LJ THE MANUFACTURE OF IRON, IN ALL ITS VARIOUS BRANCHES. INCLUDING A DESCRIPTION OF WOOD-CUTTING, COAL-DIGGING, AND THE BURNING OF CHARCOAL AND COKE ; THE DIGGING AND ROASTING OF IRON ORE ; THE BUILDING AND MANAGEMENT OF BLAST FURNACES, WORKING BY CHARCOAL, COKE, OR ANTHRACITE; THE REFINING OF IRON, AND THE CONVERSION OF THE CRUDE INTO WROUGHT IRON BY CHARCOAL FORGES AND PUDDLING FURNACES. ALSO, A DESCRIPTION OP FORGE HAMMERS, ROLLING MILLS, BLAST MACHINES, HOT BLAST, ETC. ETC. TO WHICH IS ADDED, AN ESSAY ON THE MANUFACTURE OF STEEL. BY FREDERICK OVERMAN, MINING ENGINEER. Pboenixville Iron Works. WITH ONE HUNDRED AND FIFTY WOOD ENGRAVINGS. (Kbition. PHILADELPHIA: HENRY C. BAIRD, SUCCESSOR TO E. L. CAREY, S. E. CORNER MARKET AND FIFTH STREETS. 1851. Entered according to the Act of Congress, in the year 1849, by HENRY C. BAIRD, in the Clerk's Office of the District Court for the Eastern District of Pennsylvania. \ PHILADELPHIA t T. K. AND P. G. COLLINS, PRINTERS. PREFACE. THIS book has been written with a special regard to practical utility. In what manner this object has been fulfilled, we leave the intelligent reader to judge. The character of the work is purely technological. This object we not only deemed desirable in itself, but we were necessarily restricted to it on account of space. A mere description of materials and of manipulations amounts to nothing more than an enumeration and re- cord of facts. This we considered insufficient to satisfy the wants of an inquisitive community. Therefore, each division of the book contains a philosophical investigation concerning the apparatus and manipulations applicable to specific cases, as well as the basis whence their relative advantages are deduced. No book which embodies only a collection of confused or partially developed facts is adapted either to attract or to fix the attention of a thoughtful mind. The little interest which men, even of education and intelligence, take in certain mechanical pursuits that are worthy of all notice, is probably to be attributed to the rarity of the treatises which elucidate the principles such pursuits involve. This evil we have sought to avoid, without, at the same time, making our book so scientific as to render it useless as a practical treatise. This work contains imperfections for which we cannot consistently ask the indulgence of the reader. It may even embody errors; these, on the ground of human M18G173 iy PREFACE. frailty, may be deemed, by the kind-hearted reader, excusable. The expression of one fact will, we hope, disarm critics. "We make no claims as a writer. We make this statement, not only because the language of the book is not our native tongue, but because, though it were, we doubt whether we should be able to exhibit a reasonable proficiency in its use. Many of the repetitions which the reader will observe may appear to be superfluous. Some of these were de- signed; others, despite every precaution, were unavoid- able. In verbal communications, we are enabled to draw attention to a given subject by a bold assertion, or a striking illustration. But in a technical work, designed to convey important information, a certain amount of repetition is almost indispensable. Quotations and references we consider inappropriate in a work like the present. But we have not hesitated to insert them, where this could be done without interfering with the current of the text. In addition to the authors we have quoted, we acknowledge our indebtedness to the German authors Karsten, Knapp, and Sheerer. The publisher has spared no expense in relation to the typography and engravings of this work, which have been executed in a manner equal to anything the country can afford. Woodcuts are preferable to lithographic or copperplate illustrations, on account of the facility with which they can be printed on the exact spot to which they belong. If the book, with all its incongruities, shall be accepted kindly by the public, our labors will have been more than compensated. F. OVERMAN. PHILADELPHIA, November, 1849. TABLE OF CONTENTS. CHAPTER I. IRON ORE. PAGE I. Native iron . . . . . . .17 II. Oxide of iron . ... . . . .17 a. Protoxide of iron . . . . . .18 6. Magnetic black oxide of iron . . . .18 c. Peroxide of iron . . . . . .19 d. Hydrated oxide of iron . . . . .21 III. Carburets of iron . . . . . .23 IV. Sulphurets of iron . . . . . .24 a. White sulphuret . . . . . .24 b. Yellow sulphuret . . . . . .24 V. Phosphurets of iron . . . . . .25 VI. Arseniurets of iron . . . . . .25 VII. Chlorides of iron . . . . . .26 VIII. Sulphates of iron . . . . . .26 IX. Phosphate of iron . . . . . .27 X. Carbonate of iron . . . . . .28 a. Sparry carbonate . . . . . .28' 6. Compact carbonate . . . . .29 XI. Titanate of iron ....... 30 XII. Chromate of iron . . . . Jjj . 31 XIII. Franklinite . . . . . . .31 XIV. General remarks . . . . . . .31 a. Theory of reducing ore to metals . . . .32 6. Metals and oxygen . . . . .32 c. Hydrates ....... 33 d. Reduction of oxides . . . . .34 e. Reviving of metals . . . . .34 f. Metals and sulphur . . . . .36 g. Metals and phosphorus . . . . .37 7i. Metals and carbon . . . . .38 t. Metals and acids . . 38 yi CONTENTS. PAGE XV. Roasting of iron ore . . . . .39 a. Magnetic oxide of iron . . . . .39 b. Hydrated oxide . . . . .39 c. Carburets of iron . . . . . .39 d. Sulphurets ...... 40 e. Phosphurets . . . . . .40 f. Arseniurets . . . . . .40 g. Chlorides . . . . . . .40 7i. Sulphates ....... 40 t. Phosphates . . . . . .40 k. Carbonates . . . . . .40 aa. Roasting in ovens, kilns . . . . .41 bb. Roasting in mounds . . . . .43 cc. Roasting in the open air, in heaps . . .44 XVI. Cleaning of roasted ore . . . . .46 XVII. Theory of roasting ore . . . . . .47 XVIII. Mixing of ore . . . ... . .48 XIX. Practical remarks . . . . . .49 a. Magnetic oxide of iron . . . . .50 6. Sparry carbonate . . . . . .50 c. Specular iron ore . . . . . .50 d. Hydrated oxide of iron . . . . .51 e. Compact carbonate . . . . .52 XX. Mining of iron ore . . . . . .52 XXI. Fluxes ........ 68 a. Lime ....... 68 b. Magnesia . . . . . . .71 c. Clay . . . . . . .71 d. Silex . . . . . . .71 e. Ashes of the fuel . . . . . .72 XXII. Assay of iron ore ...*'. 72 CHAPTER II. FUEL. I. Wood 80 a. Amount of water in wood . . . .80 b. Hard and soft wood . . . . ..81 c. Specific gravity . . . . . .81 d. Ashes . . . . . . .81 e. Practical remarks . . . . . .83 II. Turf or peat ...... 85 a. Ashes ...... 86 b. Chemical analysis of turf . . . .86 c. Practical remarks .... 37 CONTENTS. yii PAGE III. Fossil coal ....... 87 a. Brown coal . . . . . .88 b. Water in brown coal . . . . .88 c. Ashes ....... 88 d. Chemical composition . . . . .89 e. Bituminous coal . . . . . .89 f. Water in bituminous coal . . . .90 g. Ashes . . . . . . .90 Chemical composition . . . . .91 Ji. Practical remarks . . . . .91 i. Classification . . . . . .91 k. Mining . . . . . . .92 1. Anthracite . ... . . .99 Water ....... 99 Ashes . . . . . . .99 Chemical composition . . . . .99 Practical remarks . . . . fc .99 m. General remarks on fuel . . . . . 100 n. Quantity of coal in different parts of the world . . 100 IV. Distillation of fuel . . . . . .102 V. Charring of wood . . . . . .103 a. Charring in pits ...... 103 6. Charring in heaps ..... 104 c. Charring in mounds ..... 108 d. Charring in ovens ..... 110 e. Brown charcoal ...... 113 f. Distillation in closed vessels .... 114 g. General remarks on charring .... 114 Ti. Yield 115 i. Products of distillation ..... 117 k. Season for charring ..... 117 VI. Charring of turf . . . . . .117 VII. Charring of brown coal . . . . .119 VIII. Charring of bituminous coal . . . . 119 a. Coking in heaps ...... 119 6. Coking in rows . . . . . . 121 c. Coking in ovens ...... 122 d. Coking in iron retorts ..... 129 IX. General remarks on coking ..... 129 X. Heat liberated by fuel . . . . . 132 a. Quantity of heat . . . . . .132 6. Quality . . . . . . .136 XI. Analysis of fuel 137 viii CONTENTS. CHAPTER III. REVIVING OF IRON. PAGE I. Reviving of iron in a crucible ..... 141 II. Reviving of iron in a stuck, or wulf 's oven . . . 141 III. Reviving of iron in a blue, or cast oven . . . 143 IV. Reviving of iron in various blast furnaces . . . 144 a. Blast furnace in the Hartz Mountains, Germany . 144 6. Blast furnace at Malapane, Silesia . . . 145 c. Blast furnace for smelting bog ore . . . 146 d. Blast furnace for smelting spathic ore, Styria . . 147 e. Blast furnace in Sweden ..... 148 f. Blast furnace in Saynerhutte, Germany . . . 148 g. Blast furnace at Cold Spring, N. Y. . . 150 h. Blast furnace in Pennsylvania . . . .151 V. Modern charcoal blast furnace . . . . .151 a. Building of a stack cupola blast furnace . . 153 6. Starting of a charcoal furnace .... 164 c. Charges of a charcoal furnace .... 165 d. Practical remarks ...... 166 VI. Coke furnaces . . . . . . .174 VII. Coke furnaces in Hyanges, France .... 177 VIII. Anthracite furnaces ...... 179 IX. Management of blast furnaces ..... 183 a. In-wall, or lining ...... 183 b. Hearth . . . . . . .183 c. Application of fire ..... 184 d. Charging with ore ..... 185 e. Nature of charges . . . . . 185. /. Tap-hole . 185 g. Tuyeres . . . . . .185 Ti. Application of blast ..... 186 i. Height of damstone ..... 186 k. Slag 186 L Tools . . . . . . .187 m. Management of the tuyere . . . . 187 n. Management of the timp ..... 188 o. Accumulation of cold clinkers in the hearth . . 190 p. Tapping of iron ...... 190 q. Causes of disorder ..... 190 r. Wet bottom stone ..... 191 s. Filling of the furnace . . . . .191 t. Coal charges . . . . ... 194 u. Size of coal ...... 195 v. Ore charges ...... 195 w. Size of ore ...... 196 x. Mixing of ores ...... 197 CONTENTS. ix PAGE y. Number of charges . ... 200 z. Number of tuyeres require.d in a furnace . . 200 aa. Time of tapping iron ..... 201 bb. Quality of iron depending on the tuyere . . 201 cc. Quality of iron depending on burden . . . 202 dd. Quality of iron depending on flux . . . 203 ee. Quality shown by cinder ..... 203 //. Scaffolding in the furnace . . . 204 gg. Flame of trunnel-head ... . 204 hh. Appearance of melted iron .... 205 Theory of the blast furnace . . . . .206 a. Combustion ...... 206 b. Nature of combustion in a blast furnace . . 207 c. Composition of gases ..... 207 d. Reduction of ore . . . 209 e. Operation in the furnace . . . .211 f. Conditions required in the interior . . .211 g. Theory deduced from practice .... 212 ^. Conditions of the ore in the furnace . . . 214 i. Composition of ore best adapted to receive and retain carbon ...... 214 k. Influence of foreign matter contained in the ore . 215 I. Influence of foreign matter relative to absorption of car- bon . .216 m. Clay, silex, and lime ..... 217 n. Clay, silex, and lime in ore .... 217 o. Magnetic ore want of cinder .... 218 p. Causes of white iron ..... 218 q. Difficulty of smelting gray iron from clay and silicious ores ....... 219 r. Ores which flux themselves .... 220 s. Calcareous ore ...... 221 t. Clay ores ....... 222 u. Capacity of matter for carbon .... 222 v. Amount of fuel consumed by different ores . . 223 w. Fusibility of ores . . . . . .225 x. Effects of fusibility . . . . .225 y. Degree of fusibility of ore compositions . . . 226 z. Effect of the alkalies in cinder .... 228 aa. Fusibility of iron and cinder .... 229 bb. Nature of fluxes ...... 230 cc. Composition of cinder . . . . . 231 dd. Titanium in cinder . . . . . 235 ee. Method of working titaneous ore ... 236 ff. Cinder from steel iron ..... 236 gg. Construction of furnace for making steel metal . . 238 hh. Cinder from a coke furnace . . . 239 CONTENTS. CHAPTER IV. MANUFACTURE' OF WROUGHT IRON. PAGE I. Oriental mode of making iron ..... 243 II. Catalan forge ... .245 a. Quality of iron from the Catalan forge . . . 248 b. Stuck oven 248 III. German forge 249 a. Material employed in the German forge . . .251 b. Form of hearth .... .251 c. Influence of the quality of coal . . , .251 d. Use of fluxes 252 e. Size and form of the hearth .... 252 /. Manipulation ...... 253 g. Tools . . . . . . .255 h. Yield 255 IV. Finery fire .... . 256 a. Quality of the refined iron .... 259 V. Puddling furnaces . . . . . .259 a. Construction of furnaces ..... 260 b. Exterior 262 c. Puddling and boiling ..... 265 d. Manipulation ...... 265 e. Tools 266 /. Puddling 267 g. Boiling . . . . . . .268 Ti. Difference in the construction of puddling and boiling furnaces . . . . . .270 i. Cinder from puddling and boiling . . . 270 k. Anthracite furnace . . . . .271 Z. Heating stove ...... 274 m. Puddling at Hyanges, France .... 274 n. Improvement of puddling by addition of fluxes . .276 o. Practical application of fluxes .... 278 VI. General remarks on charcoal forges .... 280 a. Location of a forge ' . . . . . 280 b. Ores adapted to the charcoal forge . . . 281 c. Quality of metal required .... 281 d. Site of the forge . . . . . .281 VII. General remarks on puddling ..... 282 a. Quality of iron . . . . . .282 b. Boiling process ...... 283 c. Iron boshes, and brick or soapstone lining . . 283 d. Effect of iron boshes on the iron manufactured . . 285 e. Cold iron boshes ...... 288 /. Iron for specific purposes ..... 288 CONTENTS. xi PAGE g. Elements of pig iron, and their effect on the iron manu- factured ...... 291 "h. Mixing of pig iron and artificial fluxes . . . 294 z. Hearth of a furnace ..... 298 k. Roof of a furnace ..... 299 1. Depth of the bottom . . . . .299 m. Dimensions of the grate ..... 300 n. Fuel . . . . . . .300 0. Heating stoves ...... 301 p. Wages . . . . . . .301 q. Yield 301 VIII. General remarks on refining ..... 302 a. Various methods of refining .... 304 IX. Theory of refining and puddling .... 305 a. Difference between cast and wrought iron . . 305 b. Impurities contained in wrought iron . . . 307 c. Causes of cold-short iron ..... 309 d. Removal of impurities . . . . .310 e. Nature of the combination of iron and other matter . 311 f. Iron containing calcium and silicon . . . 312 g. Cause of inferior pig iron . . . .314 h. Cinder, the criterion of the quality of iron . .316 1. Influence of cinder on the permanency of the fibre in wrought iron ...... 318 k. Fusibility of iron and cinder; composition of cinder . 320 I. Critical view of the present mode of refining . . 330 m. Philosophy of improving iron .... 332 n. Qualities of wrought iron .... 333 CHAPTER V. FORGING AND ROLLING. I. Forge hammers ....... 334 a. Tilt hammer . . . . . . 334 6. T hammer . . . . . . .337 c. Brest hammer ...... 338 d. Improved jack . . . . . . 338 e. Steam hammer ...... 339 /. Shingling rods . . . . . .339 g. Faces of hammers and anvils ... . . 340 II. Squeezers ....... 341 a. Lever squeezer ...... 341 6. Rotary squeezer ...... 342 III. Roughing rollers ...... 344 a. Form of the grooves ..... 346 6. Flat bar rollers 347 xii CONTENTS. PAGE c. Construction of housings and rollers . . . 349 d. Coupling-box, cam-box, and junction-shafts . . 352 e. Flywheel . . . . . . .353 IV. Merchant mill . . . . . ... 354 a. Small rod iron ...... 354 5. Eound and square iron ..... 356 c. Hard rollers ...... 356 d. Adjusting rollers . . . . . .357 e. Flat rollers for small rods .... 357 f. Wire and hoop rollers . . . . .358 V. Heavy bar and railroad iron rollers . . . .358 a. Form of rails ...... 359 6. Rollers for rails . . . . . .360 c. Shingling of rail piles ; cutting off the fagot ends . 362 VI. Sheet iron . . . . . . .365 a. Necessity of re-heating blooms .... 366 b. Machinery for making sheet iron . . . 366 c. Wrought iron standards for thin sheet iron . . 367 d. Boiler-plate ...... 368 e. Common sheet iron ...... 369 f. Color of sheet iron ..... 371 VII. Re-heating furnaces . . . . . 371 a. Size of grate . . . . . 372 6. Quantity of iron re-heated . . . . . 373 c. Scrap furnaces ...... 374 VIII. Heating ovens ....... 374 a. Fuel in heating ovens ..... 375 6. Ancient form of the oven .... 375 c. Modern heating oven . . . . . 376 IX. Shears and turning machines ..... 378 a. Turning of rollers ..... 378 6. Hand shears ...... 378 c. Force shears : with excentric, and with crank . . 379 d. Fastening the cutters . . . . .381 X. Tools 382 XI. General remarks ...... 383 a. Catalan fires . . . . . .383 6. Hammers ..... 383 c. Rolling mills ...... 384 d. Small rod and hoop iron ..... 385 e. Wire iron ...... 385 /. Railroad, bar, and heavy iron . . . .386 g. Piling of iron ...... 387 h. Sheet iron ...... 389 t. Nails . 391 CONTENTS. xiii CHAPTER VI. BLAST MACHINES. PAGE I. Wooden bellows of the common form .... 394 II. Wooden cylinder bellows . . . . .395 a. Double stroke tubs ..... 396 b. Construction of tubs ..... 397 III. Iron cylinder blast machines ..... 398 a. Double cylinders, with beams .... 398 b. Horizontal cylinder ..... 399 c. Moving power for blast machines . . . 400 IV. General remarks on blast machines .... 400 a. Size of blast cylinder ..... 401 b. Size and form of valves ...... 401 c. Advantages of vertical cylinder .... 403 d. Packing of the piston . . . . . 403 V. Various forms of blast machines .... 404 a. Trompe . . . . . . .405 b. Chain-trompe ...... 405 c. Gasometer bellows ..... 405 d. Screw blast machine ..... 406 VI. Fan blast machines ...... 407 a. Improved fan ...... 409 c. Pressure in a fan ...... 409 VII. Receivers, or regulators of blast .... 410 a. Dry receiver . . . . . .411 VIII. Blast pipes 413 a. Nozzles ....... 415 IX. Tuyeres . . . . . . . .417 a. Tuyeres for forges . . . . . 418 b. Water tuyere ...... 419 Number of tuyeres ..... 422 X. Valves ........ 423 XI. Manometer ....... 424 XII. General remarks on blast machines .... 426 a. Effect of blast machines ..... 426 b. Location of blast machines .... 426 CHAPTER VII. HOT BLAST. I. Hot air apparatus ...... 428 a. With round pipes ...... 429 b. With straight pipes . . . . .431 CONTENTS. PAGE c. Cleaning of the tuyere . . . . ' . 432 d. Measurement of temperature .... 433 II. Theory of hot blast . . . . . .434 a. Chemical effect ...... 434 5. Chemical effect in reducing ore .... 436 c. Effect on cinder ...... 437 d. Effect on iron ...... 437 III. General remarks on hot blast ..... 440 CHAPTER VIII. WASTE HEAT AND GAS. I. Waste heat . . . . . . .444 a. Of mill furnaces . . . . . .444 6. Of blast furnaces . . . . . .445 c. Advantages of waste heat ' 446 II. Gas . . . . . . . .446 a. Carbonic oxide gas ..... 446 b. From the blast furnace ..... 447 CHAPTER IX. FIRE BRICK AND REFRACTORY STONES. I. Native refractory stones . . . . .453 a. Sandstone ....... 453 6. Clay slate . . . . . . .453 c. Talc slate . . . . . . .454 II. Artificial refractory stones . . .' . . 454 a. Fire brick ....... 454 &. Artificial sandstone ..... 456 c. Refractory mortar . . . . . 457 III. Conductors of heat ...... 458 CHAPTER X. MOTIVE POWER. CHAPTER XI. MANUFACTURE OF STEEL. I. Damascus steel ....... 465 II. German steel ....... 466 a. From ore . . . . . . . 466 6. Woots ....... 466 c. From crude plate iron ..... 466 d. Steel on wire drawplates . . . . .468 CONTENTS. XV PAGE III. Iron for blistered steel . . . . . .469 a. Ore for making the iron ..... 469 b. Iron for making blistered steel .... 470 IV. Blistered steel . . . . . . .471 a. Quality of the steel ..... 474 6. Influence of the hammer on blistered steel . . 475 V. Cast steel . . . . . . .475 VI. General remarks on steel ..... 478 Chemical composition of steel .... 481 Hardening of steel . . . . .483 CONCLUSION , 484 APPENDIX. Table I. Composition of crude cast iron .... 486 Table II. Composition of gray cast iron .... 487 Table III. Composition of steel metal ..... 487 Table IV. Composition of forge crude iron .... 488 Table V. Composition of wrought iron .... 488 Table VI. Decomposition and recomposition of materials in the blast furnace ....... 489 Table VII. Specific gravity of matter ..... 489 Table VIII. Degrees of heat generated by perfect combustion . . 490 Table IX. Degrees of heat at which substances melt . . . 490 Table X. Capacity of matter for latent heat . . . 490 Table XI. Expansion of air by heat ..... 491 Table XII. "Weight of substances . . . . .491 Table XIII. Weight of a superficial foot of sheet iron . . .491 Table XIV. Weight of rod iron one foot in length . . .492 ON THE MANUFACTURE OF IRON. CHAPTER I. IRON ORE. A GEOLOGICAL classification of the ores of iron is, in our case, not the proper way to divide the subject before us : it would not include that clear, comprehensive, practical demonstration needed for our purpose; and we choose, therefore, a division based upon the compo- sition of the material, or a Chemical classification. According to this, we shall divide the iron ores proper into Native Iron; Oxides, Carburets, Sulphurets, Arseniurets, and Phosphurets of Iron; Chlo- rides, Sulphates, Phosphates, Carbonates, and Titanates of Iron. I. Native Iron. The deposits of native iron are very limited, and the insufficient quantity of material it affords, precludes it from being ranged, for our purpose, among the iron ores. We notice it as a matter of curiosity, merely to complete the class. Native iron has been found in Canaan, Conn., in a vein or plate two inches thick; it is suffi- ciently ductile to be wrought into nails by a blacksmith. It was found in a mica slate rock, upon a primitive mountain, and very much intermixed with plumbago. In France and Germany native iron has also been found ; but there are serious doubts whether it is formed by nature ; and its existence may probably be assigned to the previous burnings of stone coal in its vicinity. II. Oxides of Iron. These constitute the most important class for the manufacture of iron. They may be considered under four distinct subdivisions, namely, Protoxide, Magnetic Oxide, Peroxide, and Hydrated Oxide of Iron. 2 18 MANUFACTURE OF IRON. a. Protoxide of Iron has never been found as a natural deposit, and it is difficult, even in the chemical laboratory, to make it. It can be made by precipitating salts of the protoxide by caustic soda ; but it is very apt to oxidize in being washed and strained, whereby a part of it is converted into oxide. The best way to produce it, is to oxidize iron heated to redness by means of steam. It is of a black color, attracted by the magnet, and very hard. It is composed of 77c23 iron 22.77 oxygen 100.00 peroxide of iron. Should, therefore, an iron ore exist of this composition, it could not contain more than 77 parts of iron in 100 parts of ore. b. The next degree of the oxidation of iron is the Magnetic Black Oxide of Iron, Loadstone. Its color is a grayish-black ; and when rubbed, it gives a black powder. It is strongly attracted by the magnet, and is magnetic itself. It is altered neither by nitric acid nor the blowpipe. It dissolves slowly in hydrochloric and diluted sulphuric acids, the former of which dissolves the protoxide, and leaves a red powder, peroxide, undissolved. This circumstance is evidence of its being no particular oxide of iron, but a mixture of the protoxide and the peroxide. Its composition is, in 100 parts, 71.79 iron 28.21 oxygen 100.00 magnetic oxide of iron ; or, it consists of 31 parts of the protoxide and 69 parts of the peroxide of iron ; and in 100 parts of ore there cannot be more than 71 per cent, of iron. This species of iron ore constitutes a large body of the native deposits. It is found in Sweden, Norway, Siberia, China, Siam, the Philippine Islands, Germany, France, and very little in Eng- land. There is a large deposit at Lake Champlain, N. Y., of the best quality. It is also found in Bridgewater, Vt., Marlborough, Vt., and Franconia, N. H.; and New Jersey and the State of New York contain it in large quantities. The exploration of the north- west of the United States promises an addition to the already known valuable deposits, for the iron mountain in Missouri appears to belong to this class. This is one of the most valuable ores, fur- nishing, by proper treatment, the best quality of iron. From it the main body of the superior iron from Sweden, Russia, and Germany IRON ORE. 19 is manufactured; but the modern improvements in manufacturing, particularly the hot blast, appear to impair its good disposition, and furnish inferior qualities of iron. We will, in the following chap- ters, explain the reasons why this ore requires particular treat- ment and attention. Magnetic iron occurs in primitive rocks, commonly in gneiss, sometimes in clay hornblende or chlorite slate, greenstone, and limestone, and is mixed with epidote, pyroxene, and garnet. We never find it in more recent geological deposits. Its crystalline form is an octahedron, and it varies, in size from an inch to the finest sand. It is seldom found in solid masses. c. Oxide of Iron, Peroxide of Iron, Iron-glance, Specular Iron, and Red Iron Ore. These subdivisions of the oxides form a very extensively distributed ore. This ore is very hard, sometimes the color of polished steel, and crystals of this kind transmit light through the edges, and appear to be beautifully red. When coarse, the oxide is of a brown color ; but its powder is always red, thus dis- tinguishing it from the magnetic oxide. It is infusible before the blowpipe, but melts with borax, and forms a green or yellow glass. Heated hydrochloric acid is the only acid able to dissolve it. By high temperatures, without the addition of any other matter, it is reduced to the magnetic ore. The magnet does not attract it, nor is the magnet attracted by the iron. Oxide of iron is composed, in 100 parts, of 69.34 iron 30.66 oxygen 100.00 protoxide of iron. This oxide of iron is used for various purposes besides the manu- facture of iron : as calcined hydrate, it forms a red-brown paint Spanish or Indian brown, which is the most durable of all paints for preserving wood and iron. In northern Europe the houses of the peasantry are mostly painted with it. It serves for polishing silver and gold, and for that purpose is manufactured from cop- peras, which is calcined along with common salt. The red color of the common brick is oxide of iron. Those varieties of specular iron ore which have lost their metallic appearance, are called red iron ore; they are either fibrous or solid, compact or ochry ; sometimes they form a firmly connected mass of a red impalpable powder. The scaly red iron, and the red iron foam belong to this class ; in masses they are but slightly coherent. The 20 MANUFACTURE OF IRON. whole variety is in close connection with the micaceous specular iron, between which and the crystallized oxide of iron is an unin- terrupted transition. If this variety of ore is mixed with foreign matter, its red color is sometimes altered and, mixed with silica, lime, &c., turns into hydrates of iron ; but an admixture of clay does not alter its red color, and the ore is called clay ore. Reddle, jaspery clay ore, columnar, and lenticular iron ore, are of this kind: the first of which is compact, friable ; the second very hard, of conchoidal fracture; the third, distinguished by its columnar forms; and the latter, by its granular composition. This variety of ore yields very unequal amounts of iron; it ranges from the red clay of hardly 12 per cent, of iron, to the rich micaceous ore, which is pure oxide of iron. In this case, the evidence of sense is no safe dependence, for a very poor clay appears sometimes as red as the richest'ore though by drying the specimens, a difference in color may be perceived: still, it would be premature to infer from this, what amount of iron a given specimen contains. The only way to ascertain the quantity of iron is by chemical analysis, and the humid is the only test we can depend upon. But this variety of ore yields always good and strong iron, and is, perhaps, on that account, the most valuable; for the iron manufactured from it is the most tenacious of all known kinds. It improves, even in small quan- tities, all inferior ores, and forms a most excellent flux in the blast furnace. The damask iron of Persia and the woots of India are manufactured from specular iron ore. Red iron ore occurs most commonly in ancient rocks, and transition clay slate is generally its locality, where the best and richest beds are deposited. The Island of Elba is justly celebrated for an inexhaustible abundance of specu- lar iron, which has been worked since immemorial antiquity. The total height of the metalliferous mountain is more than 600 feet, and never will be exhausted. Specular iron ore is found throughout Asia, Corsica, Germany, France, Sweden, and in almost every country. The United States of America have yet afforded no amount worth noticing of the better qualities ; but immense beds of infe- rior quality, for instance, the Pittsburgh coal field, are loaded with some very valuable red clay ores, interspersed with nodules of the specular kind. Massachusetts, Ohio, and the western part of New York, contain similar deposits. Specular iron ore is found in crys- tals in the craters of volcanoes, the result of the evaporation of chlo- rides of iron in the fissures of lava. It forms heavy beds in transi- tion mountains, and is frequently found imbedded in clay in the shape IRON ORE. 21 of nodules of irregular masses. It is common in beds of spathic iron, in Styria and Carinthia, and generally associated with other ores of iron or earthy minerals, as epidote, hornblende, augite, cal- careous spar, and quartz. All the red clays belong to this class, and, when they contain more than 20 per cent, of metal, may be considered an ore of iron. d. Hydrated Oxide of Iron, Brown Oxide of Iron, Brown Iron Stone, Hematite. We have here a class of iron ores which, in quantitative importance, supersede any other kind in the United States. Hydrated oxide of iron always affords a yellow powder, without any shade of red, sometimes brownish, or even velvet black. At the blowpipe it turns brown or red, and>in the reducing flame black, and melts into a black cinder. Burnt or roasted, it is strongly attracted by the magnet, but not in its raw state. Cal- cined, it yields a red powder, oxide of iron, and is employed for the same purposes as the oxide. The yellow or brown varieties contain a large admixture of water in chemical combination, and hence they are called hydrates. Hydrated oxide of iron consists, in 100 parts, of 59.15 iron 26.15 oxygen 14.70 water 100.00 hydrated oxide of iron. Brown or yellow iron ore, therefore, never contains more than 59.15 Ibs. of iron in 100 Ibs. of ore. The mineralogical term of this ore is Limonite ; it comprises a great number of compound varieties. Its forms are various glob- ular, reniform, stalactitic, and raamillary. It presents great variety of surface, being smooth, granulated, reniform, drusy, columnar ; and it is often an impalpable powder. It is a species which, on ac- count of differences in regard to mechanical composition, has re- ceived a great diversity of names ; still, all the varieties are of the same chemical composition, unless adulterated by foreign matter. The whole class is the result of the decomposition of other iron com- pounds, namely, iron pyrites, carbonates, red oxides, sulphates, &c. The fibrous limonite, or brown hematite, contains sometimes beauti- ful crystals of the hydrate, and is known under the name of pipe ore, brown ore, and shell ore; it is then reniform, and consists of alternate layers of different color, or coats of different hardness. To this species belong also a great variety of impalpable and scaly compounds. 22 MANUFACTURE OF IRON. Limonite occurs in beds and veins, generally accompanied by spathic iron, calcareous spar, aragonite or quartz. We find these beds, or veins, both in ancient and secondary rocks, in tertiary de- posits, in diluvium, and alluvium. In the older rocks, limonite is generally derived from pyrites, and in the coal measures from car- bonates ; we find it in globular masses imbedded in clay, in sand- stone, and in bogs. Limonite is very plentiful all over the globe, particularly in the United States ; vast beds are near Salisbury and Kent, in Connec- ticut, resting in mica slate ; they are of the best kind of brown hematite, and are fibrous. In the State of New York, near Beek- man and Amenia, are similar deposits. Massachusetts is favored with that kind of ore ; also Vermont, Maryland, and Ohio. The whole iron business of Hanging Rock depends upon it. Ken- tucky, Tennessee, and Alabama, abound in inexhaustible beds of the best quality. But above all, Pennsylvania has the richest varie- ties of this kind. No doubt there is more in the United States than we at present know of, and the great valley between the Rocky Mountains and the Alleghanies is a natural basin for all such valu- able deposits swept down from Canada, and the impenetrable north. Limonite is the main source of the iron of commerce all over the globe. It affords an easy and cheap material, and the better varieties are excellent iron ; but we have to be careful in the selec- tion of the ore beds. The eastern ore is generally of prime qua- lity ; so is that of Hanging Rock in Ohio ; that of Tennessee and Alabama is of as good a quality of this kind as one could desire ; but the deposits of the coal formation, the pipe ores, and bog ores, are to be carefully selected in reference to quality. This kind of ore in the older rocks is generally good, but where it is derived from more recent deposits, it contains some of the original matter from which it is decomposed. The pipe ore is decomposed sul- phuret, and frequently we find a core of pyrites in the centre ; then the ore furnishes hot-short iron ; but, carefully roasted, the sulphur of the pyrites can be mostly evaporated. The hydrates of the coal formation are mainly derived from spathic iron, and frequently con- tain carbonic and sulphuric acids, which impair the quality of the metal, but can be removed by a careful roasting of the ores. Bog ores, which mostly contain phosphoric acid, are, for the manufacture of pig metal, incurable, for the phosphorus cannot be separated by roasting; but this separation can be effected in the forge, and hence, deserves consideration. In the main, this kind of ore furnishes an IRON ORE. 23 excellent material in the blast furnace, yields cheap pig metal, and of all classes of ore is the most available for improvement in the forge as well in the charcoal forge as in the puddling furnace. III. Carburets of Iron. Iron has a great affinity for carbon, but science has yet done very little towards investigating the nature of the different compounds. In the chemical laboratories, carburets of iron are generally made by decomposing in a high heat the salts of iron of the vegetable acids; we obtain in that way various compositions, whose nature is not investigated. Those compounds of iron and carbon deserve more attention on the part of scientific men than has yet been paid to them. The investigations of such men would enable us to understand the nature of pig metal better than we do at present. Some ores, of which we are at present ignorant, may belong to the class of carburets; they are certainly not found in the older rocks, but from the period of the coal measures to the present we may expect to find them. We are not aware that there are any employed in the United States in the manufacture of iron, but, where such can be found they deserve to be employed. In Scotland, the whole iron business depends mainly upon this kind of ore ; there it is called Blackband, and was first made use of by Mr. Mushet at the com- mencement of the present century. After encountering great oppo- sition, this ore enables Scotland to be master in every pig iron market which she can supply. Carburets are black, sometimes grayish, of slaty appearance, more or less hard, but always harder than clay slate ; the powder is attracted by the magnet, and turns brown or red by being calcined. Some varieties burn in larger heaps without other fuel ; others have to be calcined, by adding coal or wood. Foreign matter is almost always mixed with the ore ; these admixtures are mainly silex or clay. Frequently this ore is classed with the magnetic oxide, on account of its black color ; but it is soluble in sulphuric acid, and with the escape of hydrogen leaves carbon, which distinguishes it from the black magnetic ore. In the coal deposits of Frostburgh (Md.), this ore is found of an inferior quality ; it generally contains but from 20 to 25 per cent, of iron. It is also found in small quantities in the Pittsburgh coal formation. This ore deserves the attention of the iron master ; for, if even 24 MANUFACTURE OF IRON. poor, it always furnishes good pig metal, and is, after being well roasted, an excellent material in the blast furnace : it is more in- clined to make gray foundry iron than any other ore ; besides that, it works exceedingly well in the furnace. IV. Sulphurets of Iron. Iron has a very great affinity for sulphur, and we are acquainted with five definite compounds. It is very difficult to separate iron from sulphur by heat alone. Of the five different compositions, two only deserve our attention the white and the yellow sulphurets. a. White Sulphur et of Iron. White Pyrites abound in coal beds, and in the accompanying strata of clay; also in regular veins along with ores of lead, copper and iron, in the transition rocks. They are very common all over the globe ; . and are found in New York, Massachusetts, Connecticut, Ohio, and other States. Before the blowpipe, sulphuret of iron becomes red ; upon charcoal, the sul- phur is evaporated, and oxide of iron remains ; it is very liable to decomposition. It is preferable to the yellow kind in the manufac- ture of copperas, and is, in coal mines, the most dangerous of any, for it often decomposes so quickly as to kindle the coal slack. There- fore, where it is frequently met with in coal mines, great cleanliness and order ought to be practiced. Its composition is, in 100 parts, 45.07 iron 53.35 sulphur 0.58 manganese 99.00 white pyrites. b. Yellow Sulphuret of Iron. Yellow Pyrites. This variety becomes red before the blowpipe, like the above ; in the reducing flame it melts into a globule, which continues red-hot for a short time, and possesses, after cooling, a crystalline appearance. In nitric acid, it is slowly soluble with the precipitation of sulphur, but in no other acid. It is composed of 47.30 iron 52.70 sulphur 100.00 yellow sulphuret of iron. Yellow pyrites is almost identical with the white pyrites, and the latter appears to be only different in containing more foreign matter. Both are widely diffused among the ores of iron. We find such in IRON ORE. 25 massive nodules, crystals, and veins, in the coal beds, clay slate, graywacke, greenstone, limestone, and in beds in primitive slate. It is the main material which is used for manufacturing copperas, alum, oil of vitriol, and Spanish brown, sulphur, and sulphuric acid. In the United States, we find iron pyrites in Vermont, New York, Ohio, New Jersey, Pennsylvania, Maryland, and, in fact, more or less in every State. This class of Iron compound does not belong to the iron ores proper, but its immense quantity, and its presence in coal beds, require especial notice, on account of the injurious effect it has upon the quality of iron, where it comes in contact with the ores or coal. The presence of pyrites is generally indicated by its sulphurous smell, either in roasting the ore or in the casting house ; and when such indication is manifest, the careful roasting of the ores, and the long exposure of the roasted ore to the atmosphere, are the best methods of removing the sulphur. If the main body of sulphur is found to be in the fuel, there is little hope of getting rid of it, for it cannot be entirely expelled where a surplus of carbon is present, as is the case in caking coal. Y. Phosphurets of Iron. Phosphorus combines readily with iron ; the compound is whiter than iron itself, can be beautifully polished, but is very brittle, cold-short. Native phosphurets are very seldom found, and we allude to them because the presence of phosphorus in the pig metal occa- sions it to be cold-short. Under the head of phosphate of iron we shall speak of the ores belonging to this class. VI. Arseniurets of Iron. A native compound of iron, arsenic, and sulphur, is called mis- pick el , or, if it contains silver, which is often the case, it is deno- minated argentiferous arsenical iron. Arsenic and iron have con- siderable affinity, and in smelting combine readily ; the composition is brittle, not magnetic. Mispickel is white, hard, of a vitreous lustre; it emits before the blowpipe arsenical fumes, and leaves asul- phuret of iron and arsenic soluble in nitric acid, and is composed of 36.04 iron 42.88 arsenic 21.08 sulphur 100.00 mispickel. 