gases, 779
cupola furnaces, 592
Charging ore and fuel, 537
Chemical reactions of the roasting
process (copper), 138
61
refining process (copper),
Chief operations of the hydrometal-
lurgical extraction of copper, 231
products of the copper smelting
process, 165
•
Chilled white iron, 281
Chills, 667
Chloride of copper, 7
Chlorination in the dry way, 238
wet way, 237
dry and wet way, 241
without employing dilute muriatic
acid, 236
Chrome iron ore, 327
Chromium in pig-iron, 316
steel, 317
Classification of the processes for ex-
tracting copper, 12
pig-iron, 295
wrought-iron, 694
Clay, 628
iron stone, 338
Cleveland iron ores, 343
Coal, 772, 777
brass, 341
dust, 629
of Mynydd Newydd, 136
Pentrifelin, 136
Tyrcenol, 136
Coarse metal, 15
Coating iron with a glass-like mass,
684
Cobalt in pig- and wrought-iron, 318,
692
Coke, 412
cupola furnaces, 593
Cold-short iron, 695
X
INDEX.
Combination of lump fining and the
breaking-up process, 758
Combined reverberatory and finery
process, 707
Common white metal, 133
Comparison of the English rever-
beratory process with the Mans-
feld cupola furnace process, 192
Complication of the copper refining
process, 62
Composition of concentrated matt, 51
furnace gases, 114
copper refining slags, 182
rosette copper, 66
Compound or universal rolling mill,
824
Concentrated matt, 120
Concentration of the matt, 119
desilverised copper matt, 123
raw matt in reverberatory
furnaces, 193
smelting, 212
of the roasted raw matt, 50
Cone and cup charger, 450
Construction of the front part of
cupola furnaces, 587
Consumption of charcoal in iron blast
furnaces, 412
coal in iron blast furnaces, 422
coke in iron blast furnaces,
414
iron in different countries, 266
Contra lodes, 8
Copper, I
at tough pitch, 56
Barilla, 3
bottoms, 174, 181
containing aluminium, 82
antimony, 82
arsenic, 82
bismuth, 82
carbon, 80
iron, 82
lead, 81
nickel, 82
phosphorus, 83
silicon, 82
sodium, 83
suboxide of copper, 80
sulphur, 82
glance, 2, 5
in pig-iron, 314
steel, 314
wrought-iron, 314, 693
lead matt, 129, 218
mica, 62
ores, I
antimonious and arsenious, 5
at Altgebirg, 101
oxidised, 6
pyrites, 2
refining process, 61
Copper salts, 6
sand, 3
scale, 83
slate, 193
smelting furnaces, 161
process, 163
smoke, 143
Coppering of cast-iron, 673
Cores, 630
Corrocorro copper ore, 3, 229
Cornish mines in 1799, IO
COWPER's hot blast apparatus, 516
Cross flookans, 8
Crucible of blast furnaces, 478
Crude lead, 120
Cupola furnaces, 586, 588
process, 604
Cupriferous raw iron, 223
Cuprobarillas, 229
Cyanide of potassium, 574
Cyano-nitride of titanium, 317, 575
DAM
AM, the, of blast furnace, 479
Dark grey pig-iron, 287
Deposits in blast furnaces, 573
Desilverisation of copper matt, 123
Dichtpolen, 210
Different methods of smelting copper
in South Wales, 177
Dioptase, 7
Double puddling furnaces, 802
Dressing of the iron ores, 352
Drontheim copper, 100
Dry copper, 80
sand, 628
puddling, 830
Drying chambers, 634
and warming blast furnaces, 532
Dünstein, 55, 93, 99, 106
EIFE
IFEL Walloon process, 762
Enamelling castings, 675
Enargite, 6
Endless chain system, 426
English copper smelting process,
133
veins, 8
Walloon process, 763
Examination of a blast furnace estab-
lishment, 576
Examples of the cupola furnace pro-
cess, б09
projection of a coke blast
furnace according to LINDAUER'S
formula, 488
production of grey and
mottled pig-iron, 555
Extraction of cement copper from
native solution, 250
copper at the desilverisation of
black copper, 124
INDEX.