26 MANUFACTURE OF IRON. In Germany it is used for the manufacture of arsenious acid, and is sometimes mixed with iron ores ; but it is very apt to choke the top of the blast furnace, for the arsenic, evaporating in the greater heat of the hearth, condenses at the cooler top. The most interesting deposits of this ore are in the United States; at Franconia, N. H., Worcester, Mass., and Chatham, Conn. A small quantity of this material, mixed with the other iron ores, does no harm to the product ; a larger quantity occasions it to be cold-short, and is troublesome in the blast furnace. VII. Chlorides of Iron. Chloride of iron should hardly be ranged among the ores of iron, but in many respects it deserves our attention. Chlorides are fre- quently found among the iron ores of the hydrates as a chloride of iron, or of sodium, and in some other unimportant combinations ; their presence is unquestionable. We find indications of chlorine on the top of burnt ore piles, and in the wash-water of iron ores. Chlorides are seldom or never found in the more ancient deposits, and occur only in the hydrates, or brown iron ore. Their pre- sence in smaller quantities is very favorable, and promotes the operations in the blast furnace ; it accelerates the motion of the charges, and furnishes a liquid, lively cinder. Such ores are very apt to furnish gray iron, of an excellent quality for the forge, though generally too cold-short for the foundry. Larger quantities of chlo- rides occasion trouble in the blast furnace, and produce white pig metal ; but the metal is always of good quality. VIII. Sulphates of Iron. Sulphuric acid has great affinity for the oxides of iron, and can with difficulty be entirely separated from such. Neither heat nor strong alkalies separate the oxide Q_f iron from the sulphuric acid, and under all circumstances a part of the acid is left in an oxide or iron ore, where it is combined. Iron masters are not very apt to make use of the sulphates of iron, either as green, white, or red copperas ; but in that large body of iron ores, the hydrates, particularly those of the coal formation, there is more or less sulphuric acid mixed with the ore ; and, as this acid cannot be expelled entirely by heat, it is a dangerous enemy to the manufacturer. Whether sul- phuric acid is, or is not, in the ore, can be ascertained by pounding, and heating it to redness along with some filings of wrought iron, IRON ORE. 27 and by dissolving the protosulphate which is formed in water; that is, wash the whole mass in rain water, and test with chloride of barium for sulphuric acid. Sulphuric acid is generally found in the yellow hydrates, but may be observed in the whole class of hydrates. The great disadvantage arising from sulphuric acid in iron ores, is its indestructibility by heat; and if, besides heat, carbon is pre- sent, then the sulphuric acid is decomposed, and leaves sulphuret of iron. This happens either in the calcining process, or in the blast furnace, and on that account sulphuric acid acts in the same manner as sulphur itself, or pyrites, and occasions hot-short iron. IX. Phosphate of Iron. * Phosphate of iron, green iron ore, is of a dull blue color, and turns yellowish-brown before the blowpipe; or, in the reducing flame, into a black, porous slag; it is not magnetic, and it is soluble in hydrochloric acid. It is often dark lake green, and of a vitreous, silky lustre. Its composition is, in 100 parts, 62.52 oxide of iron 28.50 phosphoric acid 8.98 water 100.00 phosphate of iron. This ore is seldom found in large masses, but frequently inter- spersed in other ores, and for that reason we take notice of it. It occurs in small particles of aggregated plates, sometimes only visible by means of a microscope. Generally, this phosphate is mixed with the yellow hydrates, fossiliferous and bog ores, and is the cause of very cold-short iron; and on this account is, if not to be rejected, at least to be regarded with great suspicion. Still, the ores of this kind have one great advantage that is, of furnishing a cheaper iron than that from all the other ores; the phosphorus can be completely removed in the puddling furnace. Where such ore occurs, it is generally in large bodies, and can be easily wrought; so that the price of the ore is not to be considered an objection to it. It is, of all classes of ore, the best in the blast furnace, and consumes less fuel than any other kind. Where forges are in such a condition as to work the cold-short metal into saleable bar iron, or into any particular form, it is, beyond question, the 28 MANUFACTURE OF IRON. most available. In the course of this work we shall have oppor- tunities to refer to this subject again. In Europe, particularly in the plains of Russia and Prussia, there are immense masses of bog ore, from which large quantities of iron are manufactured. These ores contain more or less of the phosphate, and the iron produced is cold-short. In the United States, we believe, there is but little of this ore ; Michigan and Ohio contain it in small quantity. There may be bog ore in Alabama, Arkansas, and Florida: but the fossiliferous ore of Pennsylvania and Maryland contains phosphate of iron. X. Carbonate of Iron. Sparry Iron; Brown Spar. This most important species contains two varieties; the spathose, or sparry iron ore, and the compact' carbonate. a. Sparry, or Spathic Iron, Steel Iron Ore, is of a lamellar sparry fracture. Color, yellowish-gray, Isabella, or even brownish-red; turns brown before the blowpipe, and is then attracted by the magnet. After being taken from the mine, it assumes a brown tint by exposure to the atmosphere; gives a slight effervescence with nitric acid, and changes to a brown color. Manganese and mag- nesia, as well as carbonate of lime, are frequently found mixed with it. It melts into a green glass with borax. Its composition varies, but a specimen from Europe contained 63.75 protoxide of iron 34.00 carbonic acid 0.75 oxide of manganese 0.72 magnesia 0.78 lime and water 100.00 sparry iron ore. Sparry carbonate belongs to the primitive formation, forming vast veins and layers in gneiss and primitive slate and limestone; it is associated with quartz, copper pyrites, gray copper, fibrous brown oxide of iron, and carbonate of lime. Beds of immense quantities are found in Styria, forming at Eisenerz a mountain as high as the snow line, from which ore was dug by the ancient Romans. These beds appear inexhaustible. In Carinthia an excellent ore of this kind exists, from which iron and steel of the first quality are pro- duced. In fact, most of the iron and steel of Austria is derived IRON ORE. 29 from this ore. It is distributed all over Germany ; and the cheap, though celebrated, German steel is manufactured from sparry iron ore. The cutlery and weapons of Solingen, in Western Germany, are made from iron and steel of the sparry ore, which is dug in Siegen, occurring in heavy veins and beds in transition slate. This ore is found in France, England, Scotland, Russia, Spain, Switzer- land, and various other countries. A very considerable vein of spathic iron is found near Roxbury, Conn., traversing a vein of quartz, imbedded in gneiss; also in Plymouth, Vt. ; and in small quantity in Monroe, Conn. This is a very valuable and interesting species. It affords steel with the greatest facility, and is one of the most favorable ores in the Catalonian forge. By proper treatment, it produces an excel- lent kind of bar iron, which is sufficiently esteemed by the black- smith. b. The Compact Carbonate of Iron spherosiderite argillaceous iron ore has no relation externally with the sparry variety ; it com- prehends most of the clay iron stones of the coal measures, parti- cularly those which occur in flattened spheroidal masses, varying in size from the dimensions of a small bean to pieces weighing a ton. The color of this ore is commonly a dirty blue or gray, brown, reddish-brown and yellowish-brown. Fracture close-grained, hard, streaked white or brown. Blackens before the blowpipe, and, if cal- cined, is attracted by the magnet. This carbonate of iron, though belonging to the coal formation, is found in various places in the tertiary strata. It is the prin- cipal ore from which iron is smelted in England and Scotland, and yields usually from 30 to 33 per cent, of metal. It is largely distributed over the United States. Pennsylvania abounds in it. It exists in Maryland, Virginia, Ohio, Illinois, North Carolina, and Kentucky. The difficulty of working this kind of ore in the blast furnace, of which we shall speak in another chapter, maybe assigned as the reason why it is not more generally in use. England and Scotland use it extensively, and work scarcely any other kind. Prof. Rogers, in his Reports of the Geology of Pennsylvania, has given a great many analyses of argillaceous ores, of which we shall select the following : 30 MANUFACTURE OF IRON. In 100 parts of ore were found : 53.03 protoxide of iron 35. IT carbonic acid 3.33 lime 1.77 magnesia 1.40 silica 0.63 alumina 0.23 peroxide of iron 3.03 bitumen 1.41 water 100.00 argillaceous ore. This may be considered an analysis of one of the best specimens. Generally these ores contain no more than 30 per cent, of iron ; and an average of the argillaceous ores of the Pennsylvania and Mary- land coal measures would not go farther than 25 per cent. The compact carbonates afford with charcoal and cold blast an excellent forge iron; by the hot blast the quality is greatly injured; but if properly calcined, and the burden not too heavy, it forms an excellent gray foundry metal. Still the operations in the yard, of roasting and those of the blast furnace, are somewhat difficult, particularly for those who are not very experienced founders, and acquainted by practice with this kind of ore. In the chapter on blast furnaces we will refer to this subject. XL Titanate of Iron. Titaniferous Iron, Iron Sand, is an oxide of iron and titanic acid, and belongs to the class of the magnetic oxides. It is attracted by the magnet, is of a deep black color, metallic lustre, very hard, and perfectly opaque; melts into a black slag by a high temperature. It is generally found near volcanoes or volcanic rocks, but seldom in quantities sufficient to justify the erection of iron works; never- theless, the quality is mostly good, and the volcanic regions around the lakes may present, in the course of time, encouraging prospects. There are two classes of iron ore which do not belong properly to our department, but are interesting as well on account of their belonging to the United States alone, as on account of their large quantity and usefulness. For this reason we shall notice them. IRON ORE. 31 XII. Chromate of Iron. Chrome ore, or chromated iron ore, is infusible before the blow- pipe; acts upon the magnet after being roasted ; of difficult smelt- ing with borax. Its composition, in 100 parts, is 43.00 oxide of chrome 34.70 protoxide of iron 20.30 alumina 2.00 silica 100.00 chromedron. Chrome ore is found in serpentine and cotemporanequs rocks, in irregular veins and beds. It is found in Europe; but in largest quantity within the United States; at the Bare Hills, near Baltimore; at Hoboken, New Jersey, and at Milford and West Haven, Conn. Europe derives its supply from these places. XIII. Franklinite. DodecaTiedral Iron Ore. Color black, and behaves before the blowpipe like the black magnetic ore : but with alkalies in the re- duction fire, it emits fumes of white oxide of zinc, and becomes green. It is composed of 66.00 peroxide of iron 16.00 red oxide of manganese 17.00 oxide of zinc. Franklinite is found near Franklin furnace, in Hamburg, New Jersey, accompanied by another variety of zinc ore, in large veins and masses ; it is a species belonging to North America alone. XIY. G-eneral Remarks. The ores of iron are distributed over the whole globe in great profusion. They are found in every latitude and in every climate. But every mineral which contains iron does not constitute an iron ore. The consideration of quality and quantity determines the application of a mineral species to the manufacture of iron. The basis upon which our arguments in this case rest, is the general theory of reducing metals, and the experience of old establishments. We will proceed to define the general theory, and to illustrate that theory by facts. 32 MANUFACTURE OF IRON. a. Theory of Reducing Ores to Metals. The metals, with the exception of gold, silver, and copper, are seldom found in their native state. They are combined with other matter in their native beds, and it is the study of the metallurgist, by dissolving this combination, to reduce them to their simple condition. The matters thus combined, are oxygen, sulphur, carbon, chlorine, and phos- phorus ; and combinations of the oxides of metals with the acids of the above metalloids. b. Metals and Oxygen. Metals, particularly iron, combine very readily with oxygen, and form oxides. In the combinations of iron with oxygen, there are four distinct grades ; the first is one atom of iron with one atom of oxygen, or FO,* the protoxide. The second, two atoms of iron with three atoms of oxygen, F 2 3 , or the peroxide. The third is a combination of one atom of the protoxide with one atom of the peroxide, FO-f F 2 3 , the magnetic oxide ; and the fourth, one atom of iron with three atoms of oxygen, or the ferric acid. The latter is a production of the chemical laboratory, and is beyond the limits of our labors. The affinity of the metals for oxygen is different in different metals, and varies with the temperatures under which the combinations are formed. Some are oxidized by a temperature below freezing, as potassium or manganium : others by the medium temperature, as zinc, tin, lead, iron, &c. Some cannot be oxidized by the atmo- sphere at all, as gold, platina, silver. Most of the metals can be combined with oxygen by being dissolved in nitric acid, or nitro- muriatic acid (aqua regia). Some metals decompose water readily ; such are potassium, sodium, and the metals of the alkalies gene- rally ; but iron and zinc decompose water slowly. If, however, an acid be added to the water which dissolves the oxide formed, the decomposition of water goes on rapidly. In all these instances the oxygen of the water is absorbed by the metal, and the hydrogen liberated. Some metals cannot be oxidized by means of acids, nor directly by the atmosphere, as rhodium and iridium, but oxidize very easily by being previously melted together with potash or salt- petre. Chrome, and a few others, are of this kind. Noble metals are those which are not oxidized by heat and ac- cess of oxygen. To this class belong gold, platina, silver, iridium. Another class of metals are oxidized in the heat of a flame, but lose their oxygen in higher temperatures ; such as palladium, * F for ferrum (iron), and for oxygen. IRON ORE. 33 rhodium, quicksilver, nickel, and lead. All other metals, when heated with access of the atmosphere, absorb oxygen and retain it. When a metal combines with oxygen, it loses its metallic, and assumes an earthy appearance, sometimes of a white, or black color. For this reason the old chemists applied to the oxides of metals the term calc that is, resembling alkaline earth. This idea is worthy of notice, for most of the oxides of the metals are electro-positive, while but few are electro-negative. This sub- ject is of great importance in metallurgy, and deserves attention. Metals whose oxides are mainly electro-positive are gold, osmium, iridium, platinum, rhodium, silver, mercury, uranium, copper, bismuth, tin, lead, cadmium, zinc, nickel, cobalt, iron, manganese, and cerium. These are at least four times heavier than water ; very few are oxidized at common temperatures of the atmosphere, but all can be deoxidized by means of carbon. Metals whose oxides are mainly electro-negative, are selenium, tellurium, arsenic, chrome, vanadium, molybdenum, wolfram, anti- mony, and titanium. The oxides of these metals take the place of acids, and form, with the above oxides and the alkalies proper, salts of definite proportions. Metals which form with oxygen the alkalies proper, are potas- sium, sodium, lithium, barium, strontium, calcium, magnesium, aluminum, beryllium, yttrium, zirconium, and thorium. We should be cautious not to conclude that this classification of the oxides of metals into electro-negative and electro-positive, is absolutely or literally true, for most metals have oxides of different composition ; combine with one, two, three, five, or seven atoms of oxygen, and are in that proportion more or less alkaline or acid. The oxides of potassium and zinc are always electro-positive to those oxides whose metals are negative to potassium and zinc. Sometimes, in fact, the first oxide of a metal is an alkali, and the second an acid ; this is the case with the oxides of tin and man- ganese, and also with iron. The protoxide of iron is a strong alkali, and its peroxide so much of an acid, that both combine and form a distinct salt, with all the characters of neutralization, the magnetic oxide. We intend to refer to this subject in the theory of fluxes. c. Hydrates. Oxides of metals form definite compounds with water, and are then called hydrates. The water in the hydrates of potash and clay is so strongly combined with its base, that the 3 34 MANUFACTURE OF IRON. strongest heat is hardly sufficient to separate them. Other hy- drates are easily decomposed, as the hydrate of iron ; while a few are decomposed in boiling water. Hydrates always decompose and combine more readily than the oxides. d. Reduction of Oxides. Most of the oxides of metals can be decomposed, that is, the metal revived by means of carbon, under various conditions ; which conditions we will explain more particu- larly in the chapter on reviving iron. There is a great difference, however, in the affinity of oxygen for metals, and that may be assigned as a cause of their different behavior with carbon ; but the main cause is, undoubtedly, the aggregate form of the oxide : for carbon is strong enough to separate potassium and oxygen, and why not silicon and oxygen, or aluminum and oxygen ? The cohe- sion of the atoms of these oxides is so strong, that the particles, or congregation of atoms, which the oxides form, resist in a body the affinity of carbon for oxygen. We find this general law of the dif- ficult decomposition of particles, particularly applicable to the oxides of iron. The solutions of the peroxide salts of iron are very easily reduced to protoxide salts, but with great difficulty to metal; It appears, therefore, that the oxygen is more firmly connected to the metal in the protoxide than in the peroxide, or, that the atoms of the protoxide are more inclined to crystallization. Most of the other oxides follow the same law, and very few the reverse. To the latter belong mercury and tin. According to this general theory, the more oxygen the metal absorbs, that is, the higher the state of oxidation, the more readily will oxides be reduced to me- tals. This theory is confirmed by experience at the blast furnace, for we know by practice that magnetic oxide is disadvantageous in its raw state, and that it is far better after being roasted or oxid- ized. The most favorable condition of iron ore for the blast fur- nace is the peroxide of iron, the reason of which we will explain, hereafter; and if we cannot find native peroxides, we must pro- duce such by art that is, by roasting and calcining in the cheap- est and most practicable manner. e. Reviving of Metals. To illustrate the foregoing principle more fully, it will be best to explain the reviving of metals in each particular case. This will furnish practical proof that the reduc- tion of the oxides is the more complete, the higher the state of oxida- tion ; it will also prove that oxides are the material from which metals can be most conveniently derived. IRON ORE. 35 Potassium is produced by mixing the oxides with carbon and heat, or, more imperfectly, by heating the hydrated oxide of potas- sium along with metallic iron. Sodium is revived by the same means as potassium, but it is not so easily evaporated as potassium, and requires more heat. It re- vives more readily if the oxide of sodium is mixed with hydrated oxide of potassium. The metals of the alkaline earths, barium, strontium, calcium, cannot be reduced by means of carbon, because the metals are more permanent, and resist evaporation. The oxides of these metals are reduced by means of electricity. Magnesium, Aluminum, Beryllium, Crlucinum, Yttrium, cannot be revived by direct application of carbon. The best way of pro- ducing these metals is by melting their chlorides together with potassium. Tellurium and Arsenic can be made by exposing their hyper- oxides, mixed with carbon, to ignition. Chrome and Vanadium are revived from their oxides and hyper- oxides by mixing the oxides with carbon and igniting the mass. Molybdenum, Wolfram or Tungsten, may be easily revived from their oxides by means of carbon. Antimony and Zirconium are not so easily revived from their combinations ; they require some skilful manipulations. Titanium can be revived from titanic acid by means of carbon. It requires a high heat to melt it. Crold can be revived from its oxide by mere heat; of course more readily by adding carbon. Osmium, Iridium, Platinum, Palladium, Rhodium, have very little affinity for oxygen, and of course are easily revived. Silver, Mercury, Uranium, Bismuth, are very easily revived from their oxides by means of carbon. Copper, Tin, Lead, Zinc, are produced by exposing their oxides, mixed with carbon, to a red heat. Nickel, Cobalt, Iron, Manganese, Cerium, are easily revived from their oxides, but require a somewhat strong heat. We here observe that experience proves the oxides are the most available for the production of metals from their combinations, and of this fact we must not lose sight, for it not only justifies the roasting of ores, but shows that to be both necessary and economical. The perfect oxides alone, that is, the red oxides of iron, should be sent to the furnace in their raw state. We will describe the most common 36 MANUFACTURE OF IRON. combination of the metals with other matters, and in that way shall arrive at the safest means to convert such combinations into oxides. /. Metals and Sulphur. Metals combine very readily with sul- phur, and such combinations are called sulphurets. Iron espe- cially has great affinity for sulphur, and does not part with it even in the highest heat. The process by which sulphur combines with metals, is analogous to the process by which oxygen combines with them. Sulphurets burn; they emit light and heat; and, in all their chemical properties, are almost identical with the oxides. They are distinguished from the oxides by their metallic lustre. They are sometimes translucent, as, for instance, sulphurets of mercury and zinc. Very few sulphurets can be reduced by carbon ; but almost all of them by adding alkalies, or a metal which has a stronger affinity for sulphur. This is the case with the sulphurets of copper and lead. If these are melted, and metallic iron added, the iron will combine with the sulphur and revive the metals. Metallic oxides are reduced to sulphurets by adding sulphuretted hydrogen, or sulphuretted carbon; and perhaps this is the mani- pulation by which, in the laboratory of nature, where sulphuretted hydrogen abounds, metallic oxides are daily reduced. Sulphurets can be reduced by heating them in an atmosphere of hydrogen ; by which means sulphuretted hydrogen is formed; this application, however, is very limited, and does not apply to iron or copper. The most common way to reduce the sulphurets, is to transform them into oxides, and then to reduce the oxides. This is most safely done in the chemical laboratory. Sulphurets are transformed into oxides by roasting and calcining. The material should be pounded to powder, and then heated with access of the atmosphere. Great care should be taken that the mass does not melt ; for if this happens, the operation is a failure, and must be repeated. The largest quantity of sulphur escapes as sul- phurous gas; and the metal remains in the highest state of oxi- dation. One part of the sulphur is generally converted into sul- phuric acid, and remains with the oxide; another part remains with the metal, and is detected by adding an acid. This especially happens with the sulphuret of iron. Such remains of sulphur can be removed by adding alkalies, or washing with water, the latter of which extracts the sulphates and carries them off; but in case a part of the sulphur is left in the form of sulphuret, the whole mass should be roasted until it is properly oxidized. One way of con- IRON ORE. 37 verting the sulphurets into oxides, is very important, and deserves attention. As above mentioned, some metals are not very easily converted into oxides ; to effect this conversion, we should melt them together with alkalies, or with oxidized bodies which have a great affinity for water. This law applies equally well to the sulphurets. Sulphurets are, like the metals, very compact, and their atoms are not exposed to the influence of oxygen or any other matter unless when dissolved. If we melt the sulphurets together with alkalies, which, besides dissolving most sulphurets, have a great affinity for water, the aggregate form of the sulphurets is de- stroyed, and the atoms offer their poles to the poles of other matter; and if heated in the meantime, most of the sulphur is expelled either as sulphurous acid or sulphuretted hydrogen. The rest of the sul- phur is generally converted into sulphuric acid, and remains with the alkali. Barytes and lime are in this case powerful agencies ; more powerful than even the alkalies. The more permanent sul- phurets, melted together with chloride of sodium, are very quickly transformed into oxides. All other salts will act in the same way ; and nitrates even better than chlorides; but nitric acid is not so permanent as chlorine, and would be more expensive. This beha- vior of the sulphurets with alkalies and the salts, is particularly applicable to iron, and may be productive of benefit to the careful manipulator. If we consider the great affinity of the metals for sulphur, par- ticularly iron, whose affinity is very strong, and consider further the injurious effects of sulphur upon iron, we shall be very cautious in preparing and selecting our ores, for it frequently happens that sul- phur exists in ore where we least' suspect it; it is not only injurious to the metal, but to the manipulation in the blast furnace. We should, therefore, pay attention to the perfect oxidation of the ores, before we make any use of them. When describing the manipula- tion in roasting ores, we shall allude to this subject again. g. Metals and Phosphorus. Most metals combine readily with phosphorus, especially iron, though not so readily as with sulphur. Carbon is necessary, in almost every case, to produce a combina- tion of phosphorus with metal. Phosphorus is easily expelled by roasting a phosphuret without carbon ; but if carbon is present, the phosphorus adheres very strongly to the metal, and its evaporation is difficult. Just so it is with sulphur, for the same manipulation which removes sulphur will remove phosphorus. Phosphorus pre- sents to us no greater difficulties than sulphur. The main difference 38 MANUFACTURE OF IRON. between them is the different effect they have upon iron ; sulphur makes iron hot-short, and phosphorus cold-short: but phosphorus is advantageous in the blast furnace; the reverse is the case with sulphur. h. Metals and Carbon; Carburets. Iron, lead, and potassium have great affinities for carbon ; but if simply carburets of iron are to be smelted, the expulsion of carbon may be easily effected by roasting the ore. i. Metals and Acids frequently exist in native ores. The Halo- gen combinations of chlorine, bromine, iodine, fluor, are either eva- porated, or combine mostly with the oxides, that is, the alkalies, whose metals are to be smelted; they are never injurious. Sul- phates are more dangerous, for the sulphuric acid is, in the presence of carbon, decomposed, and leaves the sulphur in connection with the metal. This remark applies equally to Phosphates. Nitrates are not in the least dangerous ; for a small heat, with the presence of carbon, decomposes them. Carbonates sometimes require a strong heat, as well as a long time, to be decomposed, which is particularly the case with iron; and as carbonates are most generally protosalts, we never get the higher oxides directly. This makes the roasting of carbonates of iron very difficult; for, if the heat is strong enough to expel the carbonic acid, it is generally strong enough to melt the magnetic oxide, or the protoxide, together with foreign matter. Borates are injurious to the metal, but very advantageous in the furnace. Silicates are directly of no use, particularly those of iron, but may be converted into oxides by being melted with alkalies, and then oxidized. Tellurates, Arseniates, Antimoniates, Wolf- ramiates, Titanates, and Manganates, are very easily converted into oxides, and but slightly injurious in the manufacture of iron. We meet with the whole of these compounds of metals and acids in the native hydrates of the oxides, for in case the ore or hydrate is a decomposition of a salt, the acid is never entirely removed ; and should either of the above acids have access in any way to an oxide of iron, we shall surely detect it in the hydrate. Of all the hydrates, that of iron is the most apt to retain acids, partly on account of its electro-positive character, but mainly on account of its forming a great variety of basic compounds, which are more or less difficult of solution. Such basic salts are then mechanically mixed with the hydrates, and are the cause of forming hydrates from the oxides of iron. By all means, therefore, hydrates should be roasted. IRON ORE. 39 XV. Roasting of Iron Ore. "Whether an iron ore should be roasted, is a question which very seldom arises ; at least this question seldom ought to arise. With the exception of the red impalpable oxide, the whole body of iron ores require roasting ; even the specular iron ore, if it is very com- pact ; but the best oxide, if too compact, works badly in the fur- nace. All other ores should be subjected to calcination. Some iron masters are in the habit of using the hydrates raw, but this should not be done where clay ores are smelted, for these tend to blacken the tuyere ; or where the hydrates contain either chlorides or phos- phates. In the latter case, the pig metal will be cold-short, if there is too much phosphorus. Under all circumstances, however, it is best to roast the ores if we expect good metal and well-regulated furnace operations. The object of roasting ores is either to produce higher oxidation, or to expel injurious admixtures. In both cases, liberal access of atmospheric air is required ; we should, therefore, so arrange our roasting operations, as to fulfil these conditions, from which it will appear that different ores require different treatment. To explain this more fully, we shall take a review of the various ores. a. Magnetic Oxide of Iron. This ore is very compact, heavy, and of an almost metallic appearance ; to open the textures of the ore, to make it more porous, lighter, and to oxidize it more highly, it should be roasted ; sulphur is frequently combined with it. This ore melts into a slag by a cherry-red heat ; we should, therefore, avoid a high heat, for a melted clinker is useless and injurious in the blast furnace, and a melted mass cannot be oxidized by common means. b. Hydrated Oxide of Iron, Brown Oxide, Hematite, Bog Ore. This whole class ought to be roasted, not for the purpose of oxida- tion, but in order to drive off the acids, and destroy sulphurets and phosphurets, for all the ores of this class contain more or less in- jurious matter. This ore will bear a high temperature in roasting, if there is no foreign matter mixed with it ; but of this it is very seldom free. c. Carburets of Iron are to be roasted, partly on account of the sulphur which they frequently contain, and partly for the expulsion of the hydrogen which is generally combined with the carbon. The roasting of this ore is easily effected. 40 MANUFACTURE OF IRON. d. Sulphurets of Iron. These, of course, require roasting, if designed for the manufacture of iron ; the manipulation is difficult, and requires more than usual attention and time. e. Phosphurets of Iron, where they happen to be mixed with the oxides, should be roasted, if we expect medium qualities of iron ; but if the quality is no object, and cheapness the aim, then phos- phurets, in their raw condition, will answer. /. Arseniurets of Iron. If iron ores contain arsenic, it is best to roast them ; arsenic does not injure the metal ; but if the top or shaft of the blast furnace works cool, there is sometimes danger of chok- ing at the top, or of scaffolding at the lining above the boshes. g. Chlorine contained in iron ore does no harm whatever, and may be considered beneficial in roasting. h. Sulphates of Iron should be carefully roasted with liberal ac- cess of air. This will apply also to i. Phosphates. Jc. Carbonates require careful treatment. In the furnace they melt before carbon has -any influence upon them ; and if there is any admixture of foreign matter, the carbonates are very apt to pro- duce but a small quantity of white iron, with black cinder. The roasting of carbonates is difficult ; the best means of roasting them are, low heat, and, if possible, access of watery vapors, partly to carry off the heavy carbonic acid gas, and partly to prevent a too high temperature ; for, if the heat is too strong, the carbonate melts together with the oxide, and forms a black cinder. All other ores are easily calcined ; they require no particular attention. It is evident that, as the qualities of these ores are different, they should require different treatment ; and the question which meets us is, what arrangement, in each particular case, will best enable us to arrive at the highest perfection. For roasting ores, there are three distinct modes of manipulation ovens, piles, and rows. Each arrangement may be considered perfect for a particular kind of ore ; but each is not equally applicable to all varieties of ore. We must modify our manipulations according to circumstances, in order to produce appropriate results. Under all circumstances the ore to be roasted should be broken into pieces as small as those usually put into the blast furnace, say two or three inches ; if we neglect this, of course we cannot expect a good result, for it is obvious that large pieces will not receive heat and oxygen through their whole body so soon as smaller pieces; IRON ORE. 41 and as the main object is oxidation, no means should be neglected which will accomplish the end in view. The kind of fuel required is not of so much consequence as it is usually thought to be at charcoal furnaces. Wood and small charcoal (braise) are used ; but where wood is scarce, stone coal, properly applied, will answer ; coke or anthracite is preferable. Bad or sulphurous coal should be avoided, or at least coked before used. Turf or peat, or brown coal may be used, where they can be obtained upon advantageous terms. aa. Roasting of Iron Ore in Ovens or Furnaces. There are many different forms of ovens, but all of them can be reduced to that of the blast furnace, or the limekiln. They are either per- petual, or work by charges. These ovens are commonly from twelve to eighteen feet high, and contain from fifty to one hundred tons of ore at once. Fig. 1 re- presents such an oven for perpetual work : a is the shaft or circular Fig. 1. Section of a roast-oven. hearth, where ore and fuel are thrown in ; 5, b are the grate bars, which can be removed to let down the roasted ore ; c, c are side arches, which permit access to the draft holes : d, d, d, d are four arches, including the work arch. To start operations in such an oven, the grate bars are covered with wood ; upon this either small 42 MANUFACTURE OF IRON. charcoal, or stone coal, coke, turf, brown coal, or any fuel fit for the purpose, is placed ; then a layer of coal and ore alternately, until the oven is filled, after which the fire is kindled. When the lower portions of ore are sufficiently roasted and cool, they are ta- ken out, and either carried to the furnace, or, in case the ore is not sufficiently roasted, returned to the top. The air holes d, d, d, d are designed to admit air when it is needed, and to enable us to observe the progress of the work. An oven of fifty tons capacity ought to yield thirty tons of well roasted ore in twenty-four hours ; but this depends very much on circumstances, and especially upon the quality of ore to be roasted. As the top of the ore sinks, it is replaced by fresh charges of coal and ore. This oven is well qualified to roast the hydrates, carburets, and other easily worked ores ; but will not answer for carbonates, sulphurets, or even mag- netic ore, for these ores are too soon melted. In some parts of Europe, another kind of oven is in use, which affords a better product than the perpetual oven, and may be em- ployed with great advantage. This oven is represented by Fig. 2. Its interior is a cone, wide at the base, and narrow at the top. At Fig. 2. Section of an ore roasting oven. the bottom of this cone an arch of coarse pieces of iron ore is built, which supports the body of ore charged above it. This arch will admit enough fuel to keep up a lively fire. Where wood is plenty, it may be used in its green state, but any other fuel will answer quite as well. One great advantage which this arrangement has IRON ORE. 43 over the other (Fig. 1), is that it does not bring the fuel into con- tact with the ore ; and the workmen are enabled to give just so much heat as they consider necessary. Such an oven, properly managed, may answer for any kind of ore, provided it be suffi- ciently coarse to admit the draft of air needed for oxidation. Though this arrangement makes manipulating more expensive than the ar- rangement first presented, yet the qualitative properties of the pro- duct which it furnishes for there is no doubt that a good workman will deliver a more perfectly oxidized ore from this kiln than from the other more than compensate for this expense. An improvement upon this principle has been made in Sweden and Norway by erecting large circular ovens, like porcelain kilns, at the base of which, in furnaces built around, or in the centre of the oven, the fire is applied. Such an arrangement will work continuously, like that of Fig. 1, but is expensive both in the first outlay, and in the operation. Reverberatory furnaces have been tried for roasting ores, but with little success ; the operation proved too expensive. bb. Roasting in Mounds. Sulphurets and carbonates, which cannot bear a high heat, and require sometimes several fires, are best roasted in mounds. Mounds are formed on a level ground, and consist of three stone or brick walls: see Fig. 3. The area or Fig. 3. Ground plan of a roasting mound. hearth is open on one side, so as to admit the entrance of wheel- barrows or carts : the walls are about three feet high, and have at their bases fire chambers, where the fuel is applied. This is shown at a, a, a, a, Fig. 4. Through the piled ore are draft holes or chim- neys, 5, ft, which regulate the draft; by these chimmeys, the draft may be altogether stopped when the ore gets too hot. This kind 44 MANUFACTURE OF IRON. of oven or mound is very useful for small ores, and those which cannot bear much heat. Fig. 4. Z Section of an ore roasting mound. cc. Roasting in the Open Air in Heaps. This mode of calcining ore is undoubtedly the most available, and that generally practiced. It affords by good management excellent results. To form a heap, or heaps, the ground must be leveled, and in many cases covered with beaten clay. The area of such a level depends entirely on the amount of ore to be roasted, and the time in which it is proposed to be done. It may be laid down as a rule, that the longer the fire remains in a pile, or the slower the roasting is carried on, the better will be the result. If the time is limited, rows of three feet high, from seven to eight feet wide at the base, and of convenient length, may be put up and fired. These rows may be finished in ten or twelve days ; but though they answer well enough for hydrates, sulphurets, carburets, and all those ores which calcine easily, they do not answer for magnetic ore or carbonates. For those ores which are roasted with difficulty, round or square piles of various dimen- sions are used ; some of these piles have a capacity of from one hundred to two thousand tons. The amount of ore in fire should depend mainly on the stock on hand, and on the quality of the ore. Magnetic ore may be roasted in the course of six or eight weeks ; argillaceous ores of the blue or gray kind, require at least three months; and the sparry carbonates can scarcely be roasted in one heat, frequently require different fires, and, after all, are but seldom sufficiently calcined. In Styria, Carinthia, and other places where heavy sparry iron ore abounds, and where good iron must be de- livered, the iron masters are compelled to have a stock of ore suffi- cient to supply the furnace for a number of years, and the compli- IRON ORE. 45 cated manipulations by which the sparry carbonates are oxidized, often require a period of from three to five years. The operation is there mainly conducted on the principle of oxidizing by the influ- ence of the atmosphere ; for that purpose the ores are broken into small fragments of the size of walnuts, then spread upon level plains, in a thin stratum of about two inches thick, and then exposed to the action of the sun and atmosphere ; in dry weather the ores are sprinkled with water once or twice every day. Ores oxidized in this way are, of course, far superior to those oxidized by means of artificial heat. The method of roasting ore in the open air by ar- tificial heat is as follows : Billets of wood are placed, like the bars of a gridiron, upon a previously prepared level spot ; sometimes they are laid parallel, and sometimes in a crosswise manner, so as to form a uniform flat bed. The crevices between the wood may be filled with chips of wood, charcoal, turf, or even stone coal, coke, or anthracite, so as to prevent the ore from falling between the other pieces of fuel, or, what is still worse, upon the ground. The ore, before it is put upon the fuel, should be broken into pieces of uniform size, of from three to four inches in diameter ; the larger pieces to be used inside of the pile, the smaller ones for covering. When a foundation of fuel of about eight inches high is prepared, ore may be piled upon it to the height of from eighteen inches to two feet ; upon this ore is spread a layer of small charcoal, or of turf, coke, or small anthracite coal, in a uniform thickness of two inches, or one inch of fuel to one foot of ore ; then alternate beds of fuel and ore, until a sufficient height is reached. The pile, thus prepared, whether of an oblong, square, or round form, should be covered with small ore, and then should be set on fire either in the centre for which purpose one or more holes or flues are left or around the base. After the fires are properly kindled, the piles may be covered with riddlings of ore or small coal. The combustion should pro- ceed slowly, being somewhat suffocated, so that the whole mass may be uniformly penetrated with heat. Where the fire is too intense, it must be covered with small ore or coal dust, and where it is too imperfectly developed, holes should be pierced with an iron bar, that smoke and air may have vent. In all cases of calcining in heaps, the arrangement and manipu- lation are almost the same, with hardly any other variations than those arising from the difference of ore and fuel. Fig. 5 represents the cross section of an ore pile, which is so plain as to need no description. In this plan the billets of wood are raised from the 46 MANUFACTURE OF IRON. Fig. 5. Section of an ore heap ready for firing. ground, which affords the advantage of enabling us to kindle the pile wherever we choose. XVI. Gleaning of Roasted Ores. Iron ores, after being roasted, are very apt to be mixed with foreign matter. This must be separated from them. The usual method of accomplishing this, is as follows : A movable screen, made of a wooden frame, filled with iron bars from one-fourth to three-eighths of an inch in diameter, leaving one-fourth of an inch space between the bars, is put close to the ore pile. The dry ores are thrown by means of shovels against the iron bars, when the fine ores and fine dust pass through the spaces between the bars of the screen, and the coarse ore rolls before the screen to the feet of the workman. Stones and coarse foreign matter may be separated by hand ; the fine riddlings are thrown aside, or may be used for leveling the ore yard ; mixed with lime, they make an excellent mortar. A more convenient, though more complicated contrivance than the above, is the following. It is in general use. A strong wooden frame-work, made of oak scantling five inches thick, contains the screen a, Fig. 6, made in the usual way of round iron bars from one-fourth to three-eighths of an inch in diameter, separated from each other by one-fourth of an inch space. It is a kind of flat box ; the bottom b is formed of the iron rods. This box is suspended on wires, at four points c earing ; Wormwood - - 79.00 Pine or fir - - 0.45 Poplar - 0.75 Beech - 1.45 Oak - 1.53 Willow - 2.85 Maple - 3.90 Dry beech bark - - 6.00 Besides potash, a large amount of lime commonly exists in wood ashes. Lime is very favorable to the reduction of iron ores, and deserves attention. It is generally understood that the potash or soda which exists in the ashes of plants is always in an inverse proportion to the amount of lime they contain. We give, in the following analyses, a comparative view of the amount of lime in 100 parts of different vegetable ashes : Beech. Oak. Pine. Bark of oak. Carbonic acid 38.18 31.30 18.09 38.67 Sulphuric acid 1.19 0.90 3.75 0.37 Hydrochloric acid 0.85 0.62 0.00 0.04 Silicious acid (silex) 3.38 1.67 7.59 1.08 Phosphoric acid 4.77 6.27 0.90 Potash 10.45 9.43 16.80 4.33 Lime 35.66 39.95 34.67 47.78 Magnesia 5.86 7.15 4.35 0.75 Oxide of iron 1.25 0.09 11.15 manganese 3.77 2.60 2.75 6.98 The amount of ashes differs in different plants, as the following FUEL. table indicates, and varies strikingly in trees and shrubs, and in trunks and leaves. There are, in 100 parts of air-dried Oak wood ( old > I young, 0.15 -p. , , fold, 0.30 Birch wood < I young, 0.25 Blackberry, 2.60 TV A / ld > ' 15 Pine wood < I young, 0.12 T> , , fold, 0.40 Beech wood < I young, 0.37 Wheat straw, 5.20 The amount, as well as the composition of ashes, depends, in a great measure, upon the composition of the soil in which the plant grows. But if the chemical composition of the soil is not able to furnish the vital component parts of a certain genus of plants, this genus will decay, and its place will be occupied by a class more appropriate to this composition. For this reason we often see oak growing where pine has been cut, and weeds spring up where none have been sown. The ashes of a pine tree, in one place, have contained Potash - 3.66 Lime - .... 46.34 Magnesia - - - 6.77 While ashes of the same kind of pine, growing in another spot, have furnished the following result: Potash - .... 7.36 Lime - - 51.19 Magnesia ... none. e. Practical Remarks. The foregoing investigations and tables are only designed to present to the iron manufacturer a comparative view of the relative values of wood. Therefore his attention should be closely directed to the material best adapted for his purposes. We have seen that there is a great difference in the specific gravity of wood; and that the price per cord should vary in accordance with this difference. That is to say, if a cord of pine wood is worth thirty-eight cents, then a cord of oak ought to be worth sixty-six cents, because it is the real woody fibre which constitutes fuel, and it is that which produces charcoal. Besides the attention which the specific gravity of wood demands, the consideration whether wood is old or young is very important. Young wood, saplings, if properly treated, generally produce a strong hard coal ; old wood, when sound, is not inferior; but dead or decayed wood is useless for the making of charcoal, and it is imperfect fuel for any purpose. 84 MANUFACTURE OF IRON. Therefore a higher price may be paid for young than for the same kind of old wood, when other circumstances are equal. Every attention should be paid to the proper season for cutting wood. The worst time is from February until September. It should be cut and corded in October, November, December, and January; the best time is in the two latter months. Wood cut during winter, besides being ripe, will dry fast, and furnish a strong sound coal. Wood that is fresh and green is very apt to crack in charring, and produces a small porous coal, unfit for use in the blast furnace. Besides, economy recommends the use of the winter months, for then workmen are more abundant, and wood is twenty- five per cent, more valuable. The price paid per cord for cutting wood varies according to place and time. While a woodcutter in Vermont is able to make good wages at twenty-five cts. per cord, the cutter in Missouri thinks double that amount poor compensation. From twenty to twenty- five per cent, more is paid for saplings, and crooked or thinly grown timber, than for common forest timber. Tall, and tolerably strong timber, where the trees do not average less than twelve nor more than twenty-four inches in diameter, yields the most profitable results. Hardened wood, maple, sycamore, and knotty timber are more expensive than oak, beech, hickory, pine, and tall, clear timber. Hillsides are cleared with more difficulty than plains, and demand higher wages. A good woodcutter ought to average three cords a day. Some will cut more, some less, according to their industry and ability; and wages ought to be rated accordingly. A cord of wood contains 128 cubic feet; that is, the billets must be four feet long, and the cord four feet high and eight feet long. A great deal of deception is practiced by workmen, who need close watching. The most common deceptions are these: the billets too short; the cords deficient in length and height; crooked rows; piling the wood upon rocks or upon stumps; long limbs on the billets ; and piling the billets in as open a manner as possible. These deceptions are easily detected. They often amount to twenty- five or thirty per cent. Such practices should be avoided in a well- regulated business. Managers are often as much at fault as the workmen; for many of them, by making it a rule to dock the work- men rightly or wrongly, necessarily provoke resistance, and excite cupidity. An acre, of 160 square rods, contains, on an average, thirty cords of wood; sometimes more, sometimes less. It requires excellent FUEL. 85 timber to produce forty cords; and only very close timber will exceed that. The price of wood on the ground ranges from five to ten cents per cord ; and it is clear that in many cases five cents may be too much, and in other cases ten cents may be too little, for certain wood. The best timber is always the cheapest, although it commands a higher price. Where clearings are designed, the stumps ought to be cut as low as possible, the brush piled, and, when practicable, burnt before charring commences, in order that a way for hauling the wood to the pits may be opened. Hillsides, rocks, and swamps, as well as detached patches, make the wood and coal dear. There should be more than 500 cords in one coaling ; else the business would be profitable neither to the colliers nor to the master. Ashes, and their component parts, are of too little consequence to affect the price of wood ; but little economy can be observed in relation to them. II. Turf y Peat. This mineral fuel is of but little consequence to us, because there is abundance of wood and stone coal in the country ; nevertheless, we will give it a cursory notice on account of its chemical compo- sition, which, to iron workers, is not without interest. It has been found that turf is a most excellent fuel for the blacksmith's forge, as in case-hardening, tempering, and hardening steel, forging horse shoes, and particularly in welding gun barrels. For this purpose it is pressed and charred. Turf is generally found in bogs, in horizontal layers from ten to thirty feet in thickness : sometimes in the form of a blackish-brown mud ; sometimes it is a dark peaty mass, and often a combination of roots and stalks of plants; frequently the turf layers interchange with layers of sand or clay. Sea water is better adapted to the formation of turf than rain or spring water. Turf is simply dug with spades, and then dried. If too moist to be dug, the half fluid mass is piled upon a dry spot and there left until the water leaks off, and until the mass appears dry enough to be formed into square lumps in the form of bricks. In many in- stances, however, the freshly dug turf is triturated under revolving edge wheels, faced with iron plates perforated all over their surface ; through the apertures in these plates the turf is pressed till it be- comes a kind of pap ; this pap is put into a hydraulic press, and squeezed until it loses the greater part of its moisture. It is then dried and charred in suitable ovens. The charcoal made in this way deserves the notice of the artisan. 86 MANUFACTURE OF IRON. a. Ashes. The amount of ashes in turf varies greatly; and, economically considered, ashes are of considerable importance. Some specimens contain only one per cent., while others contain thirty per cent., which, in direct proportion, diminishes the value of turf. Eut it is not so much the quantity as the quality of these ashes which interests us. Their value as a fuel to the blacksmith is indicated by their chemical composition. It is a remarkable fact that, in turf ashes, we never find any carbonated minerals ; while they contain phosphates, sulphates, and chlorides. An analysis of turf ashes gave, in 100 parts, Lime - 15.25 Alumina 20.5 Oxide of iron 5.5 Silex - 41.0 Phosphate of lime - 15.0 Chloride of sodium 15.5 Sulphate of lime - 21.0 In other kind of turf, thirty-four per cent, of phosphate of lime, and six per cent, of chlorides, were found. The phosphates and chlorides have an excellent influence upon the hardening and welding of iron and steel ; and if we use turf for these purposes, we should analyti- cally investigate the composition of the ashes which it produces. Though the elements of turf ashes are beneficial to the working of bar iron and steel, it does not follow that they are equally benefi- cial in reducing iron ore; for in the blast furnace phosphates of any kind are injurious, and produce a cold-short iron. Therefore we should be very cautious when we recommend turf for the blast fur- nace. We should recommend only such kinds of turf as contain neither too many phosphates, nor too great an amount of ashes ; otherwise, we run the risk of producing bad work in the furnace. Dug turf, that is applicable for the smelting of iron, should never contain more than five per cent, of ashes. b. Chemical Analysis of Turf. The component parts of turf differ from those of wood. This difference is owing to the fact of its being decomposed woody fibre. We present an analysis of several specimens : One hundred parts of good turf contained, besides ashes, Garbon. Hydrogen, Oxygen. No. I. 5T.03 5.63 31.T6 No. II. 58.09 6.93 31.3T No. III. 57.79 6.11 30.77 FUEL. 87 We find here less oxygen, but more combustible matter, than in wood. c. Practical Remarks. Turf is a very imperfect fuel, because it generally contains too much foreign matter ; and it is too expensive where wages are high. A great deal of it is used in different parts of Europe, where cheap labor and scarcity of wood and stone coal render it more available. But in this country, there are few places where wood and stone coal cannot be had at reasonable prices, and as yet there is no prospect of turf coming into use for the manufac- ture of iron. Still, it is unquestionably useful in working steel and bar iron. In such cases, however, it should be subjected to a che- mical analysis. Turf should never be used in its raw form, but only when charred. Where its composition is shown to be favorable by chemical analysis, we need not be harassed in relation to its price, for its utility is so obvious that a liberal expenditure may be safely hazarded. The expense of turf, in comparison with that of wood or wood-charcoal, may be estimated by weight. The specific gravity of a cord of dry wood is from two to three thousand pounds; and, if we consider that air-dried wood contains from thirty to forty per cent, of water, the real amount of combustible matter in a cord is reduced from thirteen hundred to two thousand pounds. Air-dried turf always contains more or less water, and this is to be deducted before we can know its real value. The amount of water varies exceedingly, ranging from ten to forty per cent. It can be easily expelled by weighing the turf when green, then exposing it to a boiling heat (212), and again weighing it. The difference is water. According to this, a ton of air-dried turf ought to be worth as much as a cord of wood, provided the quantity of ashes in the turf is not too great, say ten per cent. This quantity can be found by weighing a piece of turf, and burning it slowly on a plate of sheet-iron, until all the carbon is expelled. This operation requires a red heat. The remainder is ashes. If turf is dug for the pur- pose of charring, it is advisable to employ a good strong turf-press. Turf, thus pressed, chars excellently, and yields a charcoal as hard again as the best sugar maple, or hickory coal. III. Fossil Coal. G.evlogy. It does not belong to our department to treat exten- sively of the geology of mineral coal. A few remarks will be Suffi- cient to explain all that is necessary to be understood. Fossil coal may be conveniently divided into three distinct classes : The upper, 88 MANUFACTURE OF IRON. or more recent geological deposit, is called brown coal, distinguished by its color, which is mostly brown, and its texture, which is that of wood slightly charred. It occupies the same geological position as fossiliferous limestone, above chalk. The second deposit of mineral coal, generally called bituminous, or stone coal, is below chalk. This coal is black, more or less of a vitreous lustre. The third class in our arrangement is anthracite coal, characterized by its great hardness, and the small amount of hydrogen it evolves. Its position is in the transition or volcanized secondary rocks. All mineral coal varies much in chemical composition, and ranges be- tween turf, and carbon that is almost pure. a. Brown Coal. The external appearance and texture of brown coal vary as much as its chemical composition. Its color varies from a light brown to a deep black. Some specimens are very friable ; others very hard. Its structure clearly shows it to be the remains of a vegetable world ; for the identical woody fibre, the form of trunks and limbs of trees, even the minutest leaves and fruit, are exhibited with striking distinctness. Coal beds resemble an irregular pile of trees, limbs, and leaves. The powder of this coal is always brown. Sometimes brown coal is called lignite, fossil wood, or bituminous wood terms which are not sufficiently distinctive. b. Water. Brown coal generally contains a large amount of water. Some specimens contain forty- three per cent., and scarcely any contain less than twenty per cent. Exposed to a dry atmo- sphere, brown coal is very apt to fall into slack, and lose a great deal of its moisture; but it never becomes entirely dry. It is thus evident that this coal constitutes a very imperfect fuel inferior even to turf, c. Ashes. The amount of ashes is less in brown coal than in turf, and varies from 1.50 to 27.2 per cent., as in Irish coal. A remarkable difference sometimes exists in the quantity of ashes yielded by the same piece of coal. Brown coal is very seldom of any use in the manufacture of iron, partly on account of its friability and its moisture, but more particularly on account of the composi- tion of its ashes. Its ashes generally abound in sulphates, or sul- phites, which impart sulphur to the iron, and make it red-short. We should, therefore, be very careful in the use of this coal in iron manufactories. We give an analysis of two different kinds of brown coal-ashes. In 100 parts of ashes there were found : FUEL. 89 Sulphate of lime 3.6 Sulphite of potash - 1.9 Sulphite of lime - 25.4 Sulphate of iron - 50.0 Sand - 19.1 Another specimen contained, in 100 parts of ashes, Sulphate of lime - 75.50 Magnesia - - - - - 2.58 Alumina - .... 11.57 Oxide of iron - - 5.78 Carbonate of potash - - 2.64 Sand - - 2.03 The amount of sulphur, as exhibited by these analyses, is so great that it is dangerous to use such coal in the manufacture of iron. If the quality of its ashes will permit its use in the manufacture of alum, it may be considered a very cheap and useful article. d. Chemical Composition. The composition of the combustible part of brown coal forms the connecting link between turf and bituminous coal. The analysis of 100 parts of this coal gives us the following result : Carbon. Hydrogen. Oxygen. Friable brown, I. 50.78 4.62 21.38 " II. 70.49 5.59 18.93 Black lignite, I. 51.70 5.25 30.37 " II. 63.29 4.89 26.24 The residue exhibits the amount of ashes. Nitrogen is frequently found in this species of coal, but it seldom amounts to 1.50 per cent. e. Bituminous Coal Pit Coal. This species of coal possesses peculiar interest, because of the immense quantity of it which exists throughout the globe, and especially in our own country. The Pittsburgh coal field, consisting entirely of this coal, is superior in magnitude to any in the known world. Besides the coal field of Pittsburgh, there are immense coal fields in Maryland, Virginia, Alabama, and Illinois, which not only rival the largest in Europe, but which will afford, in all time to come, an inexhaustible store of fuel. This species of combustible deserves the especial attention of the iron manufacturer. Its quality is generally good, its appli- cation simple, and its price, beyond comparison, the most reasona- ble of any other kind. To extend our labor to the highly interesting geological investi- 90 MANUFACTURE OF IRON. gallons which have exhausted alike the light of science and the resources of art, would lead us too far. The lover of such re- searches will find ample information in a work of great value (Statistics of Coal, by Richard Cowling Taylor, Philadelphia, lately published by J. W. Moore), which will richly repay the time oc- cupied in its perusal. Our province embraces merely a description of the material, and of its component parts. Bituminous coal is characterized by its dark black color, and highly vitreous lustre. Its powder is black. In some of this spe- cies, fibres of wood resembling soft charcoal may be distinctly seen. Some specimens contain more or less sulphur in the form of the yellow sulphuret of iron, visible by the naked eye. This is a mechanical admixture. If the quantity of this sulphuret is very large, the coal is unfit for the manufacture of iron. This coal is mostly stratified parallel with the direction of the vein, and breaks into square, almost cubical pieces. /. Water. Good pit coal contains very little water in admix- ture. Its close texture and its resinous character prevent the penetration of air or water. But if the coal is very friable, which is frequently the case with the external portion of the veins, water may exist in the crevices ; but in amount so small as scarcely to injure the quality of the coal. g. Ashes. This article deserves considerable attention, on ac- count of its influence in the blast furnace. The ashes of bitumi- nous coal are generally composed of silex and alumina ; seldom of lime, magnesia, or any other base ; and for this reason possess much interest. The amount of ashes in this coal varies from one to twenty-five per cent. ; coal which contains more than five per cent, of ashes should scarcely be used in the blast furnace. If any is used which contains a greater per centage than that, the furnace will not work well, and a great loss of iron, or the produc- tion of bad iron, is the consequence. Even in puddling furnaces, a large amount of ashes is injurious, as we shall hereafter see. For the sake of comparison, we give two analyses of ashes from this species of coal : Ashes from a species of French seal yielded of Sulphate of lime - - 80.8 Lame - 3.S Silex - 14.2 Oxide of iron - - 1.7 White agfoes of American coal yielded of S0ex - Lime Alumina 85.7 2.5 8.2 Sulphate of lime 3.6 FUEL. ' 91 We here observe the great preponderance of the electro-negative over the electro-positive elements. How far this circumstance in- terferes with the manufacture of iron will be investigated under the heads of the theory of the blast furnace, and the philosophy of manufacturing wrought iron. Chemical Composition. ; The chemical composition of the com- bustible parts of bituminous coal ranges between that of brown coal and anthracite. An analysis of 100 parts of this coal exhibit- ed the following result : Splint coal. Cannel coal. Glance coal. Carbon 70.9 Carbon 72.22 Carbon 90.10 Hydrogen 4.3 Hydrogen 3.93 Hydrogen 1.3 Oxygen 24.8 Oxygen 21.05 Oxygen 6.5 This coal generally contains from 1 to 1.5 per cent, of nitrogen ; "which, however, for our purpose, is of no consequence. The above table shows that the quantity of hydrogen and oxy- gen is less than that in woody fibre, turf, and brown coal, a circum- stance worthy of notice. We shall refer again to this subject when we come to speak of the article anthracite. h. Practical Remarks. Much that is highly interesting might be said concerning this article, but we are forced to condense our observations as closely as possible. The thickness of the coal seams, as distributed in the stratified coal measures, varies from an inch to sixty feet. Veins less than two feet thick are hardly worth work- ing. High wages would absorb nearly all the profit derived from working them ; besides, such veins seldom afford as good a quality of coal as is needed for the manufacture of iron. These veins are generally slaty and sulphurous, and, except in cases of necessity, should be rejected. Veins more than two feet thick are generally of better quality, as well as more workable. In fact, coal three feet thick can be raised at very nearly the same cost as veins of greater thickness. In an economical point of view, therefore, a thick vein presents but little advantage. i. Classification. Geologists have classified this coal from its ex- ternal appearance, without any relation to its chemical composition. The English coal-diggers distinguish four kinds, to wit: 1, cubical coal ; 2, slate or splint coal ; 3, cannel co^l ; and 4^ glance coal. Whether this classification is a correct o&e, we will not venture to say ; for our purpose, at least, it has no specific use. The only exact basis of classification is that of chemical composition. But for the sake of usage we will adopt the classification commonly presented. 92 MANUFACTURE OF IRON. 1. Cubical Coal Pittsburgh seam is black, shining, compact, and tolerably hard. It comes from the mines in almost cubical masses. The general direction of the vein is that of the cleavage. This coal cakes with facility, and on that account is valuable to the blacksmith, for it forms very readily a wall and vault around his fire. t 2. Slate, or Splint Coal, seam next above the Pittsburgh, is of a dull black color, very compact, harder than cubical coal, and mined with greater difficulty. It splits very readily, like slate, but resists cross fracture ; it separates in large, square-edged masses, and burns without coking. It is somewhat heavier than cubical coal, and frequently yields a considerable bulk of white ashes. Where it does not contain too much ashes, it is an excellent fuel for the blast furnace. 3. Cannel Coal generally lies in seams below the Pittsburgh vein. Its color is between velvet and grayish-black ; it has a resin- ous lustre. It is as hard as splint coal, kindles like pitch, and burns with a white bright flame. This coal works very clean in the mine, and scarcely soils the fingers when rubbed. It is found in Ohio and Missouri. 4. Grlance Coal very closely resembles anthracite, and is of an iron black color. Occasionally it exhibits an iridescence somewhat resembling that of tempered steel. It has a beautiful metallic lus- tre, does not soil, and its fragments are sharply edged. It forms coke with difficulty. The classification just presented is unquestionably a very imper- fect one, because it furnishes us with no marks by which the differ- ent classes are distinctly indicated. This is evident from the fact that the same vein not unfrequently contains all the varieties in- cluded in these classes. A correct classification would include all bituminous coal, so called from its resinous aggregation, and the amount of hydrogen it contains. This class may very easily be distinguished by its property of forming coke ; for wood, turf, brown coal, and anthracite do not yield this article. Jc. Mining of Coal. Where coal fields are situated above the water level, and with the advantage of ascent, the working of coal is comparatively easy. The pit water flows off by itself, and there is but little trouble in ventilation. In such cases, all that is required is, to open a drift, to timber it well at the mouth, and to make such arrangements in train or plank roads as will afford a quick and easy hauling. A very cheap and useful arrangement is that of the Pittsburgh coal-diggers. A two wheel cart, of about twelve bushels FUEL. 93 capacity, is pushed on a plank track by a man, assisted by a strong dog, "which runs before the cart. We have never found much ad- vantage in a large, high, and wide drift or level. We have paid quite as high a price for hauling by horses or mules on an iron tram- road, as that paid for the use of the above-mentioned cart. Where a coal mine is very extensive, or where the wagons have to be pulled up an ascent, a wide track, and horses and mules, may be advan- tageous ; but, considering the cost of a spacious drift, rails, wagons, &c., very little is gained in expensive improvements. A plank track is easily removed ; it may be turned in any direction, even to the very face of the work-rooms, and will last a long time, if constructed of good white oak. We have paid twenty cents for hauling one hundred bushels from the room to the mouth of the pit, and tole- rable wages were hardly made at that ; while an equal amount readily pays reasonable wages, if the above-mentioned hand-cart or dog-cart is employed. If locality, or other circumstance, does not permit an opening or drift according to the inclination of the coal, it is necessary to drive a dead level in the coal to drain the mine of water ; and in case this cannot be done, a dead level below the coal must be drifted until the coal is reached. This is illustrated by the following diagram : a, #, c, d, e, f represent coal veins, and g, a Fig. 19. Draining level. dead or water-draining level, which, of course, can be used as a win- ning level. The shaft h may be used as a ventilator of the pit, or both ventilator and winning shaft. Coal veins, situated above the water level of the country, may be 94 MANUFACTURE OF IRON. worked at but little expense, that is, require no immediate capital ; but if they are situated below the water level, more attention and greater means are required. If the coal is so low that a mine can- not be drained by a level, machinery, either water-wheels or steam- engines, as well as pumps to raise the water sufficiently high to permit its flow into the nearest river, must be resorted to. In such cases, vertical shafts are in common use. Such a shaft is constructed of timber, walled with stones or bricks, or of iron cylinders. Its dimen- sions depend entirely on the amount of coal required to be hoisted. If the section of such a pit, or shaft, is round, it should never be less than ten feet in diameter : it may increase thence to twenty feet. Such a shaft must be divided into different compartments, one of which should be always reserved for pumps and water-pipes. The following diagrams represent its various forms : a, a are designed Fig. 20. Fig. 21. Fig. 22. Sections of shafts. for the pumps. If this shaft is made of a square, instead of a round form, it should not be less than eight by ten, or ten by twelve feet for a double pit. Where the coal is not too far below the surface, say fifty or one hundred feet, inclined planes may, in some instances, be preferable to vertical pits. In such a case, the section may be smaller, and the same railroad cars that are used above ground may be used below ground. But whatever plan is adopted, the shaft should be sunk to the lowest point of the coal vein. The working of coal by means of a shaft is, in fact, not more expensive than that of more highly located veins. It is attended with some disadvan- tages to the workmen ; these are generally balanced by a good roof. But the expense of shaft ventilation, engine, and pumps falls heavily on the proprietor; this once met, the work may be prosecuted cheaply and with facility. A good circulation of fresh air is effected only at great expense, and with considerable difficulty; this circumstance needs great attention in extensive coal mines. There is no reason, at the present time, why iron masters should go to a great depth for FUEL. 95 coal. Coal above the water level is so abundant that any farther consideration of deep coal pits would be superfluous. The mode of working a coal vein depends on several circum- stances : partly on the roof, upon the kind of coal, whether for our own use or for the market, and upon the thickness of the vein. The following are the methods practiced : 1. Working with pillars and rooms, where the pillars left bear precisely that proportion to the coal excavated which is required to support the incumbent strata or roof. These pillars are gene- rally lost. 2. Working with post and stall. Here the pillars left are of a larger size than usual, and stronger than is requisite for supporting the superior strata. These are so constructed that they may be re- moved whenever the regular work is done. This method of work- ing is best adapted for coal veins more than three or four feet thick. 3. Working with post and stall, or with comparatively small rooms. By this method, an unusually large proportion of coal is left, with a view of working backward towards the starting-point, whenever the coal field is worked to the whole extent ; then by taking away every pillar completely, if possible, the roof is per- mitted to fall in, following the miners as they retreat. 4. Taking out all the coal, and leaving no pillars at all. By this plan, the roof falls in as the diggers retire. The first two methods are practiced for the thicker coal seams. Where the veins are thicker than three feet, the two latter methods are dangerous and expensive. Where the coal, roof, and pavement are of equal hardness, the first and second methods will answer; but where the pavement is soft, the pillars should be uncommonly strong to prevent the sinking of the coal. They should be equally strong where the coal is soft, for otherwise they would be crushed, and the coal lost. The same principle may be extended to a roof that is soft and brittle. Bearing all these circumstances in mind, it may be stated, gene- rally, that, where the coal, roof, and pavement are strong, all the above methods may answer ; but where they are soft, strong pillars and rooms of moderate size are required : in this way, when the miners retreat to the starting-point, the greater part of the coal may be got out. The proportion of coal taken out to that left in the pillars, when it is our intention to remove all the coal at the first working varies O" from four-fifths to two-thirds. A loss even of that amount throughout 96 MANUFACTURE OF IRON. the whole area of the coal field, ought to be prevented. If no acci- dents happen, this can be done by adopting the third plan. When a coal field is opened, and a systematical method of working is resolved upon, we should divide the coal field into square spaces, where pillars, rooms, and roads are properly laid out. In accordance with this plan must the air pits be situated, and a system of venti- lation arranged which will secure both the safety of the working men, and the progress of the operation. Where the coal is soft and friable, particularly where it is slaty and sulphurous, perfect ventilation is indispensable. Hard, clean coal is not so dangerous, and requires, therefore, less care. Coal pits ought to be opened in summer, and continued during winter ; an air shaft should be driven during the winter, that a progressive work for the warm season may be secured. If the coal stands edgewise, or nearly perpendicular, the thickest stratum of rock is the best place for driving a shaft. The pit should be strongly timbered or walled, to prevent its being crushed. When- ever the shaft has its proper depth, galleries must be driven across all the coal strata, as shown in Fig. 23. These galleries can be Fig. 23. Opening of galleries. multiplied for the greater convenience of the winning. All the coal is then taken out at the working shaft c. For lifting coal in a shaft, chains or ropes are used ; the former are dangerous, and often unexpectedly break. Hemp ropes are FUEL. 97 more safe, but they are expensive. We doubt whether wire ropes will answer the purpose ; for, besides the friction of coal, sand, and mud, the pit water is very destructive. For these reasons, hemp or inanilla ropes are probably the cheapest as well as the best. If coal veins are not horizontal or vertical, the best plan is to follow with the shaft the dipping of the coal, and hoist on an in- clined plane. All the other arrangements, as pumps, &c., should be constructed in accordance with this principle. The price of digging coal varies much. In Frostburg, Maryland, a ton of coal is dug in the thick or twelve feet vein at twenty-five cents; in thinner veins, in the same region, at from fifty to seventy- five cents ; sometimes as high even as a dollar is paid for a ton of twenty-three bushels. In the Pittsburgh vein, the price varies from one cent and a half to two cents per bushel, and a vein of the same thickness, twelve feet, can be dug in the counties of Armstrong and Westmoreland, Indiana, at one cent a bushel. Great difference in the price is occasioned by the difference in the quality of coal whether it is designed for our own, or for market use. If the coal is to be screened, and the slack removed, workmen demand, and of course deserve, higher wages, than when mixed coal and slack are received. Wages also depend, in a great measure, on the quality, softness or hardness of coal, upon the thickness of the vein, and upon the roof and pavement. Workmen may make good wages at one cent a bushel in one place, and poor wages at five cents a bushel in another place. As a general rule, a six feet vein, with a strong, hard roof and pavement, and a strong coal, with soft undermining, is, of all others, the most favorable. The hauling of coal from the work rooms to the mouth of the pits is a matter of great import- ance, for imperfect roads, wagons, water, &c., bear heavily upon the transportation of coal. In one case, ten cents a ton may be ample remuneration ; while, in another case, respectable wages cannot be made at thirty cents a ton. Sometimes one set of hands contract both for hauling and digging. This is a good arrange- ment ; but, where the coal mines are extensive, it is not practicable. The most prevailing method of valuing coal, as well in trade as in digging, is by measurement. Any intelligent man must be con- vinced that this is a very imperfect method of valuation. The value of coal can be deduced from its specific gravity alone ; and there- fore depends upon its absolute weight. A proper deduction must of course be made for the ashes it contains. The specific gravity of coal varies, sometimes in the same vein, from 1.2 to 1.9 a differ- 7 98 MANUFACTURE OF IRON. ence of thirty per cent. That is to say, a given quantity of coal may furnish just thirty per cent, more combustible matter than ano- ther equal quantity in the same vein. Sooner or later, measurement by weight will be generally introduced in the coal trade. This will benefit the producer no less than the consumer. Whether a ton is assumed to be 2000 pounds, or 2240 pounds ; whether this or that standard of measurement by weight be adopted, it is certain that uniformity of estimation would soon settle the real value of coal. In our case, this method would be even of more consequence than to the public and the trade generally. In England, coal is sold by the chaldron; in Germany and France, by weight; in the United States, by almost every variety of weight and measure. For what reason is coal sold in Boston and New York by the ton, chaldron, and bushel ? Why is anthracite sold in Baltimore by the ton, and bituminous coal all over the West by the bushel ? The State of Pennsylvania charges toll by the 1000 pounds on its own works ; while the workmen dig mostly by measure, and the proprietors sell either by the ton or bushel. How complicated and troublesome is such an arrangement ! Some of the Eastern States recognize a ton of coal as 2000 pounds ; others as 2240. Nova Scotia coal is sold at Boston by the chaldron. Some estimate the chaldron at 2928, and others at 3000 pounds. In New York, a ton is esti- mated at 2000. pounds; in Philadelphia, at 2240; while in Pitts- burgh, coal is sold by the bushel. All United States customs are regulated at 2240 pounds. The absurdity of buying by measure will appear still more ob- vious, if we consider that coal assumes a greater bulk when it falls into slack, than when in coarse lumps ; and that wet coal is not so heavy as dry coal. A bushel of dry coal, for instance, will weigh eighty-five pounds ; but the same coal, when wet, will weigh only eighty pounds. This difference increases when coal is slaty and brittle. The buyer is the one who suffers most by measure- ment. It may be said that three bushels of coarse coal make four bushels of small coal, and, when wetted, five bushels. The only reason which may be assigned for the existence of this absurd habit of measuring coal, is the trouble which the erecting and controlling of scales occasion ; but this difficulty may be effec- tually obviated, if self-registering scales are placed at the mouth of the coal pit, or at any convenient place. The amount of bituminous coal throughout the United States is FUEL. 99 immense. We shall speak of this subject under the following ar- ticle : I. Anthracite. The application of this mineral fuel to the manu- facture of iron is of very recent date. After many unsuccessful trials and difficulties, which at one time seemed insurmountable, Pennsylvania enterprise and perseverance were crowned with suc- cess. Prof. W. R. Johnson, in his "Notes on the Use of Anthra- cite in the Manufacture of Iron," gives a very interesting account of these difficulties, and of the success with which anthracite has been applied in the blast furnace. Water. Anthracite is too compact and hard to absorb water, or to contain it in admixture. Ashes. The amount of ashes is often very considerable, varying from one to thirty per cent. Their chemical composition is princi- pally silex, with but little alumina, and sometimes oxide of iron. Chemical Composition. The composition of anthracite resem- bles very closely that of charcoal and coke. Hydrogen and oxygen, gradually diminishing, amount, in the most perfect specimens, to scarcely anything. We present an analysis of anthracite : Pennsylvania. South Wales. Massachusetts (Worcester). Carbon 94.89 Carbon 94.05 Carbon 28.35 Hydrogen 2.55 Hydrogen 3.38 Hydrogen 0.92 Oxygen 2.56 Oxygen 2.5T Oxygen 2.15 Ashes 68.65 Practical Remarks. Anthracite which contains more than five per cent, of ashes is of no use in the blast furnace, and of but little use in the puddling and re-heating furnaces. But good anthracite is undoubtedly the most perfect of all fuels for the manufacture of iron. Its application is simple; its hardness prevents it from falling into slack ; and the small amount of hydrogen it contains makes it advantageous for the blast furnace operation. By proper applica- tion, anthracite will supersede, in economy, bituminous coal. The mining operations of anthracite are more simple than those of bitu- minous coal. Less danger is to be apprehended from the effect of bad air or coal-damp ; therefore less expenditure for ventilation, as well as for pavement, roof, and coal. It is so hard that the last remains of coal can be removed. A ton of anthracite, of 2240 pounds, sells at present, in Philadelphia, at from three dollars and eighty-five cents to four dollars ; in New York, at from five 100 MANUFACTURE OF IRON. dollars and fifty cents to six dollars ; and in Boston, at from six dollars and fifty cents to seven dollars. These are market prices, on which large manufacturers generally receive a discount of from five to ten per cent. m. G-eneral Remarks on Fuel. Wood is at present so abundant throughout the United States, that charcoal furnaces and forges may Tbe carried on for a great length of time, without apparently dimi- nishing its quantity; still, so rapidly are civilization and wealth pro- gressing, so rapidly is the consumption of iron augmenting, that our attention cannot be otherwise than forcibly turned to mineral fuel as a substitute for wood in the manufacture of iron. Peat or turf is not sufficiently distributed to deserve any attention. There are peat bogs in the States of New York, Michigan, Rhode Island, New Hampshire, and Maine, and possibly in other States; but its application in our country is so limited, and will probably, for some time to come, be so limited, that it hardly deserves our notice. In France, Germany, Bohemia, and Russia, peat is used for the manu- facture of iron; but here, independent of any other cause, the price of turf would prevent its application for this purpose. In countries where no mineral coal exists, its application may be ad- vantageous. The reasons we have given against the use of turf apply equally well against the use of brown coal. Some kinds of lignite constitute a good fuel in the puddling furnace, as well as in re-heating and sheet ovens; but their application in the blastfurnace is very limited. Lignites found in the United States are very pro- perly used only for the manufacture of alum and copperas. n. Of all the coal deposits, those of anthracite and bituminous character deserve our closest attention. Their utility in the manu- facture of iron, and their extraordinary magnitude throughout the United States, give to the iron business of this country prospects the most flattering of those of any nation, or of any time. England was supposed to include, until recently, the great coal deposits -of the world; but these shrink into insignificance when compared with the gigantic deposits of the United States. The amount of coal distributed throughout the world is as follows : United States of America - 133,132 square miles. Anthracite of Pennsylvania 437 " " British America - 18,000 " " Great Britain - 8,139 " " u and Ireland (anthracite) 3,720 " " FUEL. 101 3,408 square miles. 