xi
Extraction of copper at the desilver-
isation of matt, 124
ores and matt, 117
by combined processes in
cupola and reverberatory fur-
naces, 192
from oxidised ores, 250
sulphuretted ores and
products, 256
FAR
in the wet way, 229
ALLOW ores (Fahlerz), 2, 95, 212
Fan, 601
Faulbrüchig, 696
Feathered shot copper, 182
Feeding head, 633
of the blast furnaces through the
tuyeres, 546
Ferruginous bears, 44, 90, 93: 574
Fibrous wrought-iron, 695
Fine-grained wrought-iron, 694
Fine metal, 133
Finery hearth, 741
processes, 754
Finishing wrought-iron, 869
First refining of black copper, 118
Flame from blast furnaces, 569
Flookans of the lode, 8
Flowery white pig-iron, 276, 561, 711
Fluctuations of the temperature in
blast furnaces, the reason of, 530
Fluxes for the iron ore mixtures, 405
used in the finery process, 736
puddling process, 780
re-heating iron, 781
re-melting cast-iron, 584
refining of pig-iron, 779
Forge hammers, 752
Formulæ for calculating the quantity
of blast, 497
Founding, 578
Franklinite, 327
French finery process, 760
Friction hammer, 815
Fuel for the iron blast furnace pro-
cess, 408
used in cupola furnaces for re-
melting pig-iron, 585
the finery process, 735
puddling, 771
Furnace belly, 464
Fusibility of silicates, 387
Fusion point of iron, 527
GAARSCHMELZIG,
AARSCHMELZIG, 711
Galleries, 9
Gases produced with coal, 777
turf, 775, 779
wood, 775, 779
Gas reverberatory furnaces, 623
Gulf ores, IOI, 212
General rules for the formation of ore
mixtures in copper smelting, 35
German or breaking-up process, 756
process in bloomeries, 697
GERSTENHOFFER'S furnace, 149
process, 148
Grelles Roheisen, 280
Green sand, 625
Grey pig iron, 282, 551, 713
Grossluckige flossen, 711
Growan, 8
GURLT'S method of separating sul-
phur, 144
HEMATITE, 328
Half walloon process, 760
Hammergaar, 56
Hammers, 813
Hardmalm, Sa
Hard matt, 172
ores, 84, 95
Hartzerrennen, 724
Hearth, 476
Height of blast furnaces, 470
cupola furnaces, 586
Hollow fires, 748
Hot blast apparatus for blast fur-
naces, 506
cupola furnaces, 599
Hydrates of oxide of iron, 331
Hydraulic hammers, 815
Hydrometallurgical extraction of cop-
per, 231
ILLUSTRATIONS
LLUSTRATIONS of the combined
processes in cupola and reverbera-
tory furnaces, 193
German copper smelting
process, 84
reverberatory process, 177
smelting argentiferous and
auriferous copper ores by the
German method, 117
Improvements in moulding, 668
Inclined planes, 425
Inducements for re-melting pig-iron,
579
Influence of impurities in copper upon
the electric conducting power, 78
moisture on blast, 495
the price of materials in copper
smelting, II
quality of the ore in copper
smelting, II
time used in the production
of copper, 12
Interior form of blast furnaces, 452
for easily fusible ores,
453
fuse, 454
ores difficult to
cupola furnaces, 586
Intermittent blast, 487
Iron, 264
xii
INDEX.
Iron blast furnace process, 518
deposits, 223
garnet, 344
minium, 329
ores, 323
adaptability of to the smelting
process, 347
Iridium in steel, 317
ACOB'S ladder, 426
JA
Japanese copper, 182
KARTITSCHSCHMIEDE, 766
Kernel roasting, 22, 103
Kleinluckige Flossen, 711
Kidney ore, 329
Kiehnstöcke, 130
Killas, 8
Königskupfer, 127
Kupferblau, 7
Kupfergrün, 7
Kupferindig, 5
Kupferschiefer, 193
LAKE ores, 334
Lancashire forge, 747
process, 763
LANGEN'S apparatus, 451
Laugerze, 253
Lead, influence of upon iron, 316
matt smelting, 123
Length of operations of iron blast
furnaces, 548
Lift hammers, 750
Lifts, 424
hydraulic, 429
pneumatic, 429
Manipulations in the blast furnace.
process, 532
hearth of blast furnaces, 541
Manometers, 492
Manufacture of tin plates, 871
Materials required to produce 1 cwt.
of refined copper, 95
used in moulding, 625
Matt, 55
the finery process, 711
smelting, 188
Mechanical treatment of castings, 670
Melting temperature of pig-iron, 290
Metallic moulds, 629
Methods of moulding, 635
the finery process, 754
Mine of Cross-gill-burn near Alston
Moor, 9
Ecton in Staffordshire, 9
Mineral coal, 420
Mixing the iron ores, 383
with regard to the quality
of the fuel employed, 396
slags, 385
formation of
Modifications of the copper refining
process, 63
MOLIN'S modifications of the Fahlun
process, 96
Molybdenum in pig-iron, 316
Mottled pig-iron, 286
Moulding, 578, 625
in baked or used sand, 639
green sand, 635
loam, 660
Mounds for roasting copper ores, 85
vertical by means of steam power, NATIVE copper, 3
424
Limonite, 334
LINDAUER'S formula, 466
Lixiviation of copper salts, 253
Lower hearth, 478, 529
Luckige flossen, 276
Lump fining, 758
MACHINE moulding, 668
M Magnesium in pig-iron, 318
steel, 318
wrought-iron, 318
Magnetic iron ores, 324
Magnetite, 326
MAILLARD'S cupola furnace, 598
Malachite, 2, 6
Malleable iron, 264
Dr. MALLET's views on the molecular
constitution of cast-iron, 640
Manganese, influence of upon the
removal of phosphorus, 311
in pig-iron, 311
wrought-iron, 692
Newer methods for the produc-
tion of wrought-iron directly from
the ores, 703
Nickel in pig-iron, 318
steel, 317
wrought-iron, 318, 692
Nitrogen in pig-iron, 310
steel, 310
wrought-iron, 691
Noberge, 193
Nontronite, 344
LD copper, 219
Once melting-down process,754,
765
Operations in the English copper
smelting process, 133
Ore furnace slag, 165
mixture in iron blast furnaces with
regard to temperature and pres-
sure of the blast, 401
the quality
of the pig-iron, 403
roasting (copper), 121
INDEX.
xiii
Over poled, 176
Oxides of iron, 324
Oxidised copper ores, 220
Oxidising smelting of black copper, 57
PARKES'S method of roasting, 140,
178
furnaces, 140
PARRY'S method of refining pig-iron,
718
Patterns, 630
Periods of the German finery process,
756
Phosphate of copper, 7
Phosphorus in pig-iron, 306
steel, 310
wrought-iron, 310, 691
Pig-iron, 264, 571, 711
adaptability of for castings, 578
best adapted for castings, 583,615
chemical constitution of, 320
used in puddling, 769
wrought-iron and steel, limit
between, 322
Pits, 632
Plate casting, 668
Platinum in steel, 317
Plumbiferous black copper, 214
Pneumatic hammer, 815
lifts, 429
Position of the belly, 469
Precipitation of copper, 244
Preparation of the cast-iron in iron
blast furnaces, 714
Pressure of blast, 486
Process (copper) at Altgebirg, 101
Agordo, 103
Boston, 215
Chessy, 221
Duisburg, 189
Freiberg, 214
Lower Hungary, 213
Schmöllnitz, IOI, 211
Steinwerder, 187
Sterne, 107
Szaska, 100
Thalitter, 116
at the copper works on the Rhone,
217
(copper) in Siberia, 219
of treating sulphuretted copper ores
in reverberatory furnaces, 132
Production of blue metal, 169, 180
concentrated matt, 181
copper, 263
pig-iron in Great Britain, 276
the whole World, 265
pimple metal, 171
raw matt from roasted and raw
ores, 160
white concentrated matt, 179
metal, 170
VOL. II.