1,719 " " 518 " " Spain France Belgium How great the prospect, how extensive and durable the basis of comfort, prosperity, and happiness, to the citizens of the United States, this immense wealth of mineral fuel discloses ! The dis- tribution of coal throughout the different States of the Union is as follows : Alabama - 3,400 square miles. Georgia - 150 " " Tennessee 4,300 " " Kentucky- 13,500 " Virginia - 21,195 " Maryland- 550 " " Ohio 11,900 " " Indiana - 7,700 " " Illinois - 44,000 (?) " " Pennsylvania - 15,437 " " Michigan 5,000 " " Missouri - - 6,000 " " By this table, we find that England, Ireland, Scotland, and Wales united, do not contain so much coal as the State of Ohio. We have omitted, in the above estimate, the smaller coal tracts in different States, as not worth mentioning. United States Goal Measurement. Ordinary estimate of bituminous coal: 28 bushels = 1 ton at 2240 Ibs. At some places, 30 " " " We also find it stated at 26J " " " At Richmond, Virginia, coal pits, a bushel = 5 pecks = 90 Ibs. The same coal on board a vessel " =4 " =72 " In the South, bituminous coal is sold by the barrel, weighing 172| = 13 barrels = 1 ton. In New York, as well as in Boston, and elsewhere in the East, a ton of coal = 2000 ii>s. On the State Canal u d Tidewater Canal, Pa., toll is levied at 1000 Ibs. In Pennsylvania, Ohio, and several other States, a bushel = 80 Ibs. of coal. Nova Scotia coal is sold by the chaldron = 3360 Ibs., or 42 bushels. In Boston, the retail chaldron is but 2500 or 2700 Ibs. 102 MANUFACTURE OF IRON. Prices of coal at the coal pits : In France - - $1 50 to $3 50 per ton. Germany 1 75 2 50 " England 1 50 " 2 50 " " Pennsylvania (anthracite) 2 00 " 2 25 " " " Pittsburgh (bituminous) - 50 " 1 00 " " IV. Distillation of Fuel. If raw fuel is inclosed, with exclusion of atmospheric air, in an iron or any other retort, and if to this fuel we apply heat, a decompo- sition ensues, and the result of such decomposition varies accord- ing to the kind of matter with which the retort was charged. From the moment heat is applied, the elements of the matter separate, and, according to the temperature, form new compounds, which did not previously exist in the raw material. If we charge the retort with wood, the first compound which escapes is water ; this existed in the wood either in the form of sap, that is, water combined with soluble matter, or as hygroscopic water, attracted and retained by the porous aggregation of the wood. Hydrogen and oxygen are then expelled, and form chiefly water, and partly other composi- tions. A small amount of oxygen and hydrogen is left, to form, by an increasing temperature, different compounds with carbon. Hy- drogen combines with a small amount of carbon, to form carburetted hydrogen (the fire-damp of the coal pits) ; after that, a mixture of a great many compounds, consisting of carburetted hydrogen, tar, acetic acid, messit, or wood alcohol, creasote, naphtha, &c., is dis- tilled, which can be condensed from its gaseous into a liquid state, by introducing it into a cold receiver. The same law which governs the distillation of wood, is applied to the distillation of turf, brown coal, bituminous coal, &c., with this difference in the product, that those minerals which contain the least water, hydrogen, and oxygen, will leave the greatest amount of carbon, inasmuch as carbon can- not be evaporated, without access of another element with which it unites. After distillation, a certain amount of carbon is left, ac- cording to the preponderating quantity of the elements. If there should be a great deal of water, hydrogen, and oxygen in the raw material, a small amount of carbon will be left ; should there be a large proportion of carbon in the fuel, a large amount of charcoal will remain. Wood, turf, and brown coal generally leave a char- coal in precisely the form in which the pieces were put into the retort ; their porous structure permits the evolution of the gases FUEL. 103 without disturbing their form. Bituminous coal is so close, and its aggregates so compact, that the escape of any matter from the interior is impossible ; its bulk, therefore, is increased by the ap- plication of heat ; the hydrogen and oxygen which escape form small cells, and the remaining carbon is spongy. The presence of water and ashes, as well as of hydrogen and oxygen to a large amount, in fuel, is detrimental to iron, even in the pud- dling and re-heating furnaces; and still more injurious in the blast furnace, as we shall hereafter see. To avoid these influences, at ledst in the blast furnace, we must have recourse to the charring of the fuel ; by distilling it, we get rid of the injurious admixtures. V. Charring of Wood. If a piece of wood is heated until it kindles, a flame issues, which is nourished by the decomposition of the wood ; this decomposition is the result of heat, and is continued by the heat produced by the flame itself. Within the flame is a dark body, carbon, which does not burn until all the hydrogen which protects it is consumed; and only after this protection ceases, and after the oxygen of the atmo- sphere finds access to the ignited carbon, is the carbon consumed, when it disappears gradually, leaving more or less incombustible matter, i. e. white ashes. If the access of atmospheric air is pre- vented after the hydrogen and oxygen are expelled or consumed, a black coal, charcoal, will remain. This experiment can be made in a simple way. If we take a long chip of wood, and hold the flame high until it is properly kindled, and then turn the flame sud- denly downwards into a narrow tube with a bottom, the wood burns only above the neck of the tube, and the part which is in the tube is extinguished, leaving a black charcoal. This is, in the main, the principle of charring : the gaseous matter of the wood is kindled ; its water driven off; hydrogen, oxygen, and a little carbon yield the heat by which they are expelled ; the access of atmospheric air is then excluded, and charcoal remains. There are various modes of charring wood, differing principally in arrangement and manipulation. Scientifically, there are but two methods : that is, producing the heat for charring from the material to be charred ; and applying exterior heat, by means of extra fuel, in the manner we employ it in distillation. We shall describe the different modes in historical order, and shall dwell mainly upon the most practical. a. The most ancient way of making charcoal is, simply, to dig a 104 MANUFACTURE OF IRON. hole in some dry place ; to fill it with wood, and to burn a part of the wood until sufficient heat is produced to char it thoroughly ; the wood is then covered with sod, or sand, or coal dust, to keep the air out ; and charcoal will remain in the pit. The proper time of throwing on the covering is a matter of practical importance. This mode of charring is very imperfect, and, at present, is prac- ticed only among uncultivated nations ; it makes bad, light fuel, and furnishes it only in small quantity. b. Charring in Heaps, Kilns. To build a kiln or heap, a dry, sandy, and, when possible, level spot, in the woods, is selected, protected from wind and gales, and as close to the cord wood as possible ; the earth is to be leveled, dug, and tilled, to remove stones and stumps, and sufficiently large to permit the building of a heap of thirty feet in diameter. A circular space of from forty to fifty feet diameter will thus be required. Great care should be taken that no spring, or water of any kind, is in the neighborhood of the level, and that sudden gusts of rain shall not overflow it. When the ground is leveled, and pounded as solidly as possible, the haul- ing of cord wood should be commenced ; this wood must be piled vertically around the circumference of the hearth or level, as re- presented in Fig. 24. The opening , sufficiently large for a sled to enter, is left on the most convenient part of the ground, ac- cessible by horses or oxen. Fig. 24. Hauling the wood to the hearth. After all the wood is put around the hearth, the heaviest billets always to be placed inside, the collier puts either one stick, or, what is better, three stout sticks of about ten feet in length, right in the centre of the hearth, fixes them firmly in the ground, and then fastens or binds them with withes, so as to form a chimney or draft-hole in the centre of the hearth, as seen in Fig. 25. The collier then commences to set his kiln, or pit, as it is some- times called, by ranging the heaviest billets around the centre, at first vertically, and then gradually in a slanting position by turning the butt, or thickest end of the wood, downwards. The wood should FUEL. 105 be put as closely together as possible, that it may be brought into the narrowest space. After it is all set, the pile (which may amount Setting the wood for charring. at the first burning to twenty or twenty-two cords, and may be afterwards increased to forty, even fifty cords, according to the skill of the collier and quality of the dust) is covered with small limbs, cut short, and with chips, &c., to fill the crevices ; in fact, the pile must be made as smooth and tight as possible. The whole kiln must then be covered with leaves, about two inches thick. Some colliers are in the habit of firing the kiln as soon as the leaves are thrown on ; but this practice is a dangerous one, for, should a gale hap- pen to blow, or should it become windy before the workmen are able to cover the leaves with dust, a quantity of wood and coal is lost. In addition to this, very dexterous workmen are required to manage such a proceeding advantageously. The best way to proceed is, to cover the leaves with a thin layer of dust before fire is put to the kiln. The first layer of dust should be pure earth, and somewhat short, sandy, and by no means tough or clayish, for in the latter case it will shrink, bake, and crack. Besides, it is almost impossible to keep it tight. Where no other dust than this can be found, it is necessary to take sod, which, of course, is somewhat expensive. The first dust is always imperfect, but gets better with the progress of the work ; for, after the first burning and drawing of coal, it is properly made by mixing the green dust with charcoal dust. This mixture forms a light open cover, admitting frequently a thickness of eight inches, and is of course the most secure and profitable. Where coke dust can be had at a reasonable price, it is the most available of all dusts. After the whole kiln, with the exception of a few feet on the top, is covered by a slight coating, the fire may be put in 106 MANUFACTURE OF IRON. either by means of chips, brands, or good charcoal. It should be kindled at the bottom of the centre, at a, Fig. 25, and must be watched until it has fairly started; the kiln may then be left alone for at least twelve hours, provided the weather is not so stormy as to blow the cover off. Within twelve or eighteen hours, the kiln will be sufficiently heated to permit the application of more dust, and the closing of the top ; but this depends greatly on the skill of the collier, and the kind of wood to be charred. Pine and dry wood will permit more time; but young wood, saplings, hickory, and maple, require closing very soon, or the kiln will get too hot, and a great deal of coal will thus be wasted. Three days, or, still better, four days after the fire is started, the cover gradually sinks, and this is the time that the hands should watch closely; for, should the heat not have been regularly distributed inside the kiln, the setting or sinking will be irregular, and the cover be very apt to leave openings, where the atmosphere, penetrating, would do great damage, in case such breaks are not directly closed. If the work has been well conducted, the cover sinks gradually and regularly, and the dust can be gradually increased, as the smoke ceases. After the smoke altogether ceases, the whole kiln may be closely covered, and left for cooling. Four or five days are sufficient to cool the kiln. As it is cooling, the drawing of coal commences. This is performed by raking the dust off from the coal at the foot of the kiln. We should be cautious not to open too much at once, for fresh coal, even when black and cold, is very apt to rekindle, and Fig. 26. Making charcoal in heaps. FUEL. 107 thus occasion a loss of coal and time. If the charcoal is sufficiently cold, and does not kindle, the drawing may be continued all around the kiln, but not to a greater extent than is needed to fill a wagon load ; the kiln is then to be carefully covered, and left until the next wagon is ready to take a load. Sometimes it will happen that the top sinks, or that the fire is not properly kindled. In this case, the workmen must get on the very top of the kiln. That this may be done conveniently, a rough stairs or ladder is made of a round stick six or eight inches thick, as shown in Fig. 26. In case the cover is too heavy, or in case the fire draws to one side a circumstance which, in stormy weather, frequently happens the hot places should receive more covering ; and in the cold parts, holes should be opened by removing the dust. If the kiln is too cold at the base, holes should be made all around, that the fire may be drawn towards the bottom. The foregoing description of charring applies to the standing kiln. It is a mode of working very much in use, and, by care and attention, furnishes good results. There is, however, one objection to it ; that is, if the wood is very cylindrical, or if one end is as thick as the other, many spaces will be left which cannot be filled. These are injurious to the final result ; for the circumference of the kiln will require to be sloped to make the dust stick; and to "make that slope the wood must be drawn outward at the base. To avoid such spaces between the billets, another method is frequently adopted : that is, setting the wood around the centre post verti- cally, and the last four feet at the circumference horizontally, as represented in Fig. 27 ; but the advantages afforded by this method Fig. 27. Section of a charcoal work piling the wood. 108 MANUFACTURE OF IRON. are not great ; and it is of but little consequence which method is pursued. Care and attention on the part of the workmen are the guarantee of success. The time required to finish a burning dif- fers, and depends on the season, weather, wood, dust, capacity of the workmen, and other circumstances. Winter is always a bad season for charring. Stormy, rough winds are equally unfavorable ; green wood fur- nishes a poor yield, and bad coal; green or heavy dust is disad- vantageous ; light dust is equally so. Colliers who do not under- stand their business, or who are not industrious and attentive, never make good coal, nor produce a good yield. If the work is unreasonably hurried, or if the teams are not always ready at the proper time, unavoidable losses are the consequence. c. Charring in Mounds. One of the most ancient modes of char- ring wood is by mounds. This method is practiced to a great ex-^ tent in the pine forests of Austria ; and for pine, or well-seasoned hard wood, deserves our notice. A mound is built in the following manner : Fig. 28. A row of posts, a, is firmly fastened into the Fig. 28. Wood-charring mound. ground. The ground should be previously leveled or sloped. A second row 5, parallel with the first, and well secured, must then be similarly placed ; the distance between the two rows is, by four feet long cord wood, eight feet eight inches. The length of such a row will depend upon the dimensions of the mound, say from forty to fifty feet, and the posts ought to be no more than four feet distant from each other ; the post c will be six feet, and d only three feet above ground. Three head posts, 0, are then put in, and the whole inside of the frame is lined with boards, lath, slabs, or even with split cord wood. This lining is to be fastened to the posts by wooden pins. After the frame is finished, the lining fastened, and the floor FUEL. 109 pounded solidly down, the setting is commenced, by throwing the wood crosswise on the floor; and, beginning at e, the woo$ is gradually piled to within three inches of the top of the lining. By packing the wood closely, a mound fifty feet in length, six feet in height at the head, and three at the foot, and eight feet eight inches wide, ought to take from twelve to fourteen cords of wood. The wood, if not too heavy, that is, if not more than twelve or fifteen inches in diameter, may be round ; and if straight, eight feet in length. In that case, it packs more closely than split wood. By laying the wood crosswise throughout the length of the mound, the advantage of fast work is secured, for the coal can then be drawn gradually as the fire retires. After the wood is all in, and the top leveled with limbs or chips, a dust, sod, or sand cover is thrown over the whole ; the sides, where a space all around is to be left, are equally filled between the lining and the wood with dust, which must be firmly pounded in, to prevent the lining from catch- ing fire. When the whole mound is covered, fire is kindled at the lower end/. In the mean time, some dust is removed at the top, near c. After the lapse of a few hours, the fire will be sufficiently advanced to permit the egress of smoke at e\ and the dust at/ may be thrown on again. To secure a supply of fresh air, one of the draft-holes g, g, on each side, may be opened. By this method, the charring proceeds rapidly, and requires watching. When the fire has advanced about ten feet, which will require a period of twenty- four hours, coal may be drawn at /, and continually drawn until close to the fire. This latter advantage is mainly due to the sloping of the wood pile ; and it is still greater if the pile is put on a gently sloping hillside. This process of making charcoal is a very ancient one. It was employed by the Romans to manufacture charcoal for their forges ; and at the present day, in the very country where the ancient Romans carried on their iron establishments, this mode of charring is preferred to all others. The process now carried on in southern Austria differs not in the least from that in use at the time of Pliny. It is one of those inventions which a com- prehensive genius finished, and left to posterity nothing but the task of explaining ; it is so simple, practical, and complete, that no im- provement is possible. How it happens that we, at the present day, in more cultivated countries, reflect so little upon this most simple and perfect mode of charring wood, is more than we are 110 MANUFACTURE OF IRON. able to explain. We have both observed and practiced it, and have found it to answer all the claims of the iron master. It delivers a strong coal, and yields well. It combines the advantages of the pit, the kiln, and the char-oven. At the close of this chapter, we shall elucidate this subject more fully. d. Charring in Ovens. The inducements to invent ovens for charring wood, coal, lignite, and turf, were, partly a desire of gaining the gaseous educts of the charring process, and partly a desire of security against wind and storms. The first object is only imper- fectly realized ; because, if the fuel for distillation is derived from the material to be charred, the combinations of hydrogen and carbon are mostly destroyed, while nothing but tar and heavy carbonized com- pounds are the result of the distillation. Even this is to be gained only by impairing the quality of the charcoal. The latter object is available ; a well-built oven affords perfect security against wind and storm ; and where the transport of wood to a general charring place is not too high, great advantages may be derived from it. Of the various improvements adapted to the principle of charring in ovens, the French have furnished the largest proportion. They have been compelled, by the scarcity of fuel in France, to direct their experi- ments with the especial object of economizing it. The building of stationary char-ovens has been brought to a state of perfection. With good bricklayers, all that is required to build a good oven is close joints and good sound brick. The form of these ovens has been varied to suit almost every notion ; still it is generally agreed that ovens of less capacity than fifty, and of greater capacity than sixty cords, are less advantageous than those constructed within these limits. A further distinction is made between ovens for making black, and those for making brown, charcoal. The first are heated with their own, the latter by additional, fuel. A description of the Wood-charring oven. FUEL. Ill Section of wood charring oven. various forms of these ovens would occupy too much space ; there- fore, we shall notice simply the latest improved oven of each kind. Fig. 29 represents a char-oven for wood, now in use near Baltimore. A is a side elevation, showing the binders a, a, a, made of cast iron; these stays or binders may be made of eight inch timber, hewn on two sides ; but iron has a much better appearance. The cross binders 5, 5, 5, b are made of one inch square wrought iron bar, either with head and screw, or with head and key ; either will answer the purpose. The distance between two binders should not be more than four feet, and if possible less than that ; S-o II |1 i- C w^ "O s ^1 cs! g 5 S o o {/2 C, d. III Anthracite. Beaver Meadow, Pa. - 1.6 2.38 88.9 7.11 10.4 66.8 Forest Improvement, Pa. - 1.4 3.07 90.7 4.41 10.8 69.4 1.5 5.28 89.1 5.56 9.6 61.7 Lakawana, Pa. - - - - 1.4 3.91 87.7 9.25 10.7 68.8 Coke. Midlothian, Va. - 16.55 10.3 66.2 Cumberland, Md. ... 13.34 10.3 66.2 " Mining Company, Md. 12.40 11.2 72.0 Bituminous Coal. Maryland Mining Company, Md. 1.4 12.31 73.5 12.40 11.2 72.0 Cumberland, Md. ... 1.3 15.52 74.29 9.30 11.0 70.7 Blossburg, Pa. .... 1.3 14.78 73.41 10.77 10.9 70.0 Karthaus, Pa. .... 1.2 19.53 73.77 7.0 9.8 63.0 Cambria Co , Pa. ... 1.4 20.52 69.37 9.15 10.2 65.5 Clover Hill, Va. ... 1.2 32.21 56.83 10.43 8.5 54.6 Tippecanoe, Va. ... 1.3 34.54 64.62 9.37 8.5 54.6 Pictou, Nova Scotia ... 1.3 27.83 65.98 13.39 9.7 62.3 Liverpool 1.2 39.69 54.90 4.62 8.2 52.7 Scotch 1.5 39.19 48.81 9.34 7.7 49.6 Pittsburgh, Pa. - 1.2 36.76 54.97 7.07 8.9 57.2 Dry pine wood - ... 0.3 4.7 30.2 The data in this table have been deduced from direct experi- ments made on the steam boiler, and therefore are only measurably applicable to our case. In the iron manufacturing apparatus, heat is generated and escapes at a far higher temperature than that of a steam-boiler. The above-mentioned law, that the heat, liberated by combus- tion, is in direct proportion to the oxygen consumed, is one of the most useful inculcations of chemistry. By this law, we are enabled, if we know the composition of the fuel, to calculate the amount of FUEL. 135 heat it liberates. The composition of fuel has been given at the proper place. It is only necessary to present here the formula which will enable any one to calculate the quantity of heat evolved by any process of combustion. This law has been applied by Berthier to ascertain the quantity of heat liberated by any fuel in combin- ing with oxygen. The process of accomplishing this is as follows: The combustible, properly dried, is pounded into an impalpable powder. A given amount of this powder, say fifty grains, is intimately mixed with forty times its weight of litharge, and then placed in a good Hessian crucible. The whole is to be covered with a layer of litharge, to prevent the atmospheric air from penetrating into the mixture. The crucible is carefully placed in an air furnace or common stove. No particles of coal from the fire should be suffered to fall into it. In fifteen or twenty minutes, the crucible will be red hot, and may be removed from the fire, or, what is still better, left until the litharge on the top is in complete fusion. After cooling the crucible, we find at the bottom a button of metallic lead. This must be weighed, for it is the true standard of the oxygen consumed by the fuel. One part by weight of carbon represents 2.66 parts of oxygen, which, taken from litharge, represents 34.5 parts of lead. This amount of oxygen, carbon, or lead, will heat 78.15 parts of water from 32 to 212; so that every unit of lead represents 78.15 3^-5= 2.265 units of water, heated from 32 to 212. One part of carbon combines with 2.66 oxygen to form carbonic acid; and one part of hydrogen combines with eight parts of oxygen. If, from the sum total of oxygen, we now subtract the amount of oxygen which the fuel contained, we shall find the amount needed for combustion. This latter is the measure of the power of heat of the fuel. For example, oak wood is composed of 0.4943 carbon, 0.0607 hydrogen, which takes 0.4943 x 2.66+0.0607 x 8= 1.318 + 0.485=1.803 oxygen; subtract the oxygen of the wood= 0.445, then 1.803 0.455=1.358 oxygen is left. This is equal to 17.58 lead or 39.8 water, which by one part of oak wood can be heated from 32 to 212. In this way the value of fuel, or the quantity of heat it liberates, may be very simply ascertained. But the iron manufacturer must pay particular attention also to the quality of heat, of which we shall speak hereafter. In accordance with this method, the following experiment has been made. The water is assumed to be heated from 32 to 212 Fahr. 136 MANUFACTURE OF IRON. 1 pound of bituminous coal will heat 60 pounds of water. 1 " pure carbon " 78 " 1 " charcoal " 75 " 1 " dry wood " 36 " 1 " air-dried wood " 27 " 1 " turf " 25 to 30 " 1 " alcohol " 67J- " 1 " oil, wax " 90 to 95 " 1 " ether " 80 " 1 " hydrogen " 236 " These substances naturally combine with various amounts of oxygen. Assuming oxygen to be one, and the water to be heated from 32 to 212, then 1 pound of oxygen, combining with hydrogen, will heat 29 J pounds. 1 " " carbon, " 29J " 1 " " alcohol, " 28 " 1 " " " ether, " 28 b. Quality of Heat. When a combustible combines with oxygen, that is, burns, the resulting compound contains all the heat liberated in the process. From this compound the heat is abstracted by other substances which come in contact with it. How high the tempera- ture in such cases of combustion will be, may be approximately calculated. The following demonstration is designed rather to give a clear insight into the process of combustion than a real calcula- tion of the quality of heat evolved. If one part of oxygen by weight combines with hydrogen, it forms 1.152 water, or steam of a high temperature. If gaseous water had the same capacity for latent heat as condensed water, one pound of oxygen would raise the temperature of 29J pounds of water from 32 to 212, and the temperature of the formed water would be ' 18Q = 32+4752. But the capacity of steam 1.1.25 for caloric is only 0.8407, therefore less than water; and the tem- perature of steam will be, in the moment of generation, = ^ = 1.19 times higher, or 4752 . 1.9=5654. Let us take, instead of pure oxygen, atmospheric air ; then 23.1 parts oxygen are mixed with 76.9 nitrogen, if we neglect the other compounds of atmospheric air. The nitrogen absorbs a part of the heat produced by the forming of water ; and as the capacity of nitrogen FUEL. 137 for caloric is 0.2734, or nearly one-third of that of steam, the temperature of the oxy-hydrogen flame, nourished by atmospheric air, will be only half as high as it would be if nourished by pure oxygen; that is to say, 76 parts of nitrogen will absorb as much heat as the steam formed by 23 parts of oxygen, because 23.1 of oxygen form 25.95 water, little more than one-third of the nitrogen. Inasmuch as nitrogen has only one-third the capacity of steam for heat, the temperature will be reduced to one-half of that of the pure oxy-hydrogen flame, or 2827. The foregoing is mainly de- signed to give a comprehensive insight into the process of combus- tion, and the degree of heat evolved. If we apply this simple rule to practice, we shall soon ascertain why some kinds of fuel pro- duce so low a temperature, and why wet or green wood does not produce the same degree of heat as dry wood. This calculation is easily applied to any fuel with whose che- mical composition we are familiar. It also furnishes us with a comparative idea of the temperatures produced in certain processes of combustion. In the heat of the hydrogen flame, burning in atmospheric air, we melt thin platinum wire; therefore, the melting heat of platinum cannot be more than 28.27. The temperature required for melting iron is far lower than that required for melting platinum ; from this fact, we may conclude that the heat in a blast furnace can never be higher than that by which platinum melts. XI. Analysis of Fuel. A perfect chemical analysis of fuel is, for our purpose, unneces- sary ; but an approximate analysis may, in some cases, be useful, and in other cases, prevent great evils. If we know the amount as well as the quality of the ashes, and where necessary, the amount of bitumen, contained in fuel, we may consider ourselves sufficiently safe for some enterprises. To analyze fuel, let us find the amount of water it contains, by exposing it fora time to such a heat as will not char nor kindle it. It should be weighed in its raw state, and after it is dried. If, after weighing a pound of green wood, or raw coal, we put it on the top of a puddling furnace, or on the arch of a hot air stove, leave it there one or two days, and then again weigh, we shall perceive a loss in weight ; this loss will indicate the amount of water which the fuel contains. Ashes are the residue of the combustion of fuel. To obtain them, it is best to break the specimen into small fragments; to put these fragments into an iron vessel without a cover ; then to put this 138 MANUFACTURE OF IRON. vessel over a fire. In this way, the fuel will burn very slowly. The contents of the vessel should not be stirred, for the ashes are apt to retain carbon, even though a high heat, with abundance of air, is applied. Slow combustion and low temperature are the surest means of avoiding this evil. The remains of such a com- bustion, properly calculated, will give the per centage of ashes contained in the fuel. The quality of the ashes it is of but little consequence to know. In many cases, a rough estimate of the bitumen contained in raw coal is very useful. This may be arrived at by exposing a given amount of the fuel, in an iron pot, with a fitting lid, or in a common cast iron water kettle, to a red heat. This may be done in a common grate. The vessel must be left on the fire for five or six hours. The coke which remains will be similar in amount to that derived from the process pursued in the coke yard. The losses are fugitive gases, and represent what is gene- rally understood by bitumen. The amount of sulphur contained in a given specimen of fuel, it is very difficult to determine. Even a perfect chemical analysis would be of no practical use ; because specimens selected with the greatest care from a pile of coal do not contain a uniform amount of sulphur. REVIVING OF IRON. 139 CHAPTER III. REVIVING OF IRON. IF ores contained no foreign matter, or if they were peroxides, the reviving of iron from its ^ores might easily be effected. But such is not the case. The manufacture of iron is so highly com- plicated by the great mass of impure ores, that it has been found necessary to divide it into several distinct branches. This division affords the advantage of perfecting these branches of the manufac- ture, and consequently of cheapening the product. The reviving of iron is, at present, carried to so high a state of perfection that scarcely any improvement, in this department, can be conceived, at least so far as yield of iron is concerned. But the quality of the metal, and economy of fuel, have not received a corresponding degree of attention. Among the ancients, bar iron was made directly from the ore. This method of making iron, practiced in some parts of Asia at the present day, is, of all others, the most ancient. By this method, the ore is smelted along with charcoal in a temporary smith's forge ; the bellows are urged by hand ; and the iron forged on heavy stones or anvils. Another method, at present practiced in many parts of Europe and in this country, is what is generally known as that of the Cata- lan forge, which was in use among the ancient Romans. Of these two modes of reviving iron, we shall speak at greater length in the next chapter, as they will be more properly considered when we come to speak of the manufacture of bar iron. We shall confine our attention, at present, to crude metal, pig metal, and the appa- ratus employed for its manufacture. Pig metal, or cast iron, is a mixture of different metals, metalloids, carbon, phosphorus, sulphur, &c., and oxides; in which iron and carbon are the preponderating elements. The amount of carbon and other matter varies greatly in different kinds of metal, and the quality and quantity of these ad- mixtures constitute the value of the metal. Pig metal differs from bar iron only in this respect, that it melts at a lower temperature. 140 . MANUFACTURE OF IRON. The chemical composition of bar iron differs so little from that of cast iron, that any criterion based upon this difference is practi- cally valueless. The making of pig metal is the first process in the manufacture of iron. Pig metal varies in quality according to the admixtures of foreign matter in the ore, and according to the mode of manufacturing it. Three classes of pig metal, ranging with the color of its fracture, are generally distinguished. The first is of a dark gray or black color ; it is generally sufficiently soft to receive impressions when struck with a hammer. It is not very strong ; is easily broken ; and shows, when fractured, crystals of black lead or graphite, which is crystalized carbon ; it is coarse grained. This kind of pig metal will melt at a lower temperature than any other ; and deposits, on cooling, graphite, in shining, mica-like leaves or crystals. It contains a large amount of carbon, which results from too much coal in the blast furnace. If re-melted in the air fur- nace, or in the cupola, it resolves into an excellent cast iron, which belongs to the next class. The second class is gray metal. It is tougher and stronger than the first, as well as finer in grain. It forms a good foundry metal ; castings from it are strong and smooth. The third class is white metal, of two distinct kinds. One is the result of too much ore in the blast furnace, and too heavy burden ; or, if the product of the common charge, the result of lack of blast, bad coal, wet weather, inattention of the keeper; or of ores con- taining manganese. The other kind is generally silvery white, suf- ficiently hard to scratch glass ; short, that is, easily broken ; does not receive any impressions from the hammer, of crystaline frac- ture, often very beautiful. A sudden change of temperature will sometimes break it. When struck, it emits a sound like that of a bell. The best metal for the forge is a cast between number two and number three, called mottled iron. It is white, marked with gray spots of plumbago or graphite. In the blast furnace this iron may be made advantageously ; but a furnace cannot well be kept upon such a quality. In the classification just given, we refer only to charcoal iron ; but this classification may be applied to anthracite and coke iron, if dis- tinguished in a manner similar to the above. We should add that iron of the third class made by coke or anthracite is a poor article, and, under all circumstances, furnishes inferior bar iron. Nearly every kind of pig metal alters its color, if suddenly cooled in a stream of cold water. When hot, or when in a half melted condition, the gray casts assume a whitish color. The cause of this behavior we will investigate at the close of this chapter. This change of color is not the result of REVIVING OF IRON. 141 a loss of carbon, for such iron contains as much carbon as the'gray iron from which it is derived. As a general rule, the color of the metal does not depend upon the amount of carbon it contains, for we frequently find more carbon in white than in gray pig metal. We shall now proceed to describe the various modes of obtaining pig metal, or cast iron. I. Reviving Iron in a Crucible. If we mix finely powdered pure oxide of iron with dry charcoal powder, place the mixture in a Hessian crucible, and then expose this crucible to the melting heat of iron in an air furnace, we obtain a quantity of gray cast iron. If the ore contains, besides iron, any foreign matter, as silex, clay, magnesia, and lime, substances very refractory, the result of our manipulations is seriously impaired, for the revived iron retained by the foreign matter cannot follow its gravitating tendency. If the foreign matter is clay or silex, the ore is partly reduced, forms protoxide, and combines with the clay and silex, from which it can scarcely be separated. To prevent this combination, we must have recourse to stronger alkalies than the protoxide of iron, such as lime, magnesia, potash, and soda, which, combining with the clay and silex, liberate the iron. These are the simple elements of the theory of the blast furnace. II. Stuck, or Wulfs Oven Salamander Furnace. This kind of furnace is at present very little in use. A few are still in operation in Hungary and Spain. At one time they were very common in Europe. The iron produced in the stiick oven has always been of a superior kind, favorable for the manufacture of steel; but the manipulation which this oven requires is so expensive, that it has been superseded by the furnace next described. Fig. 41 shows a cross section of a stuck oven ; its inside has the form of a double crucible. This furnace is generally from ten to sixteen feet high; twenty-four inches wide at the bottom and top ; and measures, at its widest part, about five feet. There are generally two tuyeres, a, a ; at least two bellows and nozzles, both on the same side. The breast b is open ; but, during the smelting operation, it is shut by bricks; this opening is generally two feet square. The furnace must be heated before the breast is closed; after which, charcoal and ore are thrown in. The blast is then turned into the furnace. As soon as the ore passes the tuyere, iron is deposited at the bottom of the hearth; when the cinder rises to the tuyere, a portion is suffered to escape through a hole in the dam b. The tuyeres are general!^ 142 MANUFACTURE OF IRON. kept low, upon the surface of the melted iron, which thus becomes whitened. As the iron rises, the tuyeres are raised. In about Fig. 41. Wulfs oven. twenty-four hours, one ton of iron is deposited at the bottom of the furnace. This may be ascertained by the ore put in the furnace. If a quantity of ore is charged sufficient to make the necessary amount of iron for one cast, a few dead or coal charges may then be thrown in. The blast is then stopped; the breast wall removed; and the iron, which is in a solid mass, in the form of a salamander, or stuck, wulf, as the Germans call it, is lifted loose from the bottom by crowbars, taken by a pair of strong tongs, which are fastened on chains, suspended on a swing crane, and then removed to an anvil, where it is flattened by a tilt hammer into four inch thick slabs, cut into blooms, and finally stretched into bar iron by smaller hammers. Meanwhile, the furnace is charged anew with ore and coal, and the same process is renewed. By this method, good iron, as well as steel, is always fur- nished. In fact, the salamander consists of a mixture of iron and steel; of the latter, skillful workmen may save a considerable amount. The blooms are a mixture of fibrous iron, steel, and cast iron. The latter flows into the bottom of the forge fire, in which the blooms are re-heated, and is then converted into bar iron by the same method adopted to convert common pig iron. If the steel is not sufficiently separated, it is worked along with the iron. This REVIVING OF IRON. 143 would be a very desirable process, on account of the good quality of iron which it furnishes, if the loss of ore and waste of fuel it occasions were compensated by the price of bar iron. Poor ores, coke, or anthracite coal, cannot be employed in this process. Charcoal made from hard wood, and the rich magnetic, specular, and sparry ores are almost exclusively used. III. Blue Oven Cast Oven. The furnaces of this construction are an approximation towards the blast furnace of the, present time. Fig. 42 represents the blue oven of the Germans. Its height is from twenty to twenty- five Blue oven. feet. The form of the interior resembles that of the modern blast furnace, a is the tuyere; the breast b is closed with fire brick, or fire proof stones. The bottom slopes towards the breast. This furnace is kept in continuous blast for three, six, or more months, when the hearth widens so much that further work is not deemed profitable. When the furnace is heated to a sufficient degree, the breast is entirely closed, with the exception of a hole at the bottom to let out the iron, and of a hole six or eight inches above the first, through which the scorice flow out. It is filled to the top with coal and iron, the supply of which is renewed as the charges sink. The tuyeres are seldom raised more than from ten to fourteen inches above the bottom ; should 144 MANUFACTURE OF IRON. Fig. 43. iron and cinder rise to the tuyeres, the latter may be left off. The arrangement, generally, is such that both may be let out through the tapping hole for the iron. But if the metal is designed for the making of steel, iron and cinder are let off together; in other cases, each is tapped separately. This furnace is in common use on the Continent of Europe. It is well adapted for the manufac- ture of steel, and yields an excellent forge iron ; but it requires rich ores, and an abundance of charcoal. Its management is simple ; it may be constructed at little expense ; and where rich ores and cheap charcoal are available, may be profitably used in this country. The blue oven is generally used where sparry car-, bonates abound ; and from the steel metal which it furnishes, German or shear steel is manufactured. IV. Various Forms of Furnaces. The present form of the blast furnace occupies the highest position in the scale of improvements successively made upon the Catalan forge, of which we shall speak in the next chapter. The first improvement was that of the sala- mander furnace; the second, that of the blue, or cast furnace. We shall illustrate this gradual improvement in the follow- ing notice of the many various forms of blast furnaces at present in use. It will be sufficient to describe simply the in- terior of these furnaces, for their out- ward forms present but little variation. a. Fig. 43 represents the interior of a charcoal furnace in common use in the Hartz Mountains. This furnace is a peculiar one, on account of its very heavy masonry ; the crucible c is very high and narrow ; the boshes b are ex- ceedingly flat ; the interior will receive a large body of coal and ore; the throat is often from four to five feet wide, and sometimes square. Coal and ore are ex- pensive in these regions, and this fur- nace is constructed as well for the pur- pose of saving fuel, as for producing a good quality of metal. The ores in use are the red, brown, and yellow varieties of the sparry Blast furnace, Hartz Mountains, Germany. REVIVING OF IRON. 145 carbonates frequently mixed with brown hematites, and brown hy- drates, all of which are very refractory. The furnace is generally blown with one tuyere, made of copper. Hot blast, as far as we are aware, is only applied to two furnaces in that region. This furnace is celebrated on account of the very fine, strong bar iron, and the white plate iron, from which steel is manufactured, which it produces. It is managed like any other furnace ; the cinders flow in consequence of their small specific gravity, and by the pres- sure of the blast over the damstone, which is generally square, and lined with cast iron plates. The metal is cast, by means of cool moulds made of cast iron, into plates ten or twelve inches wide, two inches thick, and from five to six feet in length. b. Fig. 44 represents a blast furnace at Malapane, Silesia. It is Fig. 44. Blast furnace in Silesia. twenty-seven feet high; it is blown with two tuyeres and hot blast. The blast is heated at the top. The crucible is seventeen inches wide at the bottom, twenty-eight at the top, and reaches five feet eight inches above the base of the furnace. The boshes are nine feet in diameter; the diameter of the top three feet eight inches, 10 146 MANUFACTURE OF IRON. The tuyeres are but fifteen inches above the bottom stone. In this furnace, pine charcoal is burnt ; the ore used is a yellow hydrate of iron, soft and friable, somewhat resembling common yellow loam. A very fine foundry iron, remarkable for its liquidity, running into the finest sand moulds, is the product of this furnace. From this metal the greater part of the fine Berlin castings are manufactured. This furnace is remarkable on account of the small amount of coal it uses. c. Fig. 45 exhibits a German blast furnace for the smelting of bog ores by charcoal. Though the bog ores of southern Prussia are celebrated for producing very cold-short iron, yet from this furnace a large number of good cannon have been cast for the use of govern- ment. The form of the inside of this furnace varies remarkably from that of other furnaces. The height of the furnace is thirty feet. The crucible a is at the bottom seventeen, and at the top eighteen, inches in width ; its height is five feet six inches. The concave boshes measure, at the widest part, 5, seven feet. The top, from the forge fires or puddling furnaces, and produce the same re- REVIVING OF IRON. 197 suits. From these ores it is almost impossible to smelt gray pig iron- x. A successful business is scarcely possible without a judicious selection and admixture of the smelting materials. Rich ores are apt to contain less foreign matter than is needed for the formation of a sufficient quantity of cinder to protect the hot iron against the influence of the blast ; the production of white iron, and the con- sumption of more fuel than is actually necessary to reduce the ore, are the results of this deficiency. In this case, an admixture of poor ores will be found advantageous. Poor ores of a refractory nature consume- much coal, and furnish a small quantity of iron; but a great deal can be accomplished by the application of hot blast. With respect to rich ores and cinder in small quantity, the hot blast is of but little advantage. We shall arrive at a thorough understanding of this question in the course of this and the fol- lowing chapter. We shall allude here to those applications which were considered useful, and generally adopted, before the science of mixing ores was established. The primary aim of the iron manufacturer is to arrive at per- fection in the smelting operations, that he may be enabled to pro- duce from a given amount of coal and ore the largest possible quantity of metal of a definite quality. This object can be realized by a judicious selection and mixing of ores; and where, through want of material, this is not practicable, by a proper selection and addition of fluxes. Nearly every material mixed with the ores is in itself more or less refractory ; but, where several are mixed under proper circumstances, they will melt together, and be, to a greater or less extent, liquid. The protoxide and peroxide of iron may be considered infusible by themselves ; but melt when mixed. Quicklime, clay, sand or silex, and magnesia, are also very refrac- tory by themselves. Protoxide of iron melts readily when mixed with silex or clay ; and forms, with these substances, a very liquid cinder, in forge fires and puddling furnaces. Lime and magnesia melt together with silex, but require a very high temperature. If, however, a little clay is added to the mixture, the melting is facili- tated; and if a small portion of the oxides of iron is added, the mixture will flux at a still lower temperature. These observations can be made at a coke or anthracite furnace. Potash and silex melt readily together; so also do soda and silex, or, what is the same thing, sand and soda; but a mixture of potash, soda, and sand melts with greater facility. If we add potash or soda, or both, to 198 MANUFACTURE OF IRON. the above mixture of lime, magnesia, and silex, the melting point of the whole will be lowered ; this is somewhat remarkable, because the sand or silex can be increased in a greater ratio than the potash and soda. From this it follows that the greater the number of ele- ments in furnace cinder, the more easily the cinder will flow ; or, in other words, that the more we mix and multiply the kinds of ore, the more regularly the cinder will flow. Silicious ores, calcareous ores, and clay ores are, singly, very refractory and troublesome in the furnace. Ore, mixed principally with silex, requires a high temperature to produce iron, on account of the refractory nature of the admixture. But it will readily make gray iron. Calcareous ore, or iron ore mixed with lime, is equally refractory by itself, requires a high heat for smelting, and is inclined to make white iron. Clay ores are not very refractory; if no lime or potash is present, or, if the ores are not very rich, they do not make iron at all, or make it in very small quantity; for a great deal of the iron is con- sumed in fluxing the clay. If we mix calcareous and silicious ores together, they will not only produce iron with greater facility than each would separately produce it, but they work with less coal ; and if to this mixture we add an ore belonging to the aluminous or clay ores, the operations in the furnace will, in every respect, prove still more satisfactory. There are many more admixtures, as may be presumed, which influence the manufacture of iron; but the above constitute the main body of foreign matter mixed with iron ores. If, through the influence of local causes, we are unable to obtain such a mixture of ore as will satisfy us, we are compelled to add such foreign matter as will produce satisfactory results. Purely silicious ores will require an addition of clay ore and pure limestone, or, if no clay ore can be obtained, argillaceous limestone; and if the latter cannot be had, any mixture of clay and iron, even blue clay, will answer. Fire clay, or any pure clay without iron, we cannot recommend ; but, if it is necessary to make use of such material, it will be advisable to dissolve it, and to mix it well with fine ore. Limestone or calcareous ores require the addition of sili- cious and clay ore; and if these cannot be obtained, ferruginous shale, which generally contains both silex and clay, will answer. But this shale is to be roasted like ore, because it frequently contains sulphurets of iron (iron pyrites). Clay ores generally contain so much silex, that no addition of sand or silicious ore is needed. For these, lime is a sufficient flux. It is a common practice to flux the ores, for which purpose limestone is, in this country, in REVIVING OF IRON. 199 most instances employed, because the main body of the ores are of a silicious and clayey nature. But if, in the case of silicious ore, an argillaceous or magnesian limestone, and, in the case of clay ore, a silicious limestone, be selected, the result will be highly favor- able. In all cases where limestone or any other flux contains a little iron, the smelting operations will be facilitated ; and a mixture of ore will produce the most perfect work. The addition of dead fluxes is thus rendered unnecessary. We cannot too much insist upon the importance of this subject, for upon it depend, .to a greater or less extent, the quality and quantity of the metal, and in consequence the success of the business. There is a point where the liquidity of the slag ceases to be of advantage. Ores which con- tain feldspar, as is generally the case with the magnetic ores, flux, in most instances, too readily ; in which case, a more refractory material, such as silex or clay, is to be added. The silicious ores of Eastern Pennsylvania require a large amount of lime; but where clay ores can be added to the lime, as in Huntingdon county, they work exceedingly well. The Eastern States do not, in this respect, enjoy equal advantages with the Western States. In fact, from the eastern boundary of Pennsylvania to the western boundary of Ar- kansas and Missouri, the coal measures to a greater or less extent everywhere accessible contain this material in abundance. Where small boxes are in use for filling and weighing ore, the distinct separation of the ores and flux is a matter of no difficulty ; but where only one box is used for the whole mixture, much atten- tion is required. The flux, as well as the ore, should be filled by weight; not, as frequently done, thrown in at random by the shovel. For, let it be well remembered that the quantitative mixture of ore, or ore and flux, is definite ; it is not a matter of indifference, in seek- ing to obtain the best result, how much we take of one kind of ore and how much of another, or how much limestone or flux we use. Too great, is as injurious as too small, a quantity of limestone. If the quantities of ore and flux are determined, it is a good practice to mix all the ores previous to being weighed. This mode of mix- ing the ores has from time immemorial been practiced in Germany. It increases to a small degree the labor of the yard, but richly repays this labor in the better work of the furnace. The process, called by the Germans Moellerung, is, simply to spread on some level place a certain quantity, say one hundred wheelbarrowsfull, of one kind of ore; upon this, half that quantity of another kind; upon the latter, one-fifth or one-sixth that quantity of a third kind; and 200 MANUFACTURE OF IRON. over the whole, the limestone or flux, if any is needed. Beds from one to two feet in height are prepared in this way, from which an amount sufficient for a charge is taken. The mixing of the ores can, in this manner, be watched, without the necessity of intrusting its management to unthinking workmen. The ore should be spread uniformly over the coal in the furnace; but where the blast is weak, or the ores wet and earthy, it maybe advisable to pile the ore in the middle of the throat, that the rising gases may escape. This should be avoided, if possible, on account of the coal consumed. Furnaces which have but one charging place, are often badly managed, because the fillers either charge in- discriminately on either side, or, what is still worse, one fille'r is in the habit of throwing the stock to one side, and the filler of the next turn to the other side. These irregularities give rise to changeable work in the hearth, to the formation of lumps in the hearth and boshes, and finally, what is generally the case when the furnace is well heated, to scaffolding in and above the boshes, which, of course, is likely to be attended with serious consequences. Ores which contain zinc, arsenic, or chlorides are apt to scaffold, at some point of the upper part of the in-wall, in charcoal furnaces. In this respect, stone coal or coke furnaces are in no danger. For this evil, small coal charges and a hot top are the best reme- dies. Sufficiently wide throats, and the heating of the ore in the middle of the coal, are required, to keep the lining as warm as pos- sible, and to permit the evaporated metals to escape. y. The number of charges brought down in twelve or twenty- four hours, or the quantity of iron produced, depends very much on the amount of blast sent into the furnace. Nevertheless, we may remark that the quantity of air does not determine with certainty the descent of charges, or the quantity of iron made. A cold hearth never produces so much iron as a properly heated furnace, where the blast, in both instances, is the same. If the hearth is too warm, nearly the same difficulty occurs. A liquid, lively cinder makes a far greater quantity of iron with a given amount of blast, than a tough, chilly cinder. Cold, black, or dark green cinder produces still less iron, and is, on the whole, the least advantageous. A clean hearth, free of clinkers and cold iron, is, of all others, the most likely to produce good metal, and in abundant quantity. z. The question, what number of tuyeres it is most profitable to us in a furnace, it is difficult to answer. It can be decided only by experience. Nevertheless, we shall present some conclusions REVIVING OF IRON. 201 drawn from experience. Where cold blast is used, we should be in favor of never applying more than two tuyeres, and of trying very hard to do with one ; but where hot blast is used, two or even more tuyeres are almost indispensable, for the following rea- sons : In the smelting process by cold blast, as strong a pressure in blast as the fuel will possibly bear is highly advantageous. This fact favors the use of as few tuyeres as possible, for, if heavy pressure is applied, the more tuyeres we have, the more coal we destroy. In addition to this, the dust in the hearth and boshes in- creases. With hot blast the matter is different. There is no need of pressure ; and by the tendency of the hot air to combine more readily with the coal, small coal, which does not burn well, is very apt to gather in the corners of the hearth, and produce difficulties that are well known. Therefore, the same reasons which are in favor of as few tuyeres as possible with cold blast, are in favor of as many as possible with hot blast. It is sometimes the case that the gray iron from the furnace is directly used for foundry purposes, such as to cast hollow ware, stove plates, &c. This mode of making use of the hot metal is practiced only at a few small charcoal furnaces. In many respects, this is a bad practice, and should be avoided. The disturbance which it occasions to the smelting operations more than counter- balances the advantage gained ; and, besides, the castings of re- melted iron are preferable to those cast directly from the furnace. aa. The time at which the iron should be let out is generally so arranged that the workmen, changing every twelye hours, have each their cast ; where the hands work by the job, they generally stay from one cast to the other, a period frequently of twenty-four hours, if the furnace is newly blown in, or if any disturbance happens. The preparation of the pig bed, moulding of pigs, is the duty of the keeper, or, at large furnaces, of the helper, or second keeper. The founder generally assumes the duty of tapping the iron. Six in the morning, and six in the evening is the time usually set apart for casting. bb. If the melted iron remains too long in the furnace, it is very apt to turn white, on account of the action of the blast. Such an accident should be avoided, for it is injurious both to the furnace and to the iron. But in charcoal furnaces, the inconvenience is not so great as in anthracite and coke furnaces. If, in an anthracite furnace, the cinder rises too high, it is very apt to adhere in lumps 202 MANUFACTURE OF IRON. to the hearthstones ; after the iron is let out, we are forced to break away these lumps with great caution. In charcoal furnaces, how- ever, the action of the blast is frequently resorted to for the produc- tion of white iron for the forges ; and should the original iron have been gray, or mottled, a strong forge iron is produced. At many European furnaces, where forge metal is manufactured, the desired effect that is, the production of white, strong metal, with the least expenditure of coal is obtained by some secret method of twist- ing and dipping a tuyere. This manoeuvre, at the wiilf J s oven and the blue oven, is applied to the ores of the primitive and tran- sition formation, as spathic and magnetic iron ores. It would be of no use, in this country, at places where oxides for the production of gray iron are principally smelted. Where white iron is smelted by a high tuyere, or, what is the same thing, where the iron cannot be reached by the free oxygen of the blast ; and where it is smelted by a weak blast, or a too wide hearth, it is always bad, weak, does not yield well, and does not make good wrought iron. cc. If no accidents or disturbances happen in the regular furnace operations, and if everything is in proper order, the quality of the metal, that is, its amount of carbon, is entirely dependent upon the burden. Small burden will produce gray, and heavier burden white iron. In the former case, the furnace will be inclined to dry the cinder, that is, to deposit balls in the hearth, by which the hearth is cooled, and the temperature frequently brought so low as to produce a tendency towards the other extreme, that is, black or dark green cinder with white iron. Such changes are very disadvantageous, and should, by all means, be avoided. A well-conducted furnace should never be too heavily, and never too lightly charged, for one extreme is as bad as the other. A medium course is the most profitable, that is, to make mottled iron, and trust to accident for the manufacture of gray or white iron ; for both, in certain cases, will be produced, In this way the furnace will carry the heavier burden, and the result will be, in either case, a good forge metal. White iron is produced by a cold furnace ; but it can be made by a hot furnace. The white iron of too heavy burden always proves a good forge iron ; but the white iron of too light burden is of a very doubtful nature, and in most instances is bad, if smelted from the ores of the coal formation. It is very bad, if made by hot blast, anthracite, or coke. The making of white iron can be prevented only by carrying as heavy a burden as possible. REVIVING OF IRON. 203 dd. The mixture of ore and flux is, with respect to the quality of the metal, a matter of great importance, for too much lime will, under all conditions, produce white iron. If the hot slag, as it flows from the furnace, blazes, and gets spongy like pumice stone when sprinkled with water, we may conclude that lime exists in too great quantity in the charge; but if the cinder appears of a dark, black, or green color, even after the temperature in the hearth is raised; and if the slag in the furnace, in spite of the heat, is inclined to form balls, to blacken the tuyere, and to stick to the hearthstones, we may conclude there is not sufficient lime in the charge. Clay ores are very apt to clinker before the tuyere, even where an abund- ance of limestone is present; but the limestone may be diminished by the application of hot blast. If the composition of the ores is such as by itself to make a very liquid cinder which, with bog ore, shell ores, or calcareous ores, is frequently the case we must not expect gray iron, until with this composition we mix silicious ore. Silicious ore is highly favorable to the manufacture of gray iron ; in fact, foundry iron can hardly be made without it. To pro- duce such iron, a strong cinder and a hot furnace are required. The least disturbance which tends to cool the furnace will cool the tough cinder, and in this way often produce very troublesome scaffoldings in the hearth or boshes. ee. The color of the cinders is not a safe criterion by which we may estimate the working of the furnace. Gray cinders may con- tain as much iron as green or black cinders. But, as a general rule, the former indicate better work than the latter. Where the charcoal furnace is in good condition, they are generally well glazed, transparent, and of a greenish color. Perfectly gray, spongy, white, and black or olive-green cinders are not the most favorable indications at a charcoal furnace. Anthracite and coke furnaces, when well conducted, generally furnish a gray, stony-looking cin- der, but always well glazed. In these furnaces, spongy, or green, or black cinders are almost as unfavorable indications as in char- coal furnaces. Those which lose their glazing, or fall to pieces, by being exposed to the influence of the atmosphere, contain too much lime, and never fail to make white iron of inferior quality. That their color is no indication of the quality of the metal, is evi- dent; for the ore or coal may contain the oxides of other metals, which generally produce various shades. Variegated cinders, like agate, indicate that the ore or flux employed is too coarse, or, what is still worse, that there is scaffolding in the furnace. Small stacks, 204 MANUFACTURE OF IRON. or narrow hearths, are endangered when they work coarse ore. In a large anthracite furnace with a wide hearth, so much pains need not be taken in breaking the stock, for there is scarcely a possibility of choking or scaffolding such a furnace. ff. If any accidents occur, such as scaffolding below or above the tuyere, or in lining, it is a bad practice to throw in at the tuyere materials either to flux or to heat the furnace. Lumps and cold cinders below the tuyere can be far more easily removed by means of the bars and ringers than by means of fluxes thrown into the tuyere, or thrown below the timp; for the addition of fluxes does nothing more, at best, than to remove the lumps where they are the least troublesome. Scaffolding above the tuyere, when it impedes the blast, or the descent of charges, is to be removed by the withdrawal of the ore charges; and, if considered dangerous, by sinking the materials in the furnace to a point very near or above the boshes, and melting away, by means of scrap iron with lime- stone, as in a cupola, any obstruction in the hearth or boshes. All difficulties may thus be removed in a very short time. Obstructions which endanger the progress of the smelting operations, by so choking or chilling the hearth that coal cannot descend, are the re- sults of inexcusable neglect inexcusable both to the manager and to the workmen. Charcoal furnaces are but little exposed to such disorders: but coke and anthracite furnaces are very much exposed to them, if they smelt gray iron; for, in the manufacture of this iron, a narrow hearth and strong cinder are required. When, in such furnaces, the least disturbance takes place, the cinder is very apt to stick to the boshes or hearth, and a green, and at last a black, cinder and white iron are produced. So long as the cinder is only of a light green color, or streaked with green, no danger need be apprehended, and the furnace maybe considered in good condition; but so soon as brownish streaks in the cinder appear, the furnace should be watched. If the brown color does not disappear within five or six hours, it is advisable to diminish the ore charges, for this color deepens so rapidly, that within twenty-four hours the cinder will become black. If light charges should not be near at hand, the difficulties would thus be greatly augmented. gg. The flame of the trunnel head, as well as that of the timp, is indicative of the nature of the operations in the furnace. At charcoal and coke furnaces, a heavy, dark top flame indicates that the furnace is cold, and that the burden is too heavy. A bright smoky flame, which throws off white fumes, indicates a too liquid REVIVING OF IRON. 205 cinder ; that too much limestone is present; or that the burden is too light. If the iron is gray, the burden can be increased ; but if it is white, this should be done cautiously. The withdrawal of a por- tion of the limestone will generally cure the evil, if the iron is white ; but if it is gray, heavier burden is required. An almost invisible, lively flame at the top is significant of a healthy state of the furnace. The strength of the top flame of an anthracite furnace is proportionate to the amount of hydrogen the coal con- tains; and therefore this is, at best, but an uncertain indication of the state of the furnace. If the flame appears to be struggling to break through the timp, we may be sure that there is something wrong in the interior. But this depends upon the ore and coal, upon the form of the stack, and upon the blast. It is common where small ore is used, and where the hearth and top are narrow. The color of the timp flame is, like that of the top flame, indicative of the work in the furnace; and the rules applicable to the one are applicable to the other. The color of the flame will be more or less modified, ac- cording to the foreign matter the ore contains. If it contains zinc, arsenic, and lead, the flame will always emit white fumes, whether the furnace be cold or warm. If the materials contain common salt, the flame will emit fumes of the same color. "Where the flame wavers, that is, where it is sometimes large and sometimes small, there is, without doubt, scaffolding in the lining. In this case, close watching of the sinking of the charges is needed. If it is found that all is not right, a reduction of the burden and an in- crease of blast must be resorted to. M. The gray metal, where the operation has been good, is very liquid; and ke"eps liquid for a long time in thepig^bed. If of good quality, it is, even in the thinnest leaf, perfectly gray ; but if in- clined to white, the corners of the pigs, and thin castings, will be white. This iron appears perfectly white when liquid ; while white metal is of a somewhat reddish, yellowish color, and throws out sparks. White metal chills very soon in the moulds, and assumes a rough, concave surface; it adheres, with much tenacity, to the iron tools used for cleaning the hearth. If metal contains sulphur, it is very apt to throw off fumes of sulphurous acid, or sulphuretted hydrogen. It throws off sulphurous acid, if smelted by coke or coal, and neutral or proper cinder ; and sulphuretted hydrogen, if lime is used in large quantity, which is generally the case, because such iron cannot be smelted without an excess of limestone. Phos- phorus can be detected only by an analysis of the metal. 206 MANUFACTURE OF IRON. eV. After the metal in the moulds is cooled, it is to be removed, weighed, and stored; and the sand of the pig bed dug up, wetted, and prepared for another cast. The cinders at small furnaces are easily removed in common carts. At stone coal furnaces, various methods have been devised to remove the large mass of cinder daily pro- duced, of which that at present generally practiced at the anthra- cite furnace may be considered the best. It is this: Dig two round basins of about five or six feet in diameter, and two feet in depth, at the side of the stack. In the centre of each basin put a piece of pig metal, in an upright position. Around this pig metal, the cinders, which run into the basin, gather. A chain attached to a crane is then fastened to the pig metal, by means of which the cold cinder is placed upon any suitable vehicle, to be carried off. A whole volume might be written without exhausting what could be said on the management of furnaces, and of blast furnaces in particular. But our space is limited, and we wish to avoid pro- lixity. Many occasions will arise, in the course of this work, which, we hope, will enable us to supply whatever deficiency our statements may, thus far, have exhibited. X. Theory of the Blast Furnace. It would be inconsistent with our object to enter, with scientific minuteness, upon this branch of our investigations. If we shall be able to convey to an intelligent mind a clear and comprehensive view of the operations which take place in the interior of the blast furnace, our design will be accomplished. It is evident our expla- nations must be somewhat of a speculative nature ; but these are illustrated and confirmed by operations performed under the cogni- zance of our senses. In the previous chapters, we have related and reasoned upon matters which can be tangibly verified; but in the present instance, we are obliged to draw general conclusions from isolated, though well-established facts, by means of pure analogy an operation frequently and daily needed, and constantly per- formed by those engaged in the management of blast furnaces. a. In the second chapter, we have spoken of fuel and its combus- tion, as well as of the different combinations which oxygen forms with fuel. We are forced to refer to that subject in the present in- stance, for the process of combustion must be well understood be- fore we can understand the chemical operations which take place in a blast furnace. The fuel used in the blast furnace is composed, to a greater or less degree, of carbon, hydrogen, sulphur, and ashes. REVIVING OF IRON. 207 Fig. 70. If oxygen or atmospheric air combines with carbon, the result is either carbonic oxide or carbonic acid; at a high temperature, with a sufficient supply of air, always carbonic acid. A suffocated com- bustion, with an excess of fuel, generally produces carbonic oxide. The result of the combustion of hydrogen and oxygen is always water ; that of the combustion of sulphur and oxygen always sul- phurous acid. b. Combustion in a blast furnace is, as may well be expected, of a somewhat complicated nature, and requires illustration to be under- stood. Fig. 70 represents a sec- tion of a blast furnace in operation, filled with coal, ore, and fluxes. If we introduce at , a, the tuyere holes, a current of air or blast, combus- tion in the lower part will ensue ; and, according to circumstances, the product will be carbonic acid of greater or less durability. But if we have an excess of fuel, and a limited supply of air, the final pro- duct of the combustion will be car- bonic oxide. The primitive or im- mediate combination of carbon and oxygen at the tuyere forms carbonic acid ; and this carbonic acid, in its progress through the coal, combines with more carbon, and forms car- bonic oxide. Carbonic acid can- not combine with any more oxygen than it already possesses ; but carbonic oxide will combine with as much more as it already con- tains. Carbonic acid is of no use in reviving iron from the ore, for the ore is a combination of iron and oxygen ; and carbonic acid could not abstract any oxygen from the ore. But carbonic oxide will combine with whatever oxygen is present in the interior of the blast furnace. c. Practical investigation has demonstrated that the more friable and tender the coal is, the more easily oxygen combines with it ; and that the more compact it is, that is, the greater its specific gravity, the greater is the difficulty with which it combines with oxygen. Heated air combines more readily with fuel than cold air, Theory of the blast furnace illustrated. 208 MANUFACTURE OF IRON. and of course is more inclined to form carbonic oxide. Soft, open fuel and heated air form carbonic oxide, the agent in the reduction of the ore, more readily than hard coal; and we may conclude that charcoal and coke are more useful than anthracite coal in the manufacture of iron. According to this statement, the atmo- sphere of oxygen and carbonic acid will be a zone of greater or less radius, of which the mouth of the tuyere is the centre, as the cir- cular lines in the engraving indicate. The radius of this zone has been found, by experiments made on furnaces, to vary, according to fuel and blast, from six inches to four or more feet. Applying what we have said to a common furnace, with grate and draft, the column of carbonic acid will be from six inches to four feet in height, if we pass a current of atmospheric air through hot and burn- ing fuel. If the column of fuel is higher than this, the carbonic acid will be gradually converted into carbonic oxide. This process is exactly the same in the blast furnace ; the oxygen of the atmo sphere is gradually converted into carbonic acid, carbon with much oxygen and then gradually into carbonic oxide, carbon with less oxygen. Where the atmosphere of carbonic acid ceases in the blast furnace, we may conclude that the working of the carbonic oxide upon the ores commences, and that it changes more or less in its course upwards. The ascending current of the gases, in a blast furnace, consists, then, of carbonic oxide, hydrogen, and combina- tions of hydrogen and carbon. These latter gases are derived directly from the fuel, above the reach of free oxygen, and constitute gaseous combustibles, ready to unite with oxygen. Mixed with the above are steam, carbonic acid, and nitrogen incombustible gasetf which have not the least influence upon the ore. The nitrogen is derived from the atmosphere. The ascending current of the gases from the tuyeres differs in composition according to height ; of course this composition will not be alike at a given height in two furnaces of different construc- tion, and in which different materials are used. Actual experi- ments on furnaces carried on by hot blast and charcoal have fur- nished the following results : Directly above the tuyere. Nitrogen. Carbonic acid. Carbonic oxide. Hydrogen. 8 feet 63.07 " 35.01 1.92 13 59.14 8.86 28.18 3.82 22J 57.80 13.96 22.24 6.00 25J " 57.79 12.88 23.51 5.82 Nitrogen. 64.58 Carb. acid. 5.97 Carb. oxide. 26.51 Hydrogen. 1.06 Carburetted hydrogen. 1.88 63.89 3.60 29.21 2.07 1.07 62.34 8.77 24.20 1.33 3.36 Nitrogen. Carbonic acid. Carbonic oxide. Hydrogen. 61.07 0.68 36.84 1.41 64.66 0.57 33.39 1.38 63.59 2.77 31.83 1.81 60.70 11.58 25.24 2.48 REVIVING OF IRON. 209 We find here, what might have been expected, a gradual in- crease of the carbonic acid. This is generated by the contact of carbonic oxide with the ore. The relative amount of the different gases is not equal in different furnaces, for, in another case, the gases were mixed in the following proportions: Directly above the tuyere. 5} feet 11} " 17} " The gases of a coke furnace exhibited the following composi- tion: Directly above the tuyere. 2 feet 17J " 28 " 31 " There are, particularly in coke furnaces, gases of a compound character ; but these have little to do with practical results, the aim of our investigations. From the above, it is apparent that the carbonic acid gas increases as the current of gas ascends ; and that, on an average, one-third of the carbonic oxide has been converted into carbonic acid before escaping at the top. If the carbonic oxide is the only reagent in the conversion of ore into iron, we may conclude that one-third of the fuel has been properly applied for the purpose for which it was designed. We here have evidence that all the fuel has not done its duty; otherwise, all the carbonic oxide would have been converted into carbonic acid, and all the hydrogen into water. But such is not the case. If a furnace works well, there will be more carbonic acid at the top of the charges than there will be if a furnace works badly; this circumstance accounts for the different appearance of the trunnel head flame. d. The theory of the reduction of ore will then be simply this : the gases ascending in the furnace leave a part of their positive elements to combine with the oxygen of the ore, that is, carbonic oxide leaves carbon, and, under peculiar circumstances, hydrogen may be retained. If carbonic oxide absorbs oxygen from the ore, it leaves of course metallic iron or protoxide, and the ore, in de- scending, will be a mixture of metallic iron and foreign matter. If 14 210 MANUFACTURE OF IRON". that process is not well performed, some oxides of iron will be left in the mixture. If an ore, to this extent prepared, but without any surplus of carbon, descends into the hearth, it cannot produce anything but white iron; for, if the iron is once heated to redness, and melts, it absorbs no more carbon. All the carbon required for making gray iron must be in the ore before it sinks into the hearth. For this and many other reasons, we are forced to assume a surplus of free carbon in the gas mixtures of the blast furnace carbon, if not chemically, at least mechanically, mixed with the gases, and so finely diifused, that it can penetrate into the pores of the ore. If we adopt this theory, that is, the presence of free carbon, we can account for many apparent irregularities in furnace operations for which we cannot account on the simple assumption that the gases ascend in their constitutional form. By adopting this theory, we account for a circumstance otherwise incomprehensible, that is, the great influence exerted by the pressure of the blast ; for if nothing else than carbonic oxide is needed, almost any pressure, even the weakest blast, will accomplish all that is desired. But we know, by experience, that the strongest blast which a given kind of fuel will bear advantageously, is the most profitable. It appears, from this, that the blast works mechanically as well as chemically, in the destruction of coal; and that a certain power will produce particles of coal of a size best calculated to penetrate the pores of the ore. If these particles are too large, they cannot reach the interior of the ore, and the iron will be white. This may be assigned as the reason why a particular pressure of the blast is required to produce gray metal. If the blast is too weak, it produces white iron from defi- ciency of carbon in the ore ; and if too strong, the consequences are equally injurious. Such an admixture of free carbon will be, of course, uniformly diffused among the gases, and penetrate the porous ores more readily than even the gases themselves, on ac- count of the superior affinity of carbon for oxygen. A further evidence of the agency of free carbon, in the smelting of gray iron, is in the fact that compact, close ores, of whatever chemical composition, will not produce gray iron. Should an atmosphere of carbonic oxide, or even carbon in any other form, alone be needed to smelt gray metal, there would be no difficulty in manufacturing gray iron from any kind of ore. But this is not the case. From com- pact specular ore, magnetic ore, the carbonates, and ores too hard burnt, we cannot make gray iron, whatever amount of coal we em- ploy, and whatever kind of blast we use. A certain aggregate REVIVING OF IRON. 211 form of the ore is, under all conditions, required for the manufac- ture of gray metal ; and this is an open, porous form. We find that pieces of ore taken from the furnace when in good condition are, towards the boshes, of a black, and, higher up, of a brown color. An analysis of such ores has never been made ; such an analysis would, of course, be attended with great difficulties ; but if the composition of the ores, in their gradual descent from the top to the bottom, were fairly tested, a great accession to our knowledge would be realized. e. The operation in a furnace is, then, as follows : In the upper part of the stack, the water of the materials is expelled, hydrogen from the coal is driven off, and the porous ore is, to a greater or less degree, saturated with carbon, by means of which the carbonic oxide serves to reduce a part of the ore to protoxide. The ore, in this condition, will appear, before entering the hearth, like a brick winch has been exposed to the interior heat of a charcoal kiln or a coke oven a mixture of iron ore, foreign matter, and carbon. This applies to cases in which the furnace operations are in good order, and in which gray iron is manufactured. All circumstances which interfere with the regular course of this work contravene favor- able results, particularly in relation to the quality of the metal. /. The conditions under which such a state of things may be ex- pected, are, a porous and dry ore, a blast neither too weak nor too strong, and a low temperature in the upper parts of the stack. A high temperature is not sufficient to produce a combination of iron and carbon, at least when the iron is once in a liquid state. On the other hand, if the metal is liquid, it is difficult to separate car- bon from it. Where the upper part of the stack is too warm, the hydrogen liberated from the coal will combine with the oxygen of the ore, and leave metallic white iron in the ore. This iron has no affinity for carbon, and will sink into the hearth without it. This also happens where the highest part of the furnace is too warm, so that, in the limited space through which the materials pass, the ore is sufficiently heated, without a protecting coat of carbon; and also where the coal contains too much hydrogen, as in the case of half charred wood, bad coke, or bituminous coal. If too much hydrogen is present, it is not entirely expelled until the materials are very low in the stack ; but then it has sufficient time, on its way to the top, to combine with some of the oxygen of the ore, even with the oxygen of the silex, or other oxides. We invariably find that the iron made under such circumstances is bad ; for, where hydro- 212 MANUFACTURE OF IRON. gen exerts any action upon ore, it deprives it of its oxygen, and thus destroys that affinity for carbon necessary to make gray metal. If the hydrogen is permitted to exert a greater degree of influence, it will decompose silex, lime, and manganese, which will combine as an alloy with the iron, and injure its strength and ductility. The best means of obviating this influence of the hydrogen at least, the best method of making it as little dangerous as possible is to em- ploy low stacks, wide throats, and an abundance of strong blast; or, if foundry metal is to be made, hot blast. Narrow tops and weak blast will not answer for bituminous fuel. We may expect to find bitumen or hydrogen in imperfectly charred wood, in soft, half burnt coke, or in anthracite of bituminous character. Under all these circumstances, we can most effectually work with a wide trunnel head, even though the other conditions cannot be complied with. A narrow throat will expose the ore, in its almost raw state, to the influence of the hydrogen ; and in that condition, without any car- bon to protect the ore, the mischief is consummated before the ore is fairly in the furnace. We will endeavor to make this subject clearly understood. Fig. TO represents a furnace with a narrow top, examples of which we have frequently seen. The arrows in- dicate the current of the gases. We may here very easily perceive that the heat and action of the gases on the ore are to a great degree lost; for it is evident that the coal will be pressed towards the lining, and that the heavier ore will remain in the centre. It is reasonable to suppose that the gases, instead of winding them- selves through the close, heavy ore, will choose the easier passage through the coal. Their action upon the ore is confined to the short, narrow passage in the throat. Narrow tops, weak blast, and high stacks may answer for good, coarse fuel, and for open, porous ores, which are not loamy; but, in all other cases, they answer im- perfectly. Where well-burnt or open ores, and dry and well-charred fuel, are available, it is advisable to have wide throats, strong blast, and stacks that are not too high. In cases in which the best mate- rials are not at our service, or in which they are too expensive, we should use all the means to arrive at favorable results which circumstances may afford us. g. The above principles are not deduced from theory : the facts on which they are based were observed prior to the existence of the science of the blast furnace, as the following considerations will establish : REVIVING OF IRON. 213 We find that, in Sweden, where magnetic ores are smelted, as well as in Russia, and at the furnace at Cold Springs, N. Y., wide trunnel heads are employed. We allude only to those establishments in which the business is in so high a state of cultivation as to pro- voke imitation. At these places, we find not only the manufacture of a superior metal, but a remarkable reduction in the consump- tion of fuel. In Styria, Western Germany, and at some establish- ments in France, the very difficult sparry carbonates are princi- pally smelted. These carbonates are never perfectly oxidized. Wide tops are employed at these places, which are celebrated on account of the small amount of fuel consumed. In our own country, scarcely any of these refractory ores are smelted ; and at a few furnaces in New Jersey, New York, and the Eastern States, where they are used, mixtures are, to a greater or less extent, worked ; and the ore charges are brought to a medium proportion of the magnetic and peroxide ores. We find an exception to this at Lake Champlain ; but of this locality, we shall speak hereafter. If we apply the above principles to the fuel, we find that some Russian furnaces, employing raw wood, make use of very wide tops, even from five to six feet square, and work with very strong blast, generally with but one tuyere, and with the exclusion of hot blast. Some French and German furnaces, which employ either red coal or kiln-dried wood successfully, have been compelled to make use of wide tops and strong blast. Of the extent to which experiments with wood in furnaces have succeeded in the United States, we have no satis- factory information : but we are inclined to believe that, in an economical point of view, such experiments would fail ; for there are few localities where both ore and fuel are found in proper con- dition. Economically, open, porous oxides work better with charcoal than with wood. Refractory rich ores can be smelted with wood. Anthracite furnaces require wider tops than coke furnaces ; while the latter require far wider tops than charcoal furnaces. This width of the top may be considered the most essential improvement on the blast furnace which is supplied by anthracite coal. The height of the stack in anthracite is much less than in coke furnaces ; and somewhat lower than in charcoal furnaces. Anthracite furnaces vary from thirty to thirty-five feet in height; charcoal furnaces from thirty to forty feet ; and coke furnaces from forty to sixty feet. The width of the trunnel head varies, in the United States, considerably. In Pennsylvania, Ohio, Kentucky, and Tennessee, the width of fur- naces at the boshes is nine, and often ten feet, and at the top from 214 MANUFACTURE OF IRON. eighteen to twenty inches ; or, in the proportion of thirty square feet at the boshes to one square foot at the top. The Cold Spring furnace measures at the boshes nine feet, and at the top thirty-two inches. Here the proportion is eleven feet at the boshes to one foot at the top. The dimensions of charcoal furnaces, in Europe, which smelt refractory ores, are generally in the proportion of five feet at the boshes to one foot at the throat ; frequently in the propor- tion of four to one. In coke furnaces, the proportion of the hori- zontal section of the boshes to that of the top is seldom less than four to one, though sometimes even 2.5 to 1. In anthracite furnaces, the diameter of the throat is six feet, and that of the boshes twelve feet ; that is, in the proportion of one to four. But sometimes the boshes measure thirteen, and the tops eight feet square ; in this case, we have the proportion of two to one. When we take into consideration the small height of the stack, and the strong blast which is applied, we shall find that this arrangement in anthracite furnaces is, in an economical point of view, very favorable ; for, instead of retarding, it facilitates the vent of the gases. Narrow tops answer where loamy ore and soft coal are used ; but in such cases, if we expect favorable results, we should employ weak blast and high stacks. But these conditions can be observed only where coal and labor are cheap. If we are in doubt concerning the proper dimensions of a furnace, our best course is to commence with a comparatively low stack, wide throat, and with as high a pressure in the blast as the fuel will possibly bear. h. The foregoing demonstrations are designed to suggest the method of producing an excellent quality of metal. It is evident that the ore should be so prepared in the upper part of the furnace that it may be brought into the crucible in the best possible con- dition for producing the best metal which circumstances will per- mit ; for we cannot expect to make gray iron from raw magnetic ore, from clinkers of ore burnt too hard, or from forge cinders. But though we are unable to smelt gray iron of good quality from these materials, nothing should prevent us from endeavoring to make the best use of the stock at our disposal. If, by means of scientific knowledge and industry, we obtain a cheaper stock, or one of better quality, we should not refuse useful material, sim- ply because our furnace is not prepared to receive it. i. If one of the conditions of success, in blast furnace opera- tions, is that the ore should be properly prepared, that is, saturated REVIVING OF IRON. 215 with carbon before it reaches the hearth, or arrives at the melting heat of iron, it is, of course, a question of great importance, what kind of ore, and what composition, are best adapted to receive the carbon, and to retain it. It is easily understood that compact ores, that is, ores of great specific gravity, even if they are per- oxides, and unburnt magnetic ores, spathic or argillaceous car- bonates, are not well calculated to absorb carbon. Silicates or alu- minatcs should not be smelted, for these ores are so compact that no carbon can penetrate them. The question is not so much one of chemical composition, as of mechanical or aggregate form of the ore. Such a form is the most easily produced in the peroxide; for, under most conditions, this oxide, if rubbed, yields an impalpable powder ; even when in compact masses, its powder is, of all others, the finest. For this reason, and for no other reason, we roast the ore. We roast the magnetic ore to open crevices ; roast the car- bonates to expel the carbonic acid gas, and open the pores ; and burn hydrates to evaporate the water belonging to their chemical composition, and thus make room for carbon. We endeavor, in roasting, to raise the oxidation of the ore to a peroxide ; with the specific object of increasing the affinity of the ore for carbon, or carbonic oxide. If the ore absorbs more carbonic oxide in one instance than in another, and if the composition of the gas, that is, carbonic oxide and free carbon, is the same in both cases, then the greater the amount of oxygen it contains, the greater will be the amount of carbon which will condense upon and penetrate the ore. For these reasons, roasted and oxidized ores are required in the manufacture of gray iron. This theory is in perfect harmony with experience; and a practical iron manufacturer will find no difficulty in arriving at evidence from facts within his knowledge. Jc. Ores are, in most cases, not only composed of iron and oxygen, but are a compound of oxide of iron and more or less foreign mat- ter. The mixtures of oxides of iron and foreign matter are innu- merable ; still, where this mixture takes place beyond a given de- gree, the compound ceases to constitute an iron ore. But the quality rather than the quantity of foreign matter in the ore de- termines this question, as we shall presently see. Nevertheless, a mineral which contains less than twenty per cent, of iron is not usually considered an ore of iron. Silex, lime, and clay are the common admixtures. Other ingredients, such as magnesia, man- ganese, and titanium, whatever influence they may have in par- ticular cases, may, in ordinary investigations, be neglected. 