Production of white extra metal, 181
wrought-iron, 685
707
at a glowing heat, 705
from ores, 697
pig-iron, 704
in fineries, 707
-reverberatory furnaces,
Products of blast furnaces, 571
the cupola furnace process, 608
finery process, 767
Profit of copper works, 186
Properties of grey pig-iron, 287
wrought-iron, 686
Puddler's ore, 329
Puddling apparatus, 781
furnaces, 782
materials, 769
process, 829
for the production of fibrous iron,
833
fine-grained iron, 858
(roughing) rolls, 821
slags or cinders, 345
Purifying black copper, 118
in reverberatory furnaces, 67
Pyritic copper ores, 217
RACHETTE'S furnace, 219, 458
Raffinad, 211
Raising the pig-iron to a glowing heat,
715
Raw matt, 90, 99, 165
smelting of, 183, 184, 185
- short iron, 695
slags, 46, 90, 99, 165
smelting, 85, 99, 178, 183, 184, 185,
186, 188
of the copper schist, 195
Reaction of oxygen upon wrought-
iron, 690
Reactions of the components of the
copper ore mixture, 36
Red copper ore, 2, 6
hæmatite, 328
matt or metal, 133
ochre, 329
short iron, 695
Reducing smelting of black copper, 57
Refined copper, 77, 94
Refineries, 720
Refinery slags, 130
Refining black copper, 56, 94, 100, 120,
129, 174, 182, 183, 184, 185, 186,
189, 202
by poling, 76
in small hearths, 58
rosette copper, 118
pig-iron, 715, 770
Regulus, 165
in reverberatory furnaces, 726
Re-heating of blooms, 862
3 N
xiv
INDEX.
Re-heating cinders, 345
furnaces, 807
in hollow fires, 865
the puddled iron, 777
866
in reverberatory furnaces,
Relations between the different parts
of iron blast furnaces, 452
Re-melting of pig-iron, 580
614, 621
in crucibles, 581
cupola furnaces, 582
reverberatory furnaces,
Repairs of the hearth, 546
Requirements of a normal process,
31
Reversing rolls, 824
Reverberatory furnaces having a hearth
inclined towards the fire-bridge,
620
Rhonitz process, 760
Rising of the copper, 73
flue, 617
Roasting in furnaces or kilns, 363
mounds, 104, 363
open heaps, 358
furnaces for copper ores, 136
having the fire-place on one
side, 375
heated by flame, 375
with gas, 377
with a grate of conical shape,
372
an interior fire-place, 375
a plane grate, 367
step grates, 372
without a grate, 365
products, 28
smelting the blue metal for the
production of pimple metal, 186
of the pimple metal, 186
the black copper, 118, 201
concentrated matt, 52, 172
copper ores, 17, 85, 98
27
in cupola furnaces, 27
heaps, 24
mounds, 26
reverberatory furnaces,
copper schist, 194
iron ores, 355
raw matt, 46, 91, 168, 178, 212
iron cinders, 345
Rohschmelzig, 711
Rohgaar, 56
Rohgang, 280
Rolls or cylinders, 818
Rosette copper, 56, 65
Rotary squeezer, 818
Rother Glasskopf, 329
Rules for the construction of iron
blast furnaces, 464
AND of the London basin, 627
SAND
ores, 193
Schaaleneisen, 617
SCHAFHAUTL'S powder, 740, 780
SCHEERER'S classification of the
regular iron blast furnace pro-
cess, 551
Scheibenreissen, 65
Scheideerze, 253
Schleissen, 65
SCHWIND'S mechanical
contrivance
for calculating the blast, 498
Scraps of cast- and wrought-iron, 583
Second refining of copper on small
hearths, 71
Selecting process, 173
Segregations in the lower parts of the
blast furnaces, 573
Shape of blown-out blast furnaces, 475
Shears, 827
Shell casting, 669
Siegen finery process, 766
SIEMENS'S regenerative gas furnace,
789, 853
Silicate of copper, 7
Siliceous iron ores, 344
Silicon in pig-iron, 296
wrought-iron, 300, 305, 692
Silver copper glance, 5
in steel, 317
Silvering castings, 673
SINDING'S method of producing sul-
phuretted hydrogen gas, 260
Skumnas, 26, 85
Slags from charcoal blast furnaces,
572
coke blast furnaces, 572
the finery process, 737
puddling process. 862
refining process, 723
re-heating operation, 867
Slitting rollers, 826
Smelting cost of the copper process,
186
in crucible furnaces, 116
sump furnaces, 84
channel furnaces with two
open eyes, 107
mass before the tuyere, 568
modification in the case of ores
containing 9 per cent of copper,
185
of copper slate at Riechelsdorf and
Friedrichshütte, III
108
pyrites at the Upper Hartz,
lead ores, 122
native copper, 226
ores with 7 or 8 per cent of
copper, 184
185
9 per cent of copper,
INDEX.
XV
Smelting of partially oxidised cop-
per ores, 183
52
the Mansfeld copper schist, 193
roasted concentrated matt,
ores, 28
slags, 180
with pyritic ores for the
production of white and red matt,
171
processes according to PERCY,
&c., 183
as described by LE PLAY, 177
process at Atvidaberg, 85
Garpenberg and Näfve-
quarn, 97
Dillenburg, IIO
Kupferberg, 109
Röraas, 98
Soft ores, 84, 95
South Wales finery process, 764
Sparry iron ore, 336
Spathic carbonate of iron, 336
Specular iron ore, 328
Specific gravity of copper, 78
pig-iron, 289
Speiss, 44
wrought-iron, 688
SPENCE'S furnace, 153
- roasting process, 146, 155
Spiegeleisen, 268, 269, 559, 712
Spleissöfen, 57, 58
Spurstein, 200
Squeezers, 816
Staffordshire furnaces, 620
Stamp-hammer, 815
Steam-hammer, 813
Steel, 265
Texture and strength of grey pig-iron,
287
of wrought-iron, 686
Theory of the iron blast furnace pro-
cess, 518
Tile copper, 182
Tilt hammers, 752
Tin lodes, 8
in pig-iron, 315
wrought-iron, 315, 692
Tinning castings, 674
iron plates, 871
Titaniferous iron ore, 327
Titanium in pig-iron, 316
steel, 316
Tough copper, 77, 182
poling, 210
Transformation of copper in a soluble
state, 231
Treatment of copper lead matt at the
Upper Hartz, 130
the cupriferous intermediate pro-
ducts and residues, 129
desilverised residues for the pro-
duction of black copper, 200
poor copper ores at Perm, 212
the products of precipitation, 248
sulphuretted substances. in
cupola furnaces, 16
ores and substances in the
dry way, 16
Tromp, 699
TRURAN'S furnace, 460
Tungsten in pig-iron, 316
steel, 317
Turf, 419, 772, 778
coal, 417
Tuyeres, 485, 479
Stilpnosiderite, 331
Stirring copper with wooden poles,
207
Stopping the blast furnaces, 547
Strength of wrought-iron, 688
Stücköfen, 702
Styrian process, 765
Walloon process, 764
Sulphate of copper, 7
Sulphides of copper, 4
Sulphur in pig-iron, 301, 305
wrought-iron, 691
Sulu finery process, 760
Swabian finery process, 759
Swedish Walloon process, 762
Synopsis of processes for the extrac-
tion of copper, 10
TAPPING off the blast furnaces,
543
Tempering castings, 671
Temperature of different parts of blast
furnaces, 417
Testing of wrought-iron, 693
UN
NDER poled or dry, 176
Universal rolling mill, 824
VAN
ANADIUM in pig-iron, 316
Variegated copper ore, 2, 5
Various plans for refining pig-iron, 716
Veins of elvan, 8
WALLOON process, 761
416
Waste gases of blast furnaces,
composition of, 416
Water balance, 426
Weathering the roasted iron ores, 380
Wet puddling, 832
White cellular pig-iron, 712
extra metal, 134
pig-iron, 267, 559
of the irregular process, 280,
560, 712
regular process, 279,
562, 712
Width of the furnace mouth, 472
Wire drawing, 874
xvi
INDEX.
Wood, 419, 772, 778
Working conditions of iron blast
furnaces, indications of, 548
of copper mines at the Isle of
Anglesea, 9
Wrought-iron, 264
behaviour of at gradually rising
temperature, 689
containing phosphorus, 696
hammered when cold, 696
produced direct from the ores,
697
YELLOW earth, 344
ZER
iron ore, 333
ERRENNSCHMIEDE, 767
Zinc in pig-iron, 315
Zones of blast furnaces, 519
Zone of carbonisation, 523
combustion or oxidation, 526
preparatory heating, 519
reduction, 521
smelting, 524
Printed by WILLIAM CROOKES, at the CHEMICAL NEWS Office, Boy Court, Ludgate Hill, E.C.
"
--


UNIVERSITY OF MICHIGAN
3 9015 06712 4936
TN
€65
.K39
Kerl
E5
1868
V.2
✓
practical treatm.
ise on metallurgy

1