211S MANUFACTURE OF IRON. I. The next important question is, what influence will any mix- ture of foreign matter have upon the iron ore, so far as the absorp- tion of carbon is concerned ? To answer this we would simply say, that, according to science, and experiments in the laboratory, clay possesses the greatest affinity for carbon ; next in order is silex ; and then lime. This classification, however perfectly true in theory, is not confirmed by practical results. In this case, theory and prac- tice appear to be at variance with each other ; but when we take into consideration that the mechanical form of the matter is the cause of this difference in the results arrived at by theory and practice, this apparent exception to a general rule of chemistry is explained. Iron manufacturers generally consider calcareous ore the most favorable of all ores for the manufacture of gray or foundry metal. Clay ore, and then silicious ore, come next in importance. But we should be cautious how far we base practical results upon this experience, for it frequently happens that the theory which we have deduced from practice fails ; and from this failure great losses ensue. The above practical rule is applicable only where the lime or calcareous ores are, as is generally the case, already mixed with foreign matter, and where silicious and argillaceous ores are in their purity. Experiments, practically confirmed, made by Mushet, and related in his papers on iron and steel, clearly prove that clay has the greatest affinity for carbon ; next to clay comes silex, and then lime. A low temperature and very little fuel will revive iron from a mixture of clay and oxide of iron ; but all the iron the mixture contains will not be revived, because clay is infusible by itself, and retains some particles of iron, and of course carbon. The iron is retained as an element, or radical, of an alkali. A stronger alkali is necessary, by combining with the clay, to thrust the iron from its hiding-place. This affinity of the carbon and iron for the clay might be dissolved, if the aggregate form of the clay would permit the formation of larger globules of iron ; for these, following the law of gravitation, would separate in spite of affinity. Nearly the same thing happens with a mixture of silex and oxide of iron, with this difference, that silex does not absorb carbon so readily as clay, and does not revive iron by so low a temperature, and with so little fuel. But if no alkali combines, at the proper time, with the clay or silex, neither would yield all its iron, even though revived and carbonized. Of these earths, lime is the very last which absorbs carbon and revives iron ; but then it precipitates all its iron at once, because carbonate of lime is fusible REVIVING OF IRON. 217 by itself, and will, when concentrated into a melted slag, squeeze the iron out. m. Agreeably to these principles, clay ores will require a low temperature in the upper part of the stack. We should endeavor to extend this temperature to as low a depth as possible. This will prevent the precipitation of iron before any lime is sufficiently heated to receive the clay, and will consequently prevent the combination of iron and clay into an aluminate, from which it is difficult to sepa- rate the iron. Silex or silicious ore is very nearly of the same character, but will permit of a higher degree of heat, without much danger. With calcareous ore we may raise the heat as high in the stack as we please, without endangering the result. These prin- ciples, deduced from theory, coincide exactly with experience at the furnace. If we smelt pure calcareous ore not what is commonly called the limestone ore, for this is generally a precipitate of iron upon a limestone bed, and contains very little lime we need a strong heat in the furnace, and an abundance of fuel. The reason of this is that the upper heat of the stack and the action of the re- viving gases are entirely lost, for lime and limestone ore condense little or no carbon. We thus find that pure calcareous ores are not the most profitable; and we shall make a better use of the fuel, if along with it we mix silicious and clay ore. In this way, we shall not only derive greater profit from the gases, but a lower tempera- ture of the stack will enable us to secure many advantages. Fo- reign admixtures are thus shown to be unaccompanied with injurious results; but this principle cannot be extended to a chemical admix- ture or combination. Chemical compositions of silex, clay, and lime are of very difficult decomposition; the very fact that their texture is so close, is the reason why no carbon can penetrate and combine with the oxygen of the iron. This is applicable also to forge and puddling cinder, and to clinkers, and to ore very hardly burnt. n. From these statements, it is evident that a proper mixture of different ores will be beneficial, so far as the use of fuel is con- cerned; and that the more closely and intimately the ores are mixed, the better will be the result. A medium temperature is a security that the furnace will work well, and guarantees economy of fuel and a favorable product. Where proper mixtures of foreign matter are already contained in the ore, the most profitable work may of course be expected. Ores of this kind are frequently met with in the coal formations, as precipitates upon limestone or clay. This is the case 218 MANUFACTURE OF IRON. at Huntingdon county, Pa., and at other places. The out-crop ores of the anthracite coal series, as well as the Western coal fields, ex- hibit generally this composition. A great majority of the Western furnaces, such as those at Hanging Rocfk, and at many places along the Alleghany River, work these ores. We have, we think, sufficiently proved that the aggregate form the mechanical composition of an ore has an important hearing upon the operations of a furnace ; but it is obvious that the chemical relations must be still more important. To arrive, by the surest and shortest method, at a clear and comprehensive conclusion, we shall describe the particular behavior of each kind of ore. o. If we charge a furnace with unroasted magnetic ore, the ore will sink with the coal charges unaltered until it arrives at a cer- tain point, when it will melt into a more or less liquid slag. This slag will pass through a column of hot coal, when a portion of the iron will be revived; another portion will combine with silicious and aluminous matter, and form cinder, which is lost. The iron which results is not gray. The carbonates, and other compact and heavy ores, exhibit the same peculiarities. If limestone is charged along with the ore, a large quantity of iron will be revived ; still, a great deal of iron is lost. In no case should we expect gray iron; for, though it should happen that some carburetted iron has been formed in the furnace about the hearth, yet so long as the cinder contains protoxide of iron, the carbon from the gray iron in the hearth will be absorbed, and iron from the cinder revived. The latter is the case when the ores contain foreign matter ; but, if the ores contain little or no foreign matter, there will not be sufficient cinder even to protect the iron from the influence of the oxygen of the blast. In this case, the iron must of course be white. The ores may be compact or porous. The result is, in both cases, the same; for, if carburetted iron is formed in the upper parts of the furnace, without a protecting cinder, it will be white before it ar- rives in the crucible. Satisfactory results cannot be obtained from these ores, unless we have a warm furnace, and unless the heat is raised to a considerable height in the stack. p. If an iron ore contains foreign matter, and if this matter is a single earth, in itself refractory, the mechanical form of the ore may be the most advantageous ; but the metal which results will always be white. When a furnace is charged with clay ore, the ore will, in its descent, absorb and condense carbon. When the carburetted metal arrives within reach of the blast, the carbon will be absorbed REVIVING OF IRON. 219 by the carbonic acid, and the iron will arrive whitened in the cru- cible ; the remaining iron yet in the clay will be highly carburetted ; but the clay cannot melt and protect the iron. The result is white iron; and, if no limestone is present, an aluminate of iron, as cinder. What we have stated is applicable, in most respects, to silicious ore ; also to calcareous ore, with this difference, that, in the latter case, no protoxide of iron is needed to flux the cinder. If, by ap- plying an excess of fuel, we try to revive all the iron from the ore, or, at least, to revive it in greater quantity, then, with clay as well as silicious ore, we receive a tenacious cinder in the hearth below the tuyere, which retains the globules of iron on its surface. If a dark gray carburet comes down, it will soon become white iron within the influence of the blast. Should the cinder not be suffi- ciently liquid to permit the iron to pass through it, the iron will oxidize, and form protoxide to the cinder, until it effects a passage. The necessity of fluxes is thus clearly seen. q. If we charge a furnace with poor ore, with an admixture of a refractory character, in a state of fine, impalpable aggregation, as is generally the case with clay ores, and particularly the case with some silicious ores, the iron will be revived by a comparatively low temperature, and for this reason will combine with a large amount of carbon. But this carbon cannot be retained, if the original globules of iron are exposed to the direct influence of the blast ; for these grains of melted metal are so small that they can pass through only a very liquid cinder. Should the cinder not be sufficiently liquid, the resulting metal will be white. This is another reason why the smelting of gray iron from clay and some silicious ores is so difficult. To arrive at desirable results, it is advisable to have fine clay ore along with silicious ore. This ore revives a portion of its iron by a low heat, and is, of course, highly carburetted. If the iron produced descends, and finds, on its way, silicious ore ready to deliver iron, it will combine with it, and form a larger mass. If this combination, in its further descent, comes in contact with a calcareous ore, which, under ordinary circumstances, would not liberate iron, the carbu- retted iron of the clay and silicious ore will draw with it a portion of iron from the calcareous ore ; this augmented combination will resist the influence of the blast, and by its ponderability will work, with greater readiness, a passage through the melted cinder below the tuyere. The remaining iron in the clay ore, which, in most pases, amounts to half the original quantity, will be enclosed 220 MANUFACTURE OF IRON. in the unaltered piece of ore until it arrives in the hearth below the tuyere. If, at that point, it meets with silicious or calcareous ore, both of which are in the same condition, the different earths, being in a high temperature, will combine, form a liquid cinder, and squeeze the iron out. The iron, having been protected from the blast by the refractory cinder which surrounds it, is now per- fectly protected against the blast by the melting cinder, composed of the foreign matter of the different ores. The case which we have described seldom happens, for there are few clay ores which do not contain a portion of silex ; few silicious ores which contain no lime, or magnesia, or clay ; and scarcely any calcareous ore which does not contain a portion of clay or silex. The above is a theoretical case, brought forward merely to illus- trate a principle. There is a possibility that similar coincidences may exist in practice ; but they can happen only very seldom. r. Experience has clearly proved that, of all ores, those which flux themselves are the most profitable. That is to say, any mix- ture of ore, or any individual ore, which produces good metal, and a liquid cinder free of iron, is more profitable than those ores which require the interference of art. What constitutes a good cinder, we shall investigate hereafter. We shall confine our atten- tion at present simply to the iron, and to the operations which take place in the furnace. If clay ore, as already explained, yields a portion of its iron very readily, we may infer that this is grayer than any other portion, because carbon combines more easily with iron at a low than at a high temperature : but this carburetted iron is destroyed on account of the refractory quality of the clay. If the clay, mixed with the ore, should contain a portion of lime and silex, its refractory character would be diminished, and the car- buretted iron in the inside of the fragment of ore would be more perfectly protected against the influence of the blast. If the car- buretted iron thus protected, should find an alkali in the cinder, below the tuyere, waiting to receive the foreign matter, it will descend with scarcely any loss of carbon. From this it is evident that we may expect gray metal from mixed clay ores, if lime or any alkali is present in the hearth ; but not otherwise. If the foreign admixtures of the ore are not of such a nature as to form a liquid cinder, the cinder must be made sufficiently liquid by the addition of flux, or by the loss of a portion of iron. In reality, there are few purely clay, silicious, or calcareous ores. REVIVING OF IRON. 221 The native deposits are, to a greater or less extent, compounds of iron ore, and of various foreign matters; still the clay and silicious ores predominate. Calcareous ores are very seldom met with on this continent. Therefore, in most cases where iron is smelted, an admixture of limestone, instead of calcareous ore, will answer every purpose. A mixture of lime and iron is always available ; for pure lime will sink into the hearth, and remain in lumps until slowly dissolved by descending clay or silex. If lime contains a quantity of iron, or other foreign matter, it will melt above the tuyere, leave the hearth free of any obstruction to the descending iron, and give the blast free play at the coal; therefore, a lime- stone which is not refractory is preferable. s. The above process takes place when silicious and clay ores are to be smelted, and when the flux is limestone. But let us consider the case in which calcareous ore is to be smelted, and fluxed with silex or clay. As mentioned above, calcareous ores require a strong heat, and permit the raising of the heat to an uncommon height in the stack. On that account, more fuel is consumed for these than for other ores. If calcareous ore is smelted by charcoal, which contains but a small quantity of silex, the ore will melt into a slag, as in the case of the magnetic ore, and in descending will lose some of its iron. If the heat is strong, and if more iron is separated, some of the lime will either be blown off at the trunnel head, which we often observe issuing in a white, fine dust, or will combine with the silex of the coke or stone coal, and descend to the hearth. Under all circumstances, a part of the lime will descend below the tuyere, and if it does not find silex or clay in the cinder, it will attack the hearthstones; and by this means the lime is saturated with clay or silex, becomes liquid, and is in a condition fit to be discharged. In this case, the iron is of no use in fluxing the limestone, for, at the high temperature of the furnace and hearth, all the iron is precipi- tated; and if there is no carbonic acid in the lime, or if no clay or silex is present, no combination between them is possible. Calca- reous ores should be fluxed by clay or silex. Pure sand and fire clay cannot be of any service ; they do not melt ; they sink gradually into the hearth; and if any iron from the calcareous ore is liberated, it has a tendency to combine with silex or clay. The first chance of receiving oxygen affords an opportunity of forming protoxide of iron, and silicates or aluminates of iron. Such disturbances happen frequently with calcareous ore. An excess of lime in the ore is, to 222 MANUFACTURE OF IRON. all appearances, not sufficient to precipitate the whole of the iron, because the blast cools off the hard, unmelted clinkers of clay and silex around the tuyere. If, in such cases, we select a clay which contains iron, or any matter capable of melting the clay at a low temperature, then such a flux, melting at a high point in the stack, will meet, in its descent through the hot fuel, the heated calcareous ore, and, combining with the lime, liberate the iron, which is then at liberty to descend. A part of the iron will be retained by the imperfectly melted lime and flux ; but, on coming in contact with the more concentrated heat of the hearth, it will be separated. t. From the foregoing demonstrations, we are enabled to draw conclusions relative to the economical working of the ores. It follows, from what we have stated, that clay ores are, of all ores, the most profitable, because of the facility with which they absorb carbon, and because of the low temperature at which they precipi- tate iron; but the refractory character of their admixtures prevents us from deriving those advantages from them which, under other circumstances, they would furnish. Silicious ores do not absorb carbon so readily ; but the foreign matter which they contain is more inclined to form liquid compounds with lime or iron, and to liberate the revived iron. On this account, they are more manage- able than the clay ores. Notwithstanding the tendency of silicious ores to smelt white iron, the fusibility of their admixtures, in con- tact with alkalies, gives them precedence over the clay ores for the smelting of gray or foundry metal. Calcareous ores do not con- dense carbon, if the amount of lime in the admixture is large ; but if the amount is small, they condense carbon like the pure peroxide of iron. However, they will not retain it absorbed, because the iron revived is not very liquid. The carbon is retained in the ore until exposed to the influence of blast, when it disappears. It is a well-known fact that ores containing much lime in admixture do not produce gray iron with facility, and consume more fuel than any other ores. u. What is the cause of the difference in the capacity of matters to absorb and retain carbon ? For the solution of this question we must refer to chemistry. But so important is this subject to the iron manufacturer, that we shall offer no apology for direct- ing, as briefly as possible, his attention to it. Observation has un- questionably proved that clay possesses the power of condensing REVIVING OF IRON. 223 carbon in the highest degree; and that there is scarcely any matter so little disposed to absorb and retain carbon as lime. As a car- bonate, lime will absorb carbon, but not as burnt or quick-lime. This may be caused, in part, by chemical affinity ; but there is no question that it is, to some extent, caused by the mechanical form of the particles of matter ; otherwise the difference between clay and silex would not be so great. There is no doubt that the same power which retains the carbon retains the iron. The particles of clay are very minute ; so also are the particles of the oxide of iron mixed with clay. When carbon penetrates the pores of such a mixture, and heat is applied, a part of the metal is retained in the interior of the ore fragment. The particles of silex are coarser than those of clay ; and if the affinity of silex for carbon were so great as that of clay, it could not retain so much iron as the latter, because of its coarse grain. From this quality, added to the other facilities which it possesses of reviving iron, it may be considered a more profitable ore than the above. Lime, in its aggregate form, is very fine-grained; but it does not absorb any carbon, and for that reason the iron is refractory, that is, it cannot separate from the lime at a low heat. The iron is not sufficiently liquid thus to separate, and is retained until the lime, becoming fluxed, leaves it. v. For these reasons, we are convinced that rich ores consume a great deal of fuel, and, on this account, are not so good as some poorer ores. If the disadvantages of their use are not compensated by cheap fuel, and by the production of a good quality of metal, it is not advisable to smelt them by themselves. Calcareous ore is equally expensive, so far as the consumption of fuel is concerned ; and, if smelted by itself, is little apt to produce good iron. The same remark applies, to nearly the same extent, to silicious ore. Clay ore, if poor, would not produce any iron, if smelted without fluxes. It thus clearly appears that no iron ore, of whatever de- scription, is, smelted by itself, so profitable as it would be when mixed with other ores. The iron which, in the clay ore, is so readily carbonized, will not separate from its foreign matter until that matter is absorbed by another element which has the power of liquefying it. This is also the case with silicious and calcareous ores. Rich ores do not smelt well, because their pores have no opportunity of absorbing carbon at a low temperature; therefore, these ores are not pre- pared for reduction when they arrive in the neighborhood of the 224 MANUFACTURE OF IRON. tuyere. The rich ores receive and absorb carbon, and produce iron, by flowing in a semi-liquid slag through a column of hot coal of greater or less height, according to the quality of the ore. These considerations lead us to the construction of the interior of the blast furnace, and to the development of the principles by which its form and dimensions are determined. Applying these principles, we should build a furnace without a hearth, that is, by sloping the boshes down to the tuyere, in case it is our intention to smelt rich ores; and we should make a partial or complete hearth above the tuyere, according as facilities presented themselves of mixing the rich ore with poorer ores of the proper kind. If we could bring the mixture to an advantageous standard, we should employ a narrow or high hearth, with the object of econo- mizing fuel, of obtaining a better yield from the ore, and of smelting gray iron. Any alterations required should be made in conformity to these considerations. If we arrive at conclusions too hastily, we shall have the mortification of finding that our anticipations will not be realized, and we shall be under the necessity of returning to the original form. The form we have suggested, that is, a furnace with- out a hearth, owes its importance to the necessity which exists of raising the temperature of the whole stack to a high degree, because, unless there is a high column of hot coal, the melted ore will not be affected by carbon. This rule is also to be applied to calcareous ore. For silicious and clay ores the hearth may be high, and the boshes flat. These ores absorb carbon in proportion to the coolness of the upper part of the furnace. When, after being saturated with carbon, they arrive in the narrow part of the hearth, the intense heat of the crucible will melt the iron and the foreign matter almost at the same moment. If the foreign matter is fluxed, the iron will thus be precipitated in the shortest possible manner. From these investigations we have arrived at the conclusion, theoretically, that no ore is perfect. This conclusion is confirmed by practice. The magnetic are not the most profitable ores, because of the amount of fuel they consume. The same remark applies to the compact oxides, to calcareous ore, and to silicious and clay ore. For this reason, the latter do not yield well. By mixing the various kinds of ore, the virtues of one will counterbalance the imper- fections of another. The desideratum is to find a proportional admixture united in a native ore. In practice, the ores are mixed in a certain ratio artificially. This conclusion leads naturally to REVIVING OF IRON. 225 the inquiry concerning the different portions of each kind of ore, and, consequently, to the constitution of cinder. w. It will be clear to every discerning mind, after reading the above, that the knowledge of the composition of the foreign matters in the ores, which, when melted together, constitute cinder, and the knowledge of the circumstances under which the most favorable results can be obtained, are highly valuable. Iron, under certain conditions, can be melted; if protected against oxygen, it is un- affected by heat. Like other metals, it is more fusible when an alloy is combined with it ; it is most fusible when combined with phosphorus, sulphur, or carbon. The latter element is preferable, because phosphorus and sulphur are considered injurious to the quality of the metal. We are thus led to conclude that, if iron could be combined with carbon under all circumstances, it would be equally liquid, no matter from what kind of ore it has been smelted. This conclusion is true : but we have seen that some ores will not make carburetted iron at all; and that others, which make it in abundance, cannot precipitate all their iron, on account of the refractory quality of its admixture. If an admixture of ore is just as fusible as the iron itself, the iron and foreign matter will separate spontaneously. This will be the surest and most profitable way of smelting. From this, it is apparent that the appropriate way of proceeding will be, so to combine different ores that the iron and foreign matter will melt at the same moment, or, what is the same thing, at the same temperature. If such a mixture is porous, it will absorb carbon, and offer a chance of smelting by a lower temperature. If its composition is favorable to the absorption of carbon, the only difficulty which remains is the production of a cinder quite as liquid as the iron. This is performed less easily than we should at first conceive; for, if we compound the material for the making of cinder, it is only under certain conditions that we arrive at the best results; and these con- ditions are, to a great extent, limited to local elements. x. Mr. J. H. Alexander, of Baltimore, tells us, in his " Report on the Manufacture of Iron," that the difference in the consumption of fuel varies according to the fusibility of the ore, or, what is the same thing, according to the iron and cinder; and he shows us that the richest ores consume the most fuel. We extract the following table from his Report: 15 226 MANUFACTURE OF IRON. Table showing the probable consumption of charcoal per 100 parts of crude iron, with ores of different sorts. Denomination. Proportion of metal Charcoal consumed per 100 metal. 66 to 90 Fusible ores yielding ^ 30 " 35 90 " 110 120 " 130 110 " 140 Ores of mean fusibility yielding^ 40 " 50 140 " 180 180 " 220 160 " 200 Ores hardly fusible yielding ^ 40 " 50 210 " 250 250 " 300 These results, applied to tons of iron and bushels of coal, would give us from 100 to 440 bushels of charcoal per ton of iron. To understand this table properly, we may remark that the above amount of fuel will be consumed, if we manufacture gray iron. The rich specular ores, the spathic ores, &c., do not consume much fuel, if we are satisfied with white metal, and suffer a portion of the ore, in combination with the foreign matter, to form cinder. y. The relative degree of fusibility of the cinder is, however, the main point to be gained. Where the cinder is too thick and pasty below the tuyere, the iron globules cannot pass the blast without injury ; where it is too liquid, it will leave the iron too soon, and thus expose the metal to the influence of the blast. The most desirable condition is that in which the cinder and iron have the same fusibility, and arrive together in the hearth before either is sufficiently heated for melting. If one should be more fusible than the other, that one is the cinder. But to secure this high state of fusibility, and at the same time to smelt gray iron, is possible only under very favorable conditions. The fusibility of earthy compounds depends principally upon their chemical relations. We do not feel sufficiently interested in this highly intricate subject to enter upon its investigation ; and we doubt whether, after all, we should derive from such an investigation more information than we have already obtained from experience. In most cases, artificial fluxes are too expensive for use; in fact, they are unnecessary, because we can produce almost any degree of fusibility we desire by means of lime, clay, and silex. All other materials which serve as fluxes are in quantities too small to be entitled to notice, and impracticable for general application : such REVIVING OF IRON. 227 as soda, potash, manganese, and magnesia. We shall cursorily notice these materials, as they are occasionally employed, and as they will assist in the explanation of the principles of fusibility. Soda is the most powerful solvent of silex or clay ; after this comes potash, then lime, then magnesia. The alkaline earths, as baryta, strontia, lime, magnesia, and alumina form with silex very refractory compounds. If but one of these earths is combined with silex, the compound is scarcely fusible in the strongest heat of the blast furnace. Such combinations exist as native deposits. Fire clay is a compound of silex and clay, a silicate of alumina ; it will resist a very strong heat. Soapstone is a silicate of magnesia, and also bears a very strong fire ; but an excessive heat is not required to melt a mixture of pounded fire clay and pounded soapstone. This principle is the leading feature in the art of mixing ores. We see here that silex, in combination with clay or magnesia, will not melt ; but a mixture of a given amount of alkali, magnesia, and clay, with a given amount of silex or acid, is fusible. If to the above two silicates we add a third silicate in itself either infusible or strongly refractory, say silicate of lime, the whole mixture is melted at a lower temperature than that at which any two of them will melt ; and if we still add a fourth silicate, the fusibility is be- low the mean temperature of the whole mixture. That is to say, if the first silicate will melt by itself at 100, the second at 90, the third at 80, and the fourth at 20, the whole, mixed together, will melt below a temperature represented by the sum of all the tem- peratures added together, and divided by the number of primary silicates. Thus, 100+90 + 80 + 20 = 72 would be the mean; but the composition would melt below the mean temperature. The fusibility of a binary compound, that is, a single base and silex, depends very much on the degree of chemical affinity of the two elements. As we have before stated, soda and silex have the greatest affinity. Then follow potash, baryta, strontia, magnesia, lime, and lastly clay, in the order of affinity. That is to say, a mixture of one pound of soda and one pound of silex will melt at a lower temperature than a mixture of one pound of clay and one pound of silex. Or, if learned chemists are not satisfied with the expression pounds, let us say equivalents. If the amount of one element increases too much, proportionally to the other, the fusibility decreases. There is a limit in the relative proportion of matter at which the greatest fusibility is produced. The fusibility of a mix- ture of baryta and silex ranges between thirty and seventy per 228 MANUFACTURE OF IRON. cent. In cases where the silex is less than thirty, and more than seventy, per cent., the mixture is equally infusible. So far does this law extend, that the most fusible compounds permit the greatest range, and the least fusible are confined to the narrowest limits. Potash is fusible by itself, and a mixture of ninety-nine silex and one soda or potash is not infusible ; while the fusibility of a lime silicate ranges only between twenty-five and forty-seven per cent, of lime, and a strontia silicate is confined to but one proportion, that is, forty-five strontia and fifty-five silex. Silicates of clay and of magnesia are not fusible at all in the heat of a blast furnace. The alkalies proper and the alkaline earths are not the only elements which form fusible compounds with silex. The metallic oxides, in obedience to the law of affinity, possess this attribute in a higher degree even than the alkaline earths. The oxides of bis- muth, lead and iron especially, form fusible compounds ; of these, however, the silicates of iron alone interest us. The protoxide of iron forms with silex a very fusible compound, which reaches from 40 to 82 per cent, of protoxide, and is not far behind the lime. Peroxide of iron silicates are almost infusible ; and sesquioxide silicates, or magnetic oxide silicates, range between peroxide and protoxide silicates. Copper, zinc, and tin silicates are scarcely fusible. Amongst the electro-positive elements (the bases) of the above enumerated compounds, we should pay particular attention to the behavior of clay under different circumstances. Clay is not a strong alkali, but possesses the remarkable property of becoming an alkali where an alkali is needed, and of forming an acid where there is a surplus of alkali in the composition. z. The tendency of the alkalies or their carbonates to dissolve metallic oxides is a fact worthy of special notice. Six parts of carbonate of potash dissolve one part of iron protoxide, and car- bonate of iron is still more soluble. The carbonates of lime and magnesia dissolve protoxide and carbonate of iron very readily. Other metallic compounds of that kind are of no interest to us. Silicates, or the melted and liquid compounds of alkalies and silex, possess the property of dissolving metallic oxides, and often to a large amount. Such solutions are, to a greater or less extent, colored ; sometimes they are white. The protoxides of iron impart a green color to the cinder, and, if in large quantity, a black color. Magnetic oxide colors the cinder brown, and, when in large amount, black. Peroxide of iron imparts a dirty yellow or reddish color to REVIVING OF IRON. 229 tlie cinder; it is but little soluble. Carbonate of iron imparts to cinder a white or yellow color. Colors imparted by other matter will be mentioned in another place. It may be mentioned that free lime, or a surplus of lime in cin- der, possesses the property of absorbing sulphur. Free alumina, or a surplus of alumina, if an abundance of alkali is present, will ab- sorb phosphorus, and carry it off in the cinders. The same remark applies to lime. After the above consideration of the general principles in the formation of cinders, we are led to inquire, what constitutes a good cinder? A positively good cinder is one which is fusible at that heat at which the iron it encloses will become liquid. The lower that temperature, the less will be the amount of fuel used in the process of smelting. From this it is obvious that the fusibility of the cinder should bear a certain relation to the mechanical form and chemical composition of the ore. An open, porous clay ore will require the most fusible cinder ; and a calcareous ore, a refrac- tory admixture. Different degrees of fusibility require distinct compositions of cinder and of iron ; therefore, the cinders from differently composed ores, and from different fuel, will require dif- ferent temperatures for smelting. aa. The degrees of heat at which iron containing more or less carbon will melt, are not accurately known. According to our own calculations in Chapter II., the fusibility of pig iron is not beyond 2700, because the fuel in the blastfurnace cannot produce a higher temperature. Many able observers have concluded that the tem- perature exceeds that point, but that it is not beyond 3000. We may, from these premises, conclude that the melting point of the different kinds of metal ranges between 2000 and 3000. In investigations concerning the fusibility of silicates, cinders, and artificial compounds, some very useful experiments have been made, from which we select the following: A furnace cinder, composed of silex 50, alumina 17, protoxide of iron 3, lime 30, melted at 2576. Another cinder, composed of silex 58, alumina 6, protoxide of iron 2, manganese 2, magnesia 10, lime 22, fused at 2500. The latter is a very complicated cinder, and ought to melt at a somewhat low heat, but its composition is of a very refractory cha- racter, as may be observed from the large amount of silex it con- tains. These cinders will bear comparison with the anthracite cin- ders of Pennsylvania. 230 MANUFACTURE OF IRON. A greenish, rather dark cinder, from a charcoal furnace, melted at 2498. We shall have an opportunity of presenting further analyses of cinder in the next chapter. It ought to be remarked that, in forming artificial cinder from very finely powdered elements, the temperature at which the smelting commences is always from 500 to 700 higher than that at which the cinder is kept liquid. The truth of this remark is sufficiently proved by the refractory character of the elements. Hence results the necessity of pound- ing the materials, as far as practicable. Where the elements of the ore are very refractory, the finest division is required; but where the elements of a very liquid cinder are contained in the ore itself, the breaking of the ore requires no special attention. bb. It is very seldom in our power to select an ore, for our smelt- ing operations, 'which contains in itself the elements of a good cin- der. Sometimes we are enabled to mix those ores which form a good cinder, and which flux each other; but in most cases we are compelled to add a dead flux to the ore we have selected. This is found to be most profitable if we are enabled to add limestone to our ore charges. The reasons why limestone is the best flux are the following: For various reasons, we generally attempt to smelt gray iron. To effect this object, it is necessary to produce, or we at least desire, a cinder but slightly more fusible than the iron itself. Gray iron is most easily produced from clay or silicious ore, or from very porous oxides. Where we have a choice between a clay and a calcareous ore, the former should be selected, for it offers greater advantages than the latter. If a clay ore is fluxed by lime, the lime will not melt in the upper part of the furnace, but will de- scend into the hearth in its original form, sometimes burnt into quick-lime, but very often as a carbonate. Now, if prepared ore de- scends, the lime is ready to receive the clay and silex, and the iron is speedily separated ; the whole mass in the hearth, hot coal and lime, through which the ore and iron are to pass, is favorable to the reviving of the metal ; and should particles of iron and foreign mat- ter even be melted together, there is a chance that a separation will be effected. But this is not the case where calcareous ore is smelted. If the amount of lime mixed with the ore is so large as to require a flux of clay or silex (under all circumstances, clay is preferable, because it contains a portion of silex), the silicious matter will de- scend to the tuyere, and there wait for the lime; the calcareous ore does not melt nor yield its iron until it arrives at the tuyere, that is, REVIVING OF IRON. 231 if the ore contains no other matter which would make it fusible. In this case, the whole process of forming cinder is accomplished very nearly before the tuyere ; while in the former case, the pro- cess will, in most cases, commence higher up in the hearth. ce. We have thus endeavored to explain, in as simple a manner as possible, the theory of the blast furnace. The composition of cinder, a subject less easily understood by those who have not studied chemistry, deserves our closest attention. To bring this subject to the comprehension of all who desire information, we shall conclude this chapter by presenting a series of applications, drawn from ore analyses, contained in " Rogers' Report on the Geology of Pennsylvania. 1 ' To convey a clear idea of the specific object we desire to accom- plish, we shall insert the following analysis of furnace cinder, taken from Mr. Alexander's Report. We shall .also attempt to recon- struct from the ores of Pennsylvania such cinders as are taken from furnaces in Europe. CHARCOAL FURNACES. COKE FURNACES. Peroxide ores. Sparry carbonate ores. Carbonates of the coal formations. 1 2 3 4 5 6 7 8 9 10 Silica 51.34 63.6 31.1 52.0 71.0 37.8 ' 49.6 40.6 43.2 35.4 Lime 21.80 24.0 14.1 30.2 7.2 32.2 35.2 38.4 Magnesia 482 1.2 34.2 5.2 5.2 8.6 15.2 4.0 1.5 Alumina 15.21 3.8 8.9 5.0 2.5 2.1 9.0 16.8 12.0 16.2 Prot, of iron 3.73 1.7 1.0 1.6 5.0 21.5 0.4 10.4 4.2 1.2 Prot. of manganese 1.16 3.9 4.4 4.7 6.5 29.2 25.8 2.6 Oxide of titanium 9.0 Sulphur trace trace 1.4 Phosphoric acid trace No. 1 is an average cinder of good iron ; No. 2 is derived from a furnace smelting bog ores ; No. 3 from a Swedish furnace ; No. 4 from France; No. 5 from Savoy, and is the result of bad work in the furnace; Nos. 6 and 7 from a German furnace the first when in bad, and the latter when in good, condition ; this furnace usually melts steel metal, from which German steel is manufactured. Nos. 8, 9, and 10 are derived from coke furnaces in Wales. The latter specimen is said to be taken from the furnace when in bad condition ; it is from the same as No. 8. To imitate these cinders, just as they appear in the above table, would be almost impossible. But this is not required. We can 232 MANUFACTURE OF IRON. arrive at the same result by another method a somewhat indirect one, to be sure, but still tending to consummate the desired result. There is a law of chemistry which governs the present case: namely, that a certain amount of silex or acid requires the saturation of a certain amount of base, such as lime, or magnesia, or protoxide of iron,. so that no base or acid be left uncombined. In such a neu- tral condition, the cinder will be most fusible. Now it matters not whether lime is replaced by magnesia, by protoxide of manganese, or by any given base ; but there should never be a large surplus either of acid or alkali in the furnace, for such a surplus will remain refractory, and finally occasion much trouble. The elements which combine to form a fusible cinder are the oxides of metals : and the amount of oxygen in the silex or acid must be equal to the amount of oxygen in* the base, or it must be present in a proportion two or three times greater, or two or three times less, than the amount of oxygen in the base ; that is to say, the one must be united with the other in definite proportions. An enumeration of the equiva- lents of the various compounds necessary to be taken into considera- tion may be found in any handbook of chemistry. In the fore- going, as well as the following demonstration, we must not be under- stood to say that, with the saturation of the silex by different alka- lies, the degree of fusibility is the same ; but that the saturation to which we allude, is a neutralization by which a surplus either of alkali or acid is prevented. In the above table, the average cinder, No. 1, may be considered a fair specimen for imitation. For practical purposes, it is not necessary that the equivalents should be numerically correct. A small surplus of oxygen in either respect should not be considered very injurious. Should we wish to imitate cinder No. 1 at a char- coal furnace situated near the canal, Westmoreland county, Pa., we shall find that, at that locality, the main body of ore is of the argillaceous kind. A specimen of this ore, taken near Blairsville> exhibited, according to an analysis of Prof. Rogers, the following composition in 100 parts : Carbonate of iron - - 71.19 Carbonate of lime ... 3.50 Carbonate of magnesia 2.72 Alumina 2.10 Silica - - - 17.55 Water, &c. - - - - - - 2.94 REVIVING OF IRON. 233 We are not concerned, at present, -with the iron and water, but with the other ingredients. The proportion of the silex in the ore is only 17.55 ; but the table above presented calls for 51.84. We must, therefore, multiply all other oxides by the number resulting from the division of 51.84 by 17.55, which is 2.9. Alumina 2.10 x 2.9 = 6.09, which leaves a deficiency of 15.21 6.09 = 9.12 alumina. Carbonate of magnesia is composed of 44.69 mag- nesia and 35.86 carbonic acid. We assume it to be a subcar- bonate, which is generally the case; then the amount of caustic magnesia in this specimen of ore will be /(44.69 + 35.86) : 2.72 = 44.69 : :A x 2.9 = 3.77. But we want 4.82 magnesia; therefore a deficiency of 4.82 3.77 = 1.05 magnesia is left. Carbonate of lime is a compound of 56 lime and 44 carbonic acid. In the table, the proportion of lime is 21.80. The ore contains but 3.50 carbonate of lime; thus making /(56 + 44) : 3.50=56 : x\ x 2.9 = 5.99: therefore there is a deficiency of 21.80 5.99 = 15.81 lime. Tak- ing the whole 15.81 lime, 1.05 magnesia, 9.12 alumina, a small quantity of iron and manganese is yet required. The ashes of the fuel will deliver a portion of the alkaline matter, but its quantity is comparatively small ; and our labors would be unnecessarily compli- cated if we should take that into account. Our next object should be to find a mineral flux which contains whatever matter we need. "No useful purpose will be served by mixing the ore with another from the same neighborhood; because none can be found which contains the requisite amount of clay and manganese. In the coal regions along the canal, among the various veins of different com- position, an ore may be found suitable to match the above. We select from Rogers' Report an analysis of a specimen found at Brighton, Beaver county. As the same strata of rock from which this specimen was taken are accessible at the Pennsylvania canal, tnere is a possibility that an ore like it, or similar to it, may be found in its vicinity. The Brighton ore contains Carbonate of iron ... 4-3.89 Carbonate of manganese - 7.20 Carbonate of lime - 42.51 Carbonate of magnesia 3.57 Silex, &c. - 0.40 Loss - 2.43 234 MANUFACTURE OF IRON. The silex, in this analysis, may be neglected. If its amount were greater, it would be necessary to add it to the silex of the first spe- cimen, and correct the basic elements accordingly; that is, to add to the silex of the first specimen the silex of the second, and to subtract from the lime, magnesia, and alumina, according to the ratio of the silex added. In the above specimen, we have 42.51 carbonate of lime, which will be equivalent to ((56 + 44) : 42.51=56 : x\ x=23.8 lime. We have also 3.57 carbonate of magnesia, which is equal to /(44.69 + 35.S6) : 3.57=44.69 : x\ and X is 1.59 magnesia. The manganese amounts to 4.75. In comparing the whole, the result is as follows: The analysis requires First ore. Second ore. Total. Silex 51.84 51.84 = 51.84 Lime 21.80 5.99 + 23.8 = 29.79 Magnesia 4.82 3.77 + 1.59 = 5.36 Alumina' 15.21 6.09 = 6.09 Protoxide of iron 3.73 Protoxide of manganese 1.16 4.75 = 4.75 The first may be considered an argillaceous, inclining to a silicious ore, and the second a true calcareous ore. Each of these ores fluxes the other almost completely, at least sufficiently for every practical purpose. The mixture of these two ores would work ex- ceedingly well, and readily produce gray foundry metal; with greater facility even than the original or model cinder in the table of Mr. Alexander. But the metal thus produced would not be so strong as that smelted by the model cinder ; neither would it be so well adapted to produce forge metal, because of the deficiency of alumina in the Bolivar ore. These ores should be mixed in the ratio of 2.9 of the first to one of the latter; the resulting mixture would yield 30.1 per cent, of iron, for the first ore contains 34.37, and the second 20.79, per cent, of iron. If no calcareous ore can be found at a reasonable price, a dead flux, such as lime, should be employed. The first, or Bolivar ore, contains clay in quantity rather too small to make forge metal, but in quantity sufficiently large to produce foundry metal. If it is our design to smelt the former, an addition of clay, or of limestone which contains clay, should be preferred to pure limestone. In the regular limestone strata of that region, limestone which contains even one or two per cent, can scarcely be found ; but in the shale REVIVING OF IRON. 235 strata of the same region, there exist small deposits of an argil- laceous limestone called cement lime. From these sources, a pro- fitable flux for the above ore may be obtained. dd. In Mr. Alexander's table, cinder No. 3 is taken from a Swedish furnace: this cinder is remarkable on account of the large amount of titanium which it contains; titanium is generally the com- panion of magnetic ores. A similar ore exists at Lake Champlain in large quantities, the working of which is different from that of most other ores. We shall therefore make some remarks on this subject, for not only the Lake Champlain ores, but most of the magnetic ores of New York, Wisconsin, and Missouri, contain titanium. It is said that the above ore contains 11 per cent, of oxide of titanium ; this will account as well for the many difficulties encountered in smelt- ing it, as for the great expenses incurred in manufacturing iron from it. Titanium, or titanic acid, neither benefits the iron ore nor injures the metal manufactured from it; still, it may occasion much trouble in the furnace. It does not combine with iron, silex, lime, potash, or anything else; avoiding all connection with contiguous atoms, it associates neither with the individuals of the alkaline nor with those of the acid series. Still there is a way of getting rid of this exceedingly troublesome substance, that is, by letting it go with the masses of cinder. Titanic acid does not melt by itself, nor with anything else. If present to the amount of ten per cent, in the ore, a neutral matter, amounting to 330 pounds, will thus exist in every ton of iron, that is, if we take only a ton and a half of ore to the ton of metal ; but, on an average, two tons of ore are re- quired to produce a ton of metal. If but a tenth part remains in the furnaces, it will speedily accumulate, and obstruct the passage of the cinder; and this obstruction no heat can remove. Prot- oxide of iron, though in a very low degree, and its isomeric com- pounds, are solvents of titanium; but the quantity required for this purpose is so large that we cannot think of making use of them in the blast furnace. On this principle the Catalan forge is con- ducted ; the titanium of the ore is carried off by the subsilicates of that process. At the blast furnace, all our endeavors are directed to the extraction of every particle of iron from the ore. We thus act in a manner precisely the reverse of that in which we ought to act in this case. Instead of being indifferent concerning the loss of a small quantity of ore, the heat is increased, and most of the iron revived ; while the titanium, thus deprived of all chance of leaving the furnace peaceably, remains with some of the cinder 236 MANUFACTURE OF IRON. as a cold, pasty mass, which the hottest blast will not soften into a fluid slag. ee. If we wish to revive most, or all of the iron, in cases where titanium is the enemy against which we have to contend, our most successful plan of operation is to increase the foreign matter, and to form a large quantity of a more or less fusible cinder, according to the quality of metal to be made. In this case, as in other cases, it is not advisable to employ pure silex, or pure clay or lime ; but to select ferruginous clay, ferruginous slate, or lime containing iron. Pure lime and pure clay are injurious in every instance. Titanium is subject to the general law of the solubility of the oxides of metals in silicates ; and the more fusible such silicates are, or the lower the temperature at which they melt, the greater is their dissolving power. Hence, where the amount of titanium is large, the amount of cinder should be uncommonly large, to produce gray iron; but if we are indifferent about the loss of a little ore, and smelt white forge metal at a low temperature, the fluxing of the furnace requires but little foreign matter. Ores containing titanium may be con- sidered very favorable for the manufacture of steel metal; in many respects, they are preferable to the spathic ores ; for, with very little attention, they will produce white iron with a large amount of carbon, the very material from which German steel is manufactured. For this purpose, a high stack, and a low hearth, or none at all, like the Styrian furnaces, are required ; as well as the addition of a flux which shall carry off the titanium. A sandstone hearth would not answer so well as a hearth of granite and gneiss. ff. Cinders No. 4 and No. 5 possess but little interest ; but Nos. 6 and 7 are taken from a furnace of which we have personal know- ledge. This furnace is known to produce a first rate article, from which German steel for the Solingen market is manufactured. No. 6 was taken while the furnace labored under too heavy a burden ; the metal produced was white, serviceable for the manufacture of bar iron. No. 7 was derived from the furnace when in good order, and while smelting gray iron, a kind of foundry metal. But for this purpose the furnace is seldom employed ; because the re- gion in which it is situated abounds in rich spathic ore, and sup- plies no ore of inferior quality. This rich spathic ore is scarcely at all adapted to produce soft gray iron. From the same furnace we have a third specimen of cinder from a different source ; this cinder was made when the furnace was smelting steel metal, that is, a white, crystallized metal, containing a great deal of carbon. REVIVING OF IRON. 237 As the manufacture of steel metal is carried on, in this country, only to a very limited extent, notwithstanding we possess ore and fuel in abundance" sufficient to relieve us from the tribute we at present pay to Europe for steel, we shall make some remarks which may be useful to those who design to engage in its manufacture. "VVe shall call the following specimen of cinder No. 12: Silex - 48.39 Lime " Magnesia - 10.22 Alumina - 6.66 Protoxide of iron - - .06 " " manganese - - 33.96 The ore employed in cinders Nos. 6, 7, and 12 was of the same composition. Nothing but its burden was changed. In No. 6 it was heaviest ; in No. 7 lightest. It will be observed that there is a considerable increase of silex from No. 6 to Nos. 12 and 7. No. 7 contains the largest amount of silex, but scarcely any iron. The iron contained in No. 6 is replaced by magnesia, alumina, and manganese. Scientific investigation shows us that the cinder from gray iron contains fifty per cent, more oxygen in its silex, in pro- portion to the oxygen of the alkali, than the cinder from white iron, No. 6, contains ; the latter is almost a single silicate, in which the oxygen in the acid is equal to the oxygen in the alkali. The characteristic feature of cinder No. 12 is that it contains no lime. This is an important circumstance. The lime is replaced by manganese ; but we cannot expect, in every instance, to find manganese in quantities sufficiently large to flux the cinder. This remark applies especially to the magnetic ores of this country. Therefore, if we wait until we find an ore which can be fluxed by the manganese it contains, before we succeed in manufacturing steel, we shall be under the necessity of waiting a long time. . An addi- tion of black manganese will be highly serviceable ; but this can be only partially applied ; partly on account of its expense, and partly because of the limited quantity in which it is found. Neither lime, magnesia, nor any of the alkaline earths, are of any use. Protoxide of iron is inefficient, because, in spite of all our efforts, it will be dissolved by the temperature of the furnace, and the amount of carbon present. The only resource which remains is the alkalies proper, that is, potash or soda. Soda is preferable to potash, as we shall, in the following chapter, more fully show. 238 MANUFACTURE OF IRON. Lime, in this instance, does not answer the purpose of a flux, for the following reasons: Metal adapted for the manufacture of German steel should contain a large amount of carbon, and be as free as possible from foreign matter; these are objects accomplished with great difficulty in the blast furnace. We are enabled to com- bine a large amount of carbon with iron, in the blast furnace ; as in the case of gray anthracite iron, or the charcoal iron of Hanging Rock. But this object is always effected by means of a strong, silicious cinder ; and such iron contains a large amount of silex. But this iron, though an excellent forge metal, is not adapted for the manufacture of steel. The combination of the revived metal with carbon may be ef- fected with comparative facility, as we have before demonstrated. But in the present case, we need a metal free from foreign matter; therefore it is requisite that we employ an ore as free as possible from foreign matter. We have an abundance of such ores in this country from Maine to Alabama, and from Iowa to Texas ; but the usual method of conducting blast furnace operations will not enable us to produce the required metal. Silex and clay, if they are pre- sent in the ore, do no harm to the metal ; but lime is injurious. Lime facilitates the reviving of iron in a higher degree than any other alkali. While it protects the bright surface of the metal, it will prevent, and sometimes even dissolve, the combination of iron and carbon. For these reasons, lime is inapplicable to our purpose. Still another reason is, that the affinity of lime for silex is not suffi- ciently strong to prevent the combination of silex and iron : and in the presence of a surplus of carbon, silex will be reduced to silicon, and, combining with the iron, will make it brittle, and useless for the manufacture of steel. The application of soda or potash in furnace operations, as a means of fluxing, has been recommended by various writers ; but we are not aware that a successful experiment has ever been made. While the application of these fluxes will improve the metal for the forge, it will impair its malleability as a foundry metal. gg. The hearth and in-wall of a furnace suitable for the manufac- ture of steel require a thoroughly different construction from those of ordinary furnaces. The material employed at common furnaces cannot resist the action of strong alkalies ; but of the material of which a hearth should be constructed, we shall speak in the next chapter. A different internal form from that of common furnaces REVIVING OF IRON. 239 is required. The interior should be high ; there should be no hearth, or a very low one. The blast should not be too strong, but in abundance. Very rich ores are desirable, in case artificial flux is to be employed, for expenses will augment in proportion to the amount of foreign matter contained in the ore. It is worthy of remark, that steel metal can be manufactured, so far as charcoal is concerned, at a very small cost; for, in Styria, where a large num- ber of furnaces produce this metal, less fuel is consumed than in any other blast furnaces in the world. The analyses of cinders Nos. 6, 7, and 12 show conclusively that cinders from the same furnace, from the same ore, and produced, with the exception of burden, under the same conditions, differ greatly in composition. Therefore we should be cautious in draw- ing conclusions from analyses of cinder, and avoid hasty imitations. We do not always know to what kind of work the cinder belongs. The theory of the artificial composition of cinder, which has of late been so highly developed, may, while it is useful to the utilitarian, seriously mislead the speculator, who, in his eagerness to secure profitable results, fails to examine whether his conclusions are drawn from sound or insufficient premises. The application of theories is accompanied with difficulty, because the science of the manufacture of iron is far in advance of the practice. The rules with which science has furnished us in relation to the rudiments of the busi- ness have, thus far, been applied only to a very limited degree; therefore, we cannot expect that improvements, based upon con- ditions thus incompletely fulfilled, will be altogether successful. Speculative minds are too little disposed to notice slight imperfec- tions; but these imperfections constitute the greatest obstacle to the progress of the business. Were they properly estimated, and due pains taken to correct them, the United States would be enabled, in a few years, to compete against the world in the manufacture of iron. hh. We conclude that cinder No. 10, on account of the large amount of lime it contains, produced red-short iron and white metal. No. 9 is decidedly of better quality; and close investigation will show that this cinder produced gray metal. To enter into details upon this subject would probably be less acceptable to the reader than to present the subject in as brief and significant a manner as possible. We infer that No. 10 produced white iron, because the amount of oxygen in the alkali is greater than that in the silex; whence it follows that the cinder is a basic or subsilicate. To make 240 MANUFACTURE OF IRON. gray metal, at least a single silicate, that is, the presence of an equal amount of oxygen in the Alkali and acid, is required. The process will be more effectual if the amount of oxygen in the silex is greater than that in the alkali. This is the case with respect to No. 9 ; and, in spite of the large amount of protoxide of iron present, the cinder is the result of a good quality of metal. If not excessively gray, it is at least goocLfoundry metal, made by cold blast. No. 10 is a peculiar cinder, and is from the same furnace as No. 9. The fur- nace is said to have been in bad order; but this cannot be true, because the amount of sulphur in this cinder is 1.4 per cent. Nearly three tons of cinder, at a coke furnace, are produced per one ton of metal; therefore, should a good cinder have been made, the iron of No. 9 ought to contain three times 1.4 per cent, of sulphur. But this amount would render the iron entirely useless, even though the largest proportion of sulphur present was expelled. Be this as it may, the coal or the ore of the furnace at the Dowlais Works, in South Wales, contained a large amount of sulphur, which is visible in No. 10. We allude to this cinder especially, because it was produced under conditions which resemble very closely those which exist at the Great Western Iron Works, in Pennsylvania. The presence of sulphur in the furnace occasions great annoy- ance. In the case before us, the furnace required a large charge of limestone to produce, even at a high temperature, a surplus of lime ; for this is the best means of carrying off a certain amount of sulphur. A high temperature will produce a white cinder, streaked with various shades. This cinder contains, besides a silicate of lime, a sulphuret of lime, and is characterized by soon losing its lustre on being exposed to the atmosphere. If, under such circum- stances, the temperature of the furnace falls below a given point, the cinder changes rapidly into a pitch black, heavy mass, contain- ing a large amount of sulphuret of iron. The same circumstance happens where too small a quantity of limestone is used. In this case, the sulphur, having no free alkali with which to combine, follows the iron into the pig bed, where its presence is indicated by the odor of sulphurous acid. If the furnace is cooled below the temperature at which gray iron is usually made, the cinder, by absorbing sulphuret of iron, is soon blackened. In such cases, the smelting of gray metal is accompanied with difficulties which absorb more attention than can well be spared. In addition to the difficulty of continuing a furnace on gray iron, the metal produced? is of infe- rior quality, and unsuitable for the market. To get rid of the sul- REVIVING OF IRON. 241 phur is indispensable, for, whether we bring the pig iron to the forge or to the foundry, it is, in all cases, exceedingly troublesome. There is no resource left but an excess of limestone. This will, of course, produce white metal, and, if hot blast is employed, of very inferior quality. In this case, it is necessary to work the fur- nace with light burden, to prevent the formation of black cinder, which will absorb too much iron. The revived iron, or iron ore, in the upper part of the furnace, will be saturated with carbon ; and at the high temperature of the hearth, the silex, and even the lime, will be reduced to their corresponding metals. These metals will combine with the iron ; and having, where hot blast is employed, little chance of being oxidized, they will of course follow the iron to the bottom, and be troublesome both in the forge and in the foundry. Metal, thus produced, is so brittle and hard as to be unfit for foundry use. We have thus presented an instance in which white iron, smelted by a high temperature, contains little or no carbon. In charcoal furnaces, on the contrary, the metal contains a large amount of car- bon ; but this applies only to those cases in which no limestone, or limestone in very small quantity, is used. In this, as in every case, the disappearance of carbon results from the large quantity of lime in the furnace. The result is the same, under similar circumstances, in charcoal furnaces. The weakness of the metal is to be attributed principally to an admixture of silicon, and even of calcium both very bad admixtures with the use of hot blast. In this case, the cold blast will produce better metal than hot blast, because of the oxidation of silex by the former. By the latter the oxygen is not so quickly absorbed; the iron which sinks is more exposed to oxidation ; and of course calcium, silicon, and carbon will be sooner oxidized than iron. But of this matter we shall speak hereafter. As we have seen, cinder No. 10, compared with No. 9, contains very little iron. This cinder may be considered the regular mixture of ore and flux for the location whence it was derived ; because, if it contained less limestone, the metal, in addition to being very hot-short, would be produced in very small quantity. A surplus of limestone would produce a better yield, more easy work, and metal of good quality, however white it might be. The application of cold blast in smelting is, so far as the quality of iron is concerned, un- doubtedly preferable to that of hot blast, because of the large quantity of lime which is exposed to its action. Where the blast 16 242 MANUFACTURE OF IRON. is cold, the limestone will be chilled ; and the cinder and iron, in their passage through them, will also become chilled. Where hot blast is employed, there is a more uniform heat in the hearth, and no obstacle prevents the passage of the cinder ; because even the unmelted parts are sufficiently warm to facilitate the process of smelting, and the discharge of the fused mass. From what we have stated, we deduce the following conclusion: that the best method of using sulphurous materials is to smelt them by an excess of alkalies. The resulting metal may be gray or white. This is both theoretically and practically true. We may add, that the smelting of sulphurous minerals should, where prac- ticable, be avoided. But where we cannot avoid using them, we should employ the hot blast, and work with as low a temperature as possible, with the view of expelling silicon. Any experiment made with the object of improving the metal in the blast furnace will be likely only to augment the expenses of the iron master, without benefiting his operations. With these remarks, we shall conclude this chapter ; trusting that whatever deficiencies exist will be supplied by the intelligent reader. We are conscious of having omitted to state several slight matters ; but, though these omissions will be of little consequence to an accomplished manager, we shall, to make our work as com- plete as possible, notice them in the following chapters. We flat- ter ourselves that we have mentioned every fact and theory which has an important bearing upon the successful operation of a fur- nace. MANUFACTURE OF WROUGHT IRON. 243 CHAPTER IV. MANUFACTURE OF WROUGHT IRON. THE manufacture of wrought iron involves two fundamental operations : namely, the removal of impurities from the ore and from the crude metal ; and the oxidation of the metal to a degree sufficient to form fibres. The first, which consists in the removal of impurities, and the vitrification of the earthy admixtures in the ore, is effected by a variety of methods. If the amount of im- purities in the ore is large, an intermediate method is employed to remove them ; that is, metal of greater or less purity is manu- factured in the blast furnace. But if the amount of impurities is excessive, or if the metal from the blast furnace is very impure, as is often the case in gray charcoal pig, and quite generally the case in anthracite and coke iron, the refining fire is resorted to, before the metal is subjected to the process by which it is converted into bar or fibrous iron. The second operation, though effected by a variety of methods and by a diversity of apparatus, consists mainly in a semi-fusion of the metal. In this condition, it is stirred and worked by manual labor, with the object of exposing the smallest particles of the metal to the influence of the atmospheric oxygen. In this chapter, we shall endeavor to describe the principal forms of apparatus at present in use in various parts of the world, and especially in the United States ; we shall attempt to indicate the methods by which the operations in the manufacture of wrought iron are performed ; and we shall close the chapter by some theo- retical investigations to which we would invite the attention of the manufacturer no less than the philosopher. I. Persian Mode of making Iron. The most ancient method of manufacturing iron is at present practiced in Persia ; and, as far as we can ascertain from the pub- lished reports of travelers, whose descriptions, while they slightly vary in detail, agree in relation to the uniformity of the principle, 244 MANUFACTURE OF IRON. this method is practiced throughout Asia. The manipulation is as follows : A hearth, or a mould with fine charcoal, or clean charcoal dust that is, a semicircular hole from six to twelve inches in depth, and from twelve to twenty-four inches in width, as represented in Pig. 71 is formed. The darker shading in the figure illustrates Fig. 71. Ground-plan of a Persian forge fire. the lining of charcoal dust ; the form of this lining is sometimes round, and sometimes square. Before the dust is put into the hearth, it is moistened, well mixed, and pounded as closely as pos- sible. The lining will, of course, be perfect, in proportion to the fineness of the dust. The bottom especially should be hard, to resist the action of the blast. Into this basin, the blast is conducted by means of a clay tuyere, or a piece of crockery, situated a short distance above the bottom of the basin. The bellows are urged by Fig. 72. Asiatic or Persian method of making iron, manual power. Fig. 72 exhibits a section of the basin, and the situation of the bellows. In the bottom of the basin, medium-sized charcoal is laid to the height of several inches, covered by a layer MANUFACTURE OF WROUGHT IRON. 245 of ore in pieces of the size of hazelnuts. Where no compact ore can be obtained, the fine ore may be cemented by being moistened, and then dried and broken. But the native compact ore is prefera- ble, because it contains fewer impurities. Upon this layer of ore a layer of charcoal is placed, and then alternately ore and charcoal until five or six strata are piled. The whole is covered by charcoal of moderate size, firmly pounded. Fire is then introduced at the tuyere, and the bellows gently moved, so as to expel all the water jcontained in the mass, before a full heat for the reduction of the ore is given. When the water is supposed to be driven off, the bellows are urged more strongly, and the heat increased. The ore is then reduced, and iron liberated in a metallic state. The whole process lasts from three to four hours, at the end of which time twenty-five or thirty pounds of iron may be removed by tongs, and forged by means of sledgehammers. Of course, the desirable shape is not produced until the metal is heated and re-heated seve- ral times. After the iron from one heat is forged, the clinkers are removed, and another coating of charcoal thrown on ; in fact, a renewal of the whole process is required. In this process, none but the best kind of red iron ore, or specular iron, is used; and it is questionable whether any but the richest of this ore can be employed. The iron manufactured is very strong and tenacious; from which the sabres of Damascus, and the neat and delicate, though very powerful Damascene gun barrels, as well as weapons of nearly every kind, are wrought. In the States east of the Mississippi, no ore or at least no ore in large quantity suitable for the manufacture of such articles, is found ; but it is probable that it may be obtained in Iowa, Missouri, and along the borders of the Rocky Mountains. In Arkansas, large deposits exist. This ore is seldom found anywhere else than in transition clay slate, or roofing slate. II. Catalan Forge. This forge is extensively employed in Vermont and New Jersey, to smelt the magnetic ores of these States. It is there called the blomary fire. The form of this fire is nearly uniform everywhere. Fig. 73 represents a Catalan fire, seen from above. The whole is a level hearth of stonework from six to eight feet square, at the corner of which is the fire-place, from twenty-four to thirty inches square, and from fifteen to eighteen, often twenty, inches in depth. 246 MANUFACTURE OF IRON. Inside it is lined with cast iron plates, the bottom plate being from Fig. 73. Ground-plan of a forge fire. two to three inches thick. Fig. 74 represefats a cross section through the fire-place and tuyere, commonly called the iron, a Fig. 74. Blomary fire. MANUFACTURE OF WROUGHT IRON. 247 represents the fire-place, which, as remarked above, is of various dimensions. The tuyere b is from seven to eight inches above the bottom, and more or less inclined, according to circumstances. The blast is produced by wooden bellows of the common form, or, more generally, by square, wooden cylinders, urged by waterwheels. The ore chiefly employed is the crystalized magnetic ore. This ore very readily falls to a coarse sand, and, when roasted, varies from the size of a pea to the finest grain. Sometimes the ore is employed without roasting. In the working of such fires, much depends on the skill and experience of the workmen. The result is subject to considerable variation; that is to say, the result depends on the circumstance whether economy of coal or that of ore is our lead- ing object. Thus, a modification is required in the construction either of the whole apparatus, or in parts of it. The manipula- tion varies in many respects. One workman, by inclining his tuyere to the bottom, saves coal at the expense of obtaining a poor yield. Another, by carrying his tue iron more horizontally at the commencement, obtains a larger amount of iron, though at the sacrifice of coal. Good workmen pay great attention to the tuyere, and alter its dip according to the state of the operation. The gene- ral 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, in case any is at our 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 urged gently, and directly upon the ore ; while the coal above the tuyere is kept cool. Four hundred pounds of ore are the common charge, two-thirds of which are thus smelted ; and the remaining third, generally the finest ore, is held in reserve to be thrown on the charcoal when the fire becomes too brisk. The charcoal 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 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 scoriae are permitted to flow through the tapping- hole a, so that but a small quantity of cinder remains at the bot- tom. 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 248 MANUFACTURE OF IRON. the bottom directly before, or to a point above the tuyere, until it is welded into a coherent ball twelve or fifteen inches in*diameter. This ball is brought to the hammer or squeezer, and shingled into a bloom, which is either cut in pieces to be stretched by a hammer, or sent to the rolling mill to be formed into marketable bar iron. a. A mixture of fibrous iron, cast iron, and steel an aggrega- tion of unavoidable irregularities is the result of the above pro- cess. The quality of the iron depends entirely upon the quality of the ore. No opportunities are presented by which any skill or in- genuity can create improvements in this process. Poor ores cannot be smelted at all ; but rich ores, like those at Lake Champlain, or in Missouri, or even the hydrates of Alabama, may be smelted to advantage ; the latter with a prospect of economy. In some coun- tries, where much larger fires than the one we have mentioned are employed, balls of 200 or 300 pounds weight are produced ; but such large masses cannot be worked with facility, and are always of inferior quality. It is not advisable to make, at one smelting, balls heavier than 100 pounds. In Vermont, where the rich magnetic ores are employed for this kind of work, a ton of blooms costs about forty dollars. To produce this quantity, four tons of ore and three hundred bushels of char- coal are required. Wages of workmen per ton ten dollars. b. An improvement upon the Catalan forge is the stuck oven described in our third chapter. But little explanation is required to exhibit the connection between the two manipulations. So heavy are the masses of iron in the stiick oven, that powerful ma- chinery, as well as a large number of workmen, is required in working them. The salamander, when lifted out of the furnace, is cut into pieces of 100 or 150 pounds weight. These pieces are re- heated in a common forge, or Catalan fire ; a portion of the cast iron melts out of it ; and what remains is generally the best iron, and called the "blume" or flower. From this term it is probable that our English word " bloom" is derived. Sometimes the bloom is in part steel, according to the state of the furnace, and the kind of ore used; but, generally, it is fibrous iron. The cast iron which results from the re-heating is worked, by the common method, into fibrous iron. So expensive is the operation in the stiick oven, and so im- perfect the iron which it produces, that this furnace is now generally abandoned. Both the Catalan forge and the stiick oven are imprac- ticable where ores containing less metal than forty per cent, are to be smelted. Any foreign matter in the ore is injurious. MANUFACTURE OF WROUGHT IRON. 249 Many ingenious contrivances have been devised to convert ore, by one manipulation, directly into bar iron or steel. These con- trivances are local, and vary according to the quality of the ore, and the intelligence of the operator. They are not worthy of our notice ; in our country, at least, they are of no practical utility. III. German Forge. The most successful method of manufacturing charcoal wrought iron is by means of the German refining forge, or the blomary fires of Pennsylvania. These forges not only produce good iron at reasonable prices, but they afford all the facilities presented by differently constructed apparatus. Fig. 75 represents a section of Fig. 75. Forge fire. the German forge, in which the location of the hearth is shown. The hearth is lined with cast iron plates ; the bottom is generally kept cool by a current of water circulating in pipes below it. Three or four inches above the bottom, there is a row of holes, through which the cinder is let off. Through the tuyere, which is hollow, a current of water circulates. Frequently, the back part of the hearth is raised, for the purpose of putting in hot air pipes, as shown 250 MANUFACTURE OF IRON. at c, Fig. 77. Fig. 75 exhibits a better arrangement for heating the blast, d represents a hollow roof plate, of sheet iron, through Fig. 76. Forge fire. *vhich the blast passes. The blast for this kind of fire is produced either by wooden or by iron cylinder bellows. Fig. 77. Heating the blast on a forge fire. MANUFACTURE OF WROUGHT IRON. 251 a. The material employed in this forge are the various kinds of pig metal, from the white pig which contains a small amount of carbon, to the gray and white metal which contains carbon in con- siderable quantity. The construction of the apparatus depends to scarcely any extent upon the kind of metal worked. By varying the treatment of the material, in the course of the manipulation, the same results may, in the great majority of cases, be produced. Gray metal requires a higher heat at first than white metal; besides, more time is required, and a greater amount of fuel consumed, in working it. In this, as in every case, white metal containing a large amount of carbon is, when smelted by hot blast, the worst of all metals for the manufacture of a satisfactory iron. Should gray metal, smelted from the same stock as white metal, by small bur- den, work slowly, and consume more fuel than the latter, there is a greater prospect of producing a better article. An exception from this rule occurs only where ore of the first quality is smelted by charcoal and cold blast. White metal, smelted by small burden and poor ore, or anthracite or coke, is always inferior; it works so fast that no time is afforded for the removal of the impurities com- bined with it. Where gray metal is worked, a greater chance of purifying the melted iron before it comes to nature is presented. These remarks do not apply to white iron made by charging the blast furnace with a heavy burden of ore. At the close of the chapter, we shall speak at greater length on this subject. In this place, we simply wish to draw the attention of the manufacturer to this important matter; because his success depends, in a very high degree, upon a clear understanding of the qualities of the metal, and of the modes of working it. b. The form of the basin of the hearth, the height of the tuyere, and the pressure and the quantity of the blast, cannot be fixed by any general rule, and depend on coal, metal, and the workmen. We shall endeavor to explain the leading principles in each case; but their application is to be varied according to local circum- stances. c. The quality of the charcoal determines, to a certain extent? the dimensions of the hearth, the dip of the tuyere, and the pres- sure of the blast, as well as the amount of metal to be smelted by one heat. Soft coal, from pine wood and poplar, requires a larger and deeper fire, a greater dip of the tuyere, and weaker blast, than charcoal from hickory and maple. Soft coal works more slowly than hard coal. Dirty or sandy coal, which has been received from the 252 MANUFACTURE OF IRON. coalings in wet weather, or which has been exposed to sand or mud in the yard, should be refused altogether, or at least carefully dried and cleaned; because each pound of sand removes three or four pounds of iron, for no conceivable purpose whatever. Coal that is not larger than an egg will answer tolerably well for a blomary fire ; but strong and heavy coal is far more serviceable. In the prepara- tion of coal for the blomary and the blast furnace, we are guided by different rules; that is to say, the small coal from the coalings is taken to the former, and the large coal to the latter. It should always be remembered that coal which contains sand and dirt is far more injurious in the forge than in the blast furnace. d. So great are the practical difficulties in this kind of work, that fluxes are employed to a very limited extent. Hammer-slag is some- times used as a flux. This is a very useful material ; it facilitates the work, and improves the iron. Water is another means of flux- ing. In the first place, it cools the iron, and thus gives the blast a chance of playing upon it to advantage. In the second place, it is decomposed by red hot iron; its oxygen is retained, and forms peroxide of iron, which readily combines with silex. For the pur- pose of retarding the working of the iron, workmen are in the habit of throwing sand, dry clay, or even loam upon the fire. This is inexcusable ignorance, or at least gross sluggishness. If proper care and industry are exercised, the iron will not come too fast, and if it does, the application of sand will only augment the evil. At any rate, it is highly injurious to the quantity and quality of the iron produced. e. The fire-place is always of a square form, varying in size ac- cording to circumstances, and lined throughout with cast iron plates. To secure the stability of the apparatus, these plates should be well screwed together. The plate nearest to the workman," called the cinder or work-plate, should especially be solid, because a great deal of the manipulation is done on this plate. The plates around the fire are generally laid in a somewhat sloping position, more for the purpose, we presume, of facilitating the lifting out of the bloom, and of making the dust stick, than for anything else. The tuyere plate is generally inclined into the hearth, partly with the object of facilitating the dipping of the tuyere, and partly with the object of cooling that plate. The blast is regulated by a valve (Fig. 76), which should be as close as possible to the tuyere, to facilitate the labors of the workmen. The tuyere, as before re- marked, is a water tuyere, which, after its inclination has been de- MANUFACTURE OF WROUGHT IRON. 253 termined, is firmly fastened in its place. The nozzle of the blast pipe is to be movable; it can thus be dipped into the hearth, or moved horizontally, so as to drive the blast towards any portion of the hearth. The width of the tuyere and the nozzle depends on the quantity of metal melted at once ; but it seldom exceeds two superficial inches. The form of the tuyeres is generally that of a half circle Q ; but sometimes that of a circle. The depth of the fire-place varies according to the quality of the metal to be refined. Nine inches will be sufficient for mottled or gray iron : but white iron, of small burden, requires a depth of ten or twelve inches. The better the metal, the less the depth of hearth required. A deep hearth consumes more fuel, and works more slowly, than a shallow hearth ; but if worked with proper attention, the product is supe- rior. Inferior workmen require deep fires, as well as hot-short and cold-short metal. Shallow fires work fast, but require excellent metal, or very expert workmen ; because, if anything goes wrong in them, heavy losses are experienced. A number of workmen draw the tuyere, to a greater or less extent, into the hearth. This is productive of no real benefit. Though by this means we are enabled to carry the blast a little farther into the fire, the advantage we derive from it scarcely amounts to anything. If it is our desire that the metal should flow down, directly through the current of the blast, our object may be effected if the tuyere reaches but a short distance into the coal. Besides, by drawing the tuyere, its plate is sooner destroyed than where the tuyere is close. We must regulate the dip of the tuyere according to the metal we are smelting. Gray, and hot or cold- short metal, and soft coal, require a greater dip than good metal and hard coal. Nevertheless, the whole dip does not vary more than from 7 to 15. /. The manipulation at a blomary fire varies according to the quality of the metal with which we have to deal. Therefore, our descriptions must be restricted to one particular metal, or at least to a metal smelted from the same kind of ore. We shall confine our attention to gray or half gray metal, and introduce such remarks occasionally as will indicate the method by which other kinds of metal are worked. When the hearth is in proper order, and when the blast and everything else are prepared, the interior of the fire-place is lined with charcoal dust, which ought to be free from sand and other 254 MANUFACTURE OF IRON. impurities. The finer the dust, the more serviceable it will be. Good fire clay water thrown over it will, when the whole mass is well pounded, make it more adhesive. Pieces of refractory ore, of good quality, and even good qualities of cinder, are sometimes used in lining. These materials are decidedly preferable to char- coal; but they require more attention on the part of the workmen. Coal is then thrown on, and the fire kindled. As the fire rises, iron may be melted down. The amount of coal, or the size of the heap above the tuyere, varies according to the metal. A height of from twelve to eighteen inches is sufficient for gray metal ; but for white metal the height of the coal should be twenty-four inches. When the fire has thoroughly penetrated the coal, the degree of which should be higher for white than for gray metal, the broken metal is thrown in parcels weighing from seventy to eighty pounds upon the top of the coal above the tuyere. But white metal is thrown more from the tuyere, in case a charge of more than 120 pounds is melted at once, which is frequently accomplished where the metal is good, and competent workmen are engaged. Where the charges are small, all the metal is thrown on at once. The blast is applied moderately; the tuyere is made to dip very slightly; coal is con- stantly supplied ; and in case iron is left to be smelted, it is thrown on the heap. In due time, the iron melts into the bottom of the fire-place, and is more or less liquid. If the metal is gray, and if it contains a large amount of foreign matter, the process proves to be slow, for there are no indications of solidification. In this case the workman proceeds to increase the dip of the tuyere, to blow upon the iron, and to stir it repeatedly by means of iron bars. If, in consequence of this manipulation, the cinder increases on the top of the iron, it may be let off, and thrown away, for such cinder is useless. When this is removed, the iron will be exposed in a greater degree to the action of the blast. If the metal still shows no signs of becoming pasty, hammer-slag or rich iron ore may be thrown upon the fire, and melted down upon the iron. The metal, when mottled iron, must be bad and gray indeed if hammer-slag or rich ore fails to bring it to nature ; or if, when it is gray iron, either of these fails to bring it to a state of boiling. Iron and cinder will then rise spontaneously, and move before the blast. The workmen should take care that no metal remains in the corners of the hearth. By degrees, the cinder will subside, and the iron will become a pasty, tough mass. If it is very hard, and feels like lead, the blast MANUFACTURE OF WROUGHT IRON. 255 is increased, for a strong heat is required to weld this tough mass into a ball. By continually raising and turning the iron, it will become uniformly heated. When it becomes tenacious, it may be removed to the hammer or squeezer, and reduced to a rectangular prism, five or six inches square ; and if too long, it may be cut into pieces not exceeding in length fifteen or sixteen inches. These prisms form the blooms of our markets, and are usually sent to the rolling mills to be transformed into bars, or, generally, into sheet iron or boiler-plates. The cinder in the hearth, unless present in too large quantity, which is seldom the case, may be suffered to remain. When the scraps of iron are removed, and the lining of the hearth secured, and, if necessary, repaired, coal may again be filled in, and the blast turned on. White iron, and even mottled iron, very seldom boil ; but, by proper treatment, arrive as a pasty mass at the bottom of the hearth. This mass should be broken up, and brought, to a greater or less extent, within the range of the blast. But this manipulation requires caution, where we have to deal with white anthracite iron ; because this metal commonly works too fast, and, if its propensity is favored, bad iron results. Taking care to prevent the iron from touching the tuyere, we should keep it in a pasty state as long as possible, to afford the impurities an opportunity of combining with the cinder. When the bottom of the hearth is very cold, it is possible that the metal may be gray or fusible, though the part which touches the hearth may be hard and cold. In this case, the iron, whether in mass or in pieces, should be carefully brought above the tuyere, and once more melted down. In the mean time, it is advisable to discontinue the cooling of the bottom plate. g. The tools required at the blomary fire are very simple. A few implements like crowbars ; several pairs of tongs for lifting the bloom from the fire ; and a couple of chisels shaped like hatchets, for cutting blooms, are all we require. Of the means employed to reduce the balls to a proper size, we shall speak elsewhere. 7i. The results and the expense of this branch of iron manufac- ture of course vary greatly. One kind of metal will yield ninety or ninety- two per cent, of blooms; while another kind will yield but eighty per cent., or even a less per centage than that. The same difference may be observed in relation to the quantity of fuel re- quired. The number of bushels of charcoal varies from 150 to 250 per ton of blooms. A blomary fire, conducted night and day 256 MANUFACTURE OF IRON. for a week, furnishes from four to seven tons of blooms. Wages of workmen from six to seven dollars per ton. At various places, hot blast is applied with success; but at other places, with but little advantage. In most instances, it is not employed. In Europe, the varieties of blomary fires, or refineries, are in- numerable. These varieties depend, in a greater or less degree, on the nature of the coal and iron used, and upon the habits of the people. Many arrangements produce better iron than that we have described. But the advantages which these possess depend upon circumstances which are not available in the United States, The only ore which does not follow the general rule of the oxides and hydrates is the magnetic ore of the different States. This ore might be made to produce as good iron as the best Swedish. But our furnace owners, in their inconsiderate eagerness to realize every possible advantage, often produce a metal whose quality cannot be at all relied on. Our forge owners, therefore, when they desire a good article, are generally compelled to do the best they can in relation to economy of fuel. IV. Finery Fire. We shall now describe a process intermediate between the blast furnace and the forge ; that is, the finery or run-out fire. A descrip- tion of this should have followed that of the blast furnace ; but as it is of later origin, and will, besides, be better understood after the explanation of the German finery, we thought it advisable to delay our notice of it. This invention is the result of necessity. The introduction of stone coal, coke, and hot blast occasioned so much bad pig iron, that some means which should remove a portion of the impurities in the metal before its removal to the charcoal forge or to the puddling furnace were eagerly sought. The necessity of an intermediate process will be readily admitted : but a more awk- ward and unprofitable invention than that we are considering could not have originated from the most unskillful intellect. The appa- ratus is so worthless as scarcely to deserve notice. In fact, when we see the large amount of iron which is converted into slag ; when we see the best charcoal iron wasted by the Western manufacturers, we are justified, we think, in wishing that the apparatus had never been invented. But the invention exists, and there is no imme- diate prospect of getting rid of it ; therefore it is our duty to re- cord its existence, and to exhibit its construction. MANUFACTURE OF WROUGHT IRON. 257 Fig. 78 represents a vertical section, and Fig. 79 a ground-plan, of a finery. It is erected on a platform of brick, about twenty inches in height, in the middle of which is the hearth or fire-place A. At each of the four corners an iron column is erected, upon Fig. 78, Finery, or run-out fire. 'which a brick chimney, two feet in width inside, is built. This fire generally works with four tuyeres, that is, two on opposite sides; or with four nozzles, and but two tuyeres, on the same side. When the latter is the case, two currents of blast are conducted into each tuyere, that the whole surface of the melted metal may be ex- posed to the action of the blast. The sides of the hearth are formed of hollow cast iron plates, through which a current of cold water is constantly running, to prevent their melting. The hearth is gene- rally from three to three and a half feet in length, twenty-four inches in width, and twenty-four or thirty inches in depth. Around the fire are sheet iron doors, fastened to the columns; these are alternately used to prevent the disturbance occasioned by strong draughts of wind. Such fires produce a great deal of dust, heat, and rubbish, 17 258 MANUFACTURE OF IRON. and are generally removed from the main buildings. The bottom of the hearth is formed of coarse sand, and often of coke dust. The nozzles a a are from an inch to an inch and a quarter wide. The blast, of which about 400 cubic feet per minute are required, is pro- Fig. 79. Ground-plan of a finery. duced in iron cylinders. In a prolongation of the tapping hole 5, is the chill mould ; this is a heavy, cast iron trough, sufficiently large to receive the contents of the hearth. It is commonly ten feet in length, thirty inches in width, and four inches in depth. A current of water is led around it to keep it cool. Pipes, suffi- ciently large for throwing a strong current of water upon the hot iron, should be at our disposal at all times. When the hearth is ready for operation, fire may be placed on it, and coke, or, as in many instances, charcoal thrown on ; the blast is then applied, and pig iron, to the amount of five or six hundred pounds, charged at once. If the iron is very gray, a greater dip MANUFACTURE OF WROUGHT IRON. 259 of the nozzles and of the tuyere is given ; this secures stronger blast upon the metal, which, after being charged, soon comes down. When the iron disappears from the top, another charge is given, and melted down, care being taken that the coke is duly supplied. In this way, twenty pigs, or generally one ton of iron, are melted, ^ome time is required, where the iron is gray, before the metal can be let out ; and when sparks of burning iron appear to be thrown off from the top of the coke, this time is supposed by the workmen to have arrived. After the lapse of about two hours, the time re- quired for one heat, the tapping hole is opened, and the iron runs into the chill mould, or, as it is called by the workmen, the pit. This mould has been previously washed with a thin clay solution, to pre- vent adhesion of the refined metal to its surface. Cinder also flows off with the iron. As soon as the metal becomes solid, a strong cur- rent of cold water is permitted to flow upon it, the reason for which we shall explain in another place. In the mean while, fresh coke and pig iron are charged, and the process continued as before. a. The quality of the refined iron depends principally upon that of the pig metal ; while the quality of the latter depends upon the previous manipulation in the blast furnace. Good soft gray or white iron generally furnishes metal of excellent quality ; but white, hard, and brittle pig is very little improved in the finery. There have been cases in which 2300 pounds of pig produced a ton of metal ; and we have known instances in which 3000 pounds of coke iron were used to produce the same amount. It is beyond human skill to suggest any method by which a waste of iron, to a greater or less degree, can be prevented. To what extent this kind of work answers its purpose as a fore- runner of the finery forge and puddling furnace, we shall investi- gate at the close of this chapter. Some years since, Mr. Detmold, of New York, introduced an improvement upon this mode of refining. He constructed a reverberatory furnace, resembling in form the puddling furnace. The pig iron was melted on a large hearth, and the blast thrown upon its surface to whiten it. But there is little merit in either of these refineries. V. Puddling Furnaces. The reverberatory or puddling furnace is, unquestionably, of all arrangements, the best adapted to convert cast iron into bar iron. The imperfect results which have hitherto been obtained with re- spect to the quality of iron, have, as might have been expected, 260 MANUFACTURE OF IRON. depended upon a variety of circumstances. We shall endeavor to give a clear and comprehensive insight into the whole manipulation, practically and theoretically ; for we consider this the most import- ant subject in our treatise. How far we shall succeed in our in- tentions, the intelligent reader will, of course, be most competent to judge. An historical sketch of this branch of the iron manufacture would require more space than we can spare. It would, besides, be of little interest to show in what manner the puddling process is per- formed on a sand bottom, or even on a bottom of coke dust. We shall, therefore, simply describe the process of the present day ; and, while we shall principally dwell upon the arrangements em- ployed in the United States, we shall notice some of the most in- teresting ones in Europe. a. At Pittsburgh, and throughout the West, the single furnace, and on the eastern side of the Alleghany Mountains the double furnace, are generally employed. The former is the most an- cient form of the puddling furnace, and for this reason is probably generally used in the Western States. Fig. 80 exhibits a side ele- Fig. 80. Elevation of a puddling furnace. MANUFACTURE OF WROUGHT IRON. 261 vation of such a furnace, including the stack. It represents the work side at that point of view from which the door to the interior can be seen. The stack, or chimney, is generally from thirty to forty feet in height, and erected upon a solid foundation of stones ; this foundation is covered with four, or, in many cases, with but two cast iron plates ; upon these plates, four columns of cast iron are erected, forming four corners of the chimney. A square frame, formed of four cast iron plates, is laid upon these columns ; and upon this frame the chimney is erected. The exterior or rough wall of the chimney, width nine inches, is made of common brick, and well secured by iron binders, which are generally flat hoops, from one- eighth to three-sixteenths of an inch thick. These hoops occupy no more space than a layer of mortar ; and they should be placed at intervals of two or three feet, or sometimes of even three or four feet, while laying the brick. The binders, into which an ob- long hole should be made, should overlap the brick about two and a half inches; through this hole a bar three-fourths of an inch square may be pushed. Two of these upright bars, which should extend the whole height of the stack, are required at each corner. The top of the chimney is covered with a cast iron plate ; but this is sometimes dispensed with. Such a top-plate is a useful appendage, for it secures the bricks: but, if not properly made, it is troublesome ; it is apt to break into halves, and fall down, under the influence of heat. To prevent this accident, it is advisable that the top-plate should be formed of four pieces screwed together ; the points in the corners should be left open, to give room for ex- pansion from the centre of the plate. Fig. 81 represents such a plate from above, and Fig. 82 in a ver- tical section, with a portion of the brick work of the chimney. The interior of the stack should be built of good fire brick; for single furnaces sixteen, and for double furnaces from eighteen to twenty, inches square. The frequent expansion and contraction of this lining under a high heat affect its durability. A space of an inch or an inch and a half, left between the rough wall and the Plan of a chimney lop< in-wall, with a brick occasionally pro- jecting, will, to a great degree, prevent contraction. Fig. 82 ex- hibits the arrangement of the in-wall and rough wall distinctly. A 262 MANUFACTURE OF IRON. wire reaches from the damper on the top to the side of the furnace , the most convenient place for the workmen. Fig. 82. Fig. 83. Chimney top. I. The exterior of the furnace, eleven or twelve feet in length, and about five feet in height, is composed of cast iron plates. Into the small square hole, coal is thrown. The large one is a sliding door for the charge and discharge of the iron ; the hole in this door is designed for the introduction of the tools. The door is suspended on a chain, fastened to a lever, which is above the head of the workman. Fig. 83 re- presents the door on a large scale, in which a front view, a vertical and horizontal section, are shown. The average size of this door is twenty-two inches in width, and twenty-seven inches in height. Its inside towards the fire is filled with fire brick, tightly wedged in. The square work hole is very much sloped inside, to enable the work- man to reach every part of the furnace hearth. Fig. 84 exhibits a vertical sec- tion of the furnace and the stack. The whole arrangement is a judicious one. The structure is built of fire brick and common brick ; the former is indicated by the lighter, and the latter by the Door of a puddling furnace. MANUFACTURE OF WROUGHT IRON. 263 darker, shade of lining. The fire-place is a separate chamber, de- signed for nothing else than the combustion of fuel. Behind the fire-place is the hearth, where the iron is charged, melted, and puddled. The hearth is heated in part directly bj the flame, but Fig. 84. Vertical section of a puddling furnace. chiefly indirectly by the reflected heat from the roof, for which reason this furnace is called a reverberatory furnace. For western bituminous coal, a grate measuring three by two feet is sufficiently large; but for anthracite coal, a much larger grate is required. The hearth is five feet, sometimes six feet in length, and three and a half or four feet in width, and of an irregular form. Its bottom and sides are made of cast iron, and prevented from melting by a constant current of cold air. Where this is not sufficiently strong, a dish of water is sometimes thrown under the bottom. By care on the part of the workman, the application of water is unneces- sary. If the bottom plates are so thin as to be in danger of bend- ing, they should be supported by props made of iron rods. 264 MANUFACTURE OF IRON. After heating the hearth, the flame is conducted through the inclined flue into the stack. The size of the flue depends on that of the hearth, and upon the interior dimensions and height of the stack. A flue ten by twelve inches square is considered to be sufficiently large for a single furnace. A large hearth with a narrow and low stack requires a larger flue than a small hearth with a high or wide chimney. The dimensions of the grate increase with the incombustible, and decrease with the inflammable, nature of the fuel we employ. A grate measuring one square foot is large enough for dry wood ; while for anthracite coal a grate of twenty square feet is required. Behind the furnace, on one side of the stack, a small fire is seen burning. This fire is to be kept up at those furnaces where the fire bricks produce cinders, or where the slag from the furnace hearth passes the flue bridge. The accumulation of cinder obstructs the passage of the flame; and a small fire at the flue, with a slight draught into the chimney, keeps that part of the furnace sufficiently warm to prevent such accidents. Fig. 85 represents a section of a Fig. 85. Vertical section of a single puddling furnace. furnace on a larger scale than the above ; the furnace also is shown more distinctly. The ground-plan of this furnace is exhibited by Fig. 86, in which the form of the hearth, the plan of the fire cham- ber, grate, and the fire bridge, are clearly shown. In these illus- trations, the cast iron plates which enclose the hearth are also clearly MANUFACTURE OF WROUGHT IRON. 265 shown. These plates, about ten or twelve inches high, are made to cross the bridges, as well as to secure whatever else needs c ;?uri Fig. 86. Ground plan of a single puddling furnace. c. At Pittsburgh, and at most of the Western Works, charcoal iron is exclusively used in the puddling forges. The process con- sists of puddling and boiling. Puddling is very nearly the same thing as boiling, with slight differences in manipulation. In pud- dling, metal from the run-out fire is worked, and sometimes mixed with good white charcoal metal from the blast furnace. In boiling, the gray or mottled pig iron is brought directly to the furnace, and refined by means of slag; this iron, in the course of the manipula- tion, rises along with the cinder, and its motion is like that of boil- ing water. The latter process would, of course, be the more pro- fitable, if generally effected ; but on account of cinder, there is a limit to the boiling operation. Therefore, in a rolling mill forge, half the furnaces are employed for boiling, and half for puddling ; the latter supplies cinder for the former. d. The process of operation, in these furnaces, is as follows: A new furnace is dried slowly ; that is, a small fire is put in the grate, not quite filled with coal. This fire is usually kept up for three or four days. After the furnace is dry, which is indicated by the ces- sation of vapors from the brick work, the grate is cleared from clink- ers. A good stone coal fire is then kindled, which, in the course of four or five hours, will bring the furnace to a heat proper for charging the metal. Previous to this, the iron bottom of the hearth is covered with finely pounded cinders from a charcoal forge, or from another puddling furnace, or from a re-heating furnace. If none can be obtained, cinder from a blast furnace will answer. This cinder is broken into uniform pieces of about an inch in size. A 266 MANUFACTURE OF IRON. portion of it is thrown around the sides and bridges, and covers the bottom to the height of three or four inches. Fire should then be applied to the cinder for about five hours. By pounding it, when it gets soft, so as to fill all the crevices, the cinder will not only melt more readily, but the furnace will become more thoroughly heated. A perfect fusion of the cinder is required before iron is charged ; otherwise, it will not form a solid lining over the iron plates and bottom. But for this object alone it is employed. If crevices are left in the cinder, drops of melted iron will find them, and penetrate to the iron bottom of the hearth. Thus the thickness of the bottom is not only unnecessarily increased, but it is made rough, and occa- sions troublesome manipulation ; besides, a portion of the iron is lost. When the cinder is melted, and the bottom and sides pro- perly protected, the door is lifted, and cold cinder mixed with the melted mass. When the bottom is so far cooled that the tools make no impression on it, the metal is thrown in, the door shut, and the fire brought to good order. The door, which, as shown in the drawing, moves in a frame, is fastened by two wedges, one on* each side. These wedges are driven in between the frame and the door; for which reason, the door is about an inch smaller than the frame. Fine cinder, or hammer-slag, is thrown around the door, to prevent a draught of cool air through the crevices. In the work hole a piece of coal is laid, covered by a small plate of sheet iron. Meanwhile the door is secured by the puddler, and the helper charges coal, cleans the grate, and heats the furnace as strongly as possible. Within a quarter of an hour, the iron, in some places, begins to get red ; the helper then takes a bar, and turns the iron, that is, he moves the warm iron to a cold, and the cold iron to a warm place; after which, a fresh charge of coal is supplied. Within half an hour, if everything is in good order, the metal be- comes white, and ready to melt, when the helper, by means of a hook, breaks the pieces, and mixes the iron with the half liquid cinder ; at the same time, the puddler stirs the grate, with the ob- ject of augmenting the heat. In forty-five minutes the iron may be brought under the protection of the cinder. At this point, the divergence in the manipulations of puddling and boiling com- mences. We shall first speak of puddling; but, preliminary to this, we shall describe the tools applicable to this process. e. Most of the tools consist of iron bars and hooks. Five or six are required at each furnace. Fig. 87 represents a bar from five to six feet in length. One end of it is sharpened and square ; the MANUFACTURE OF WROUGHT IRON. 267 other end terminates in a round knob, which enables the workman to handle it with facility. The lengthier portion of the ba-'ai*." Fig. 87. Puddling bar. hook is eight-sided, for a bar of this shape is held more firmly than one that is round. These tools suffer greatly from the heat of the Fig. 88. Puddling hook. furnace, particularly when used for too long a time at once by care- less workmen. The heat is apt to slit and break the iron. For this reason, charcoal forge iron is preferable to puddled iron for tools. A water trough, six feet in length, twelve inches in depth, and fifteen inches in width, is attached to each furnace. This trough should be constantly supplied by a stream of cold water, to cool the heated tools. A large pair of tongs is also required to grasp the hot balls in the furnace. These balls are either dragged on iron slopes to the hammer or squeezer, or, as is more commonly the case, they are loaded on iron wheelbarrows made expressly for the purpose, and wheeled by the helper to their appropriate desti- nation. A flat bar, with a round handle, for stirring the fire, and cleaning the grate ; a coal shovel; a small hammer ; and an oblong, sheet iron dish for throwing water or hammer-slag in the furnace, complete the list of implements requisite at a puddling furnace. /. When the metal is heated to such a degree that a blow from a hook will break it, the damper should be lowered. If the iron is not of the best quality, the damper should be very nearly closed, so as to prevent the access of oxygen until the metal is thoroughly mixed with the cinder. By this means, the iron is protected, time is given to the workman to break it, and an opportunity afford- ed for a combination of the impurities with the cinder. Where the metal is of good quality, so much attention is not required at this stage of the process. When the iron is well worked into the cinder, the damper may be slightly raised ; and if but little flamo 268 MANUFACTURE OF IRON. is in the furnace, a small quantity of coal may be thrown into the grate, and the fire stirred. At this point, the duties of the assistant workman cease. The puddler, then, with a good sharp bar, frees the bottom and sides of the furnace of any lumps of metal, or lumps of iron already refined ; and in case the bottom is not per- fectly smooth, he takes away the projecting parts, which are gene- rally metal, adhering to the cinder. Gradually the mixture of iron and cinder rises spontaneously, and exhibits a kind of fermentation. This may be kept down by raising the damper; or, by stirring the fire, it may be permitted to rise still higher. If all the iron is melted, and the furnace in good order, the rising must be prevented ; but if the furnace is not quite clean, it is preferable to maintain a low temperature until all the iron is mixed in small particles with the cinder. When this is fairly accomplished, the damper may be slightly raised, so that, in addition to the heat, a small quantity of oxygen may pass through the iron. Should the metal have been of good quality, but little time is required to separate the iron and cinder ; this stage of the operation is called coming to nature, and is characterized by the iron forming at first small, and to all appear- ances round particles of the size of peas, which swim in the cinder. When these particles of refined iron begin to grow larger, by ad- hering one to another, the damper may be raised, and the heat in the furnace brought, by degrees, to the highest point. The accumulation of the particles then proceeds rapidly. Active mani- pulation is required to prevent the formation of too large masses. By breaking up, and turning, the whole mass is uniformly heated. After a short time, by squeezing the small lumps, by means of the bar or hook, round balls, twelve or fifteen inches in diameter, or seventy or eighty pounds in weight, are formed. After all the balls are finished, the work hole is shut for a few minutes, that a final and thorough heat may be given to the iron. When this is accomplished, the wedges at the door are loosened, the door is lifted by the helper, and the puddler takes one ball after another to the hammer or squeezer, or loads it on an iron hand-cart, which the helper wheels to its place of destination. g. If the metal charged is gray or mottled, a somewhat different method of working it is pursued. So far as the heating of the iron is concerned, but little difference in the treatment is required, though the heat, before commencing operations, must be stronger than in the puddling process. It requires some skill to hit the proper time for commencing operations. If we commence too soon, the iron will MANUFACTURE OF WROUGHT IRON. 269 divide into small particles, and assume a somewhat sandy appear- ance; in this case, the work will not only proceed slowly, but the iron will be of inferior quality. If, on the other hand, the metal is melted perfectly, the result will be rapid work, and an excellent quality of iron. Melting of the metal may be accomplished by leaving the damper open until the iron and cinder have become sufficiently liquid, after which it must be shut to exclude atmo- spheric air. At this time the interior of the furnace appears dark and smoky, and black fumes issue from the almost closed top of the stack. The melted mass is continually stirred, and at intervals of a few minutes, fluxes, consisting of hammer-slag, or pounded ore and water, are applied. If these act their part well, the surface of the mass will be covered, to a greater or less extent, with blue flames. Within twenty minutes, the cinder commences to rise; a kind of fermentation takes place beneath its surface; and the mass, at first but two inches high, rises to a height of ten or twelve inches. Whilst the cinder and iron are thus rising, constant stirring is required, to prevent the settling of the iron on the bottom, which is now deprived of the direct influence of heat. If the process goes on well, no iron is yet visible. When the cinder rises to its proper height, the duties of the helper cease. The puddler then com- mences, by means of a sharp bar, to free the bottom and sides of the furnace of lumps of metal. At this point the damper may be slightly raised; and, by the addition of a small quantity of coal, a bright flame may be produced. Soon after this, the iron is seen in small, bright spots at the surface of the cinder, and then alternately appears and disappears. Brisk stirring at the bottom and at the sides is now requisite to prevent the iron from remaining at the cold bottom, after having once been at the surface. The iron and cinder, when in lively motion, have a striking resemblance to the boiling of corn; from this resemblance the term lolling is derived. At a well-managed furnace, the boiling lasts about a quarter of an hour; the cinder gradually sinks; and the iron appears in the form of porous, spongy masses, of irregular size, which are to be stirred, to prevent their adhering together in lumps too large to be formed into balls. At this stage of the process, the heat should be raised as high as practicable. The iron, even in its spongy form, will be quite hard, and a good heat is required to soften it sufficiently for welding. If the heat is not strong, the iron is not apt to stick; and if put together by squeezing, it will not bear shingling ; besides, the balls are likely to break under the hammer, or in the squeezer. 270 MANUFACTURE OF IRON. The method of removing the balls is the same as that before de- scribed. Puddling and boiling differ mainly in the method of bringing the iron to nature ; that is, producing that transformation of metal which constitutes bar iron. The difference between white and gray iron does not produce the difference in the work, but the degree of fusi- bility of the iron, and the time required to crystallize it. The de- scription we have given of boiling and puddling applies only to cases in which good wrought iron is produced. Instances occur in which both processes are applied in the same case; and we think we shall but slightly err if we state that the puddling opera- tion is generally conducted, to a greater or less degree, to a state of boiling. h. The construction of boiling and puddling furnaces does not vary materially except in the depth of the hearth ; that is, in the distance from the work-plate below the door to the bottom plate. In the latter, a depth of six inches is sufficient ; while in the former, a depth of eleven or twelve inches is required. In the puddling furnace, the distance between the bottom and top seldom exceeds twenty inches ; in the boiling furnace, it varies from twenty to thirty inches. In the former, the iron boshes do not always reach all round the hearth, but are frequently confined to both bridges ; in addition to which, the sloping sides are of fire brick. i. In puddling, the furnace is charged with metal alone; but in boiling, cinder is charged along with the metal. When the balls are removed from the boiling furnace, a large mass of fused cinder remains in the bottom, a part of which is let off, through the tap- hole below the work door, into a two-wheeled iron hand-cart. A small portion of the liquid cinder is left in the furnace. A large quantity of cold cinder, from the hammer or squeezer, is now thrown upon the pasty cinder; and upon this cinder the pig metal is placed. The cinder which results from boiling is of inferior quality, but it is improved when mixed with that from the pud- dling furnace. For this reason, puddling furnaces are used at the western puddling establishments. Charcoal forge cinder, added to the above hammer cinder, is still better than that from the puddling furnace. At the Pittsburgh works, it is customary for the puddlers to make six, and the boilers to make five heats in a turn, of a charge weigh- ing 350 pounds. This is accomplished in eight or nine hours. The MANUFACTURE OF WROUGHT IRON. 271 workmen make but two turns in twenty-four hours ; therefore an interval of from six to seven hours, during the night, is left, in which the furnaces are stopped up. The workmen change every day at twelve o'clock ; the first set begin at three or four o'clock in the morning, and the second cease at about ten at night. Jc. The construction of the western puddling furnaces does not differ materially from that of the single furnace generally in use in England ; but they are distinguished by iron boshes, by which the hearth is lined all round, which is not the case anywhere else in single furnaces. In the Eastern States, there are scarcely any single puddling furnaces in use. Where anthracite is employed, the construction of the fire-places is modified. The following illustrations will serve the purpose of description : Fig. 89 represents an anthracite furnace dissected vertically through the grate, hearth, and chimney. The arrangement varies but slightly from that of the single furnace we have already described, with the exception that the grate is deep- er. In this furnace, coal can be filled to the depth of from twenty to twenty-four inches ; while, in the bituminous coal furnace, a depth Fig. 89. ce, Puddling furnace for anthracite coal. of ten or twelve inches is sufficient. The cross binders, which we omitted to mention in our description of the single furnace, are marked a a. These binders are a necessary element in the con- struction of a furnace. They are wrought iron square bars, either with screw and nut, or with a key, and serve to bind together the cast iron plates of the enclosure. They prevent the sinking of the roof caused by the expansion and contraction of the fire brick. 272 MANUFACTURE OF IRON. The two holes below the grate serve for the passage of the blast. For this purpose, one orifice is usually deemed sufficient. The blast machines are fans ; and, as pressure of the blast is unneces- sary, they serve every purpose. We shall speak of these in another chapter. The incombustibility of anthracite coal makes the application of blast necessary. A chimney cannot draw through a high column of coal an amount of air sufficient to give it the requisite heat. If the column of coal in the grate is left low, all of the oxygen of the air is not absorbed, and the quality of the heat is impaired. An- thracite can be most successfully burnt, when blast is applied to it. Fig. 90 exhibits a horizontal section of the furnace. The hearth Fig. 90. Anthracite double puddling furnace, horizontal section. and grate are seen from above. In a double furnace, the grate commonly measures three by five, and in a single furnace, three by four feet. The width of the furnace externally is from five and a half to six feet. Some furnaces measure even seven feet ; but this is rare. The hearth is generally six feet in length, and its width accords with that of the furnace. The flue ought to mea- sure at least 150 square inches ; and more than that, if the chim- ney is narrow. However, a flue twenty-four inches in width, and seven inches in height, may be considered of good size. The chim- ney is sometimes of larger dimensions than necessary. A lining sixteen inches square is sufficiently wide for a double or single furnace. A chimney high enough to carry the hot gases out of the furnace is, under all circumstances, sufficient. The draught, and consequently the heat, depend upon the blast, for which reason it matters very little what kind of chimney is employed. The main difference between this and the single furnace is, that in the former there are two work doors, one directly opposite the MANUFACTURE OF WROUGHT IRON. 273 other. Therefore, two sets of workmen are required at the same time. In this furnace, double the quantity of metal is charged, and of course the yield is twice that of a single' apparatus. The advantages of this arrangement are obvious. Rooms, building ex- penses, and fuel are economized, and much of the labor of the workmen saved. Besides, but one good puddler is required for managing the operation ; while at a single furnace two are needed. Of course, no more repairs are required for one furnace than for the other. The arrangement of the hearth in a double furnace varies con- siderably. In Pennsylvania and the anthracite region, the boshes are made of soapstone, a refractory material found in eastern Penn- sylvania and New Jersey. In some places, they are made of a refractory ore, magnetic oxide, mixed with soapstone. In the State of New York, and the New England States, the furnaces are provided with hollow iron boshes ; and where anthracite is em- ployed, the blast is led through these boshes, and the air, thus heated, applied to the coal. In many cases, where the boshes are of iron, iron ore is used, partly to protect the boshes, and partly to flux the iron. On the Hudson River, the crystalized magnetic ore from Lake Champlain, an excellent article, is employed for this purpose. The following illustration (Fig. 91) of the cast iron Iioi- Fig. 91. Double furnace with air boshes and heating stove. low boshes will be understood without any description. Their height is usually from twelve to fifteen inches ; their width at the bottom six, and at the top from three to four inches ; the inside slopes toward the centre. These plates are generally so arranged 18 274 MANUFACTURE OF IRON. that the whole is cast in two parts, and divided at the doors. Each part forms a bridge, and its two wings serve to form the sides. There is no difference between the manipulations at this, and those at the single furnace. It can be used either for puddling or for boiling ; or, at least, a process analogous to boiling. That is to say, the fermentation is carried to half the extent of that usual in regular boiling. At one time, this furnace labored under a serious disadvantage. The quantity of iron it contained at once, sometimes amounted to 900 pounds. Therefore, the time necessary for shin- gling at the hammer, or the old-fashioned squeezer, was not only injurious to the iron, but occasioned a loss of time to the workmen. This difficulty is at present effectually removed by Burden's rotary squeezer. I. At the Eastern establishments, the heating stove is commonly applied to the puddling furnace. It forms an appendage or pro- longation of the hearth. Its location is generally between the pil- lars of the stack. It is charged from behind, and on this account is very convenient. Fig. 91 shows the arrangement of this stove. With experienced workmen, it affords facilities for economizing fuel and time ; but with awkward workmen, it is of doubtful utility. Before we give the general practical rules which should guide us in our manipulations, we shall present two very interesting illus- trations of puddling. m. Fig. 92 represents a section of a single furnace in operation Fig. 92. Single puddling furnace at Hyanges. at Hyanges, France. The general appearance of this resembles that of any other puddling furnace, with the exception of the manner in which the heating stove is applied. In this instance, it forms a prolongation of the hearth, while the flue is behind it, leading to the stack. Bituminous coal is used, and the grate is constructed in MANUFACTURE OF WROUGHT IRON. 275 accordance with this circumstance. Thus far there is nothing un- usual in this furnace. Its characteristic feature is, that its bottom is of cast iron, which is from four to five inches thick. The fire bridge is about six inches high ; the flue bridge, formed by the stove, is of the same height. At the centre, the bottom is four inches deeper than at the sides, and is about four and a half feet in width by five feet in length. It is secured from below by iron props, and therefore, when burnt or cracked, may be replaced by a new one. In this furnace, the worst kind of coke iron is converted into fibrous bar iron of very fine appearance ; but for the blacksmith's use, this article is of poor quality. To those manufacturers who desire to produce cheap iron, with no special regard to quality, this furnace is worthy of imitation. The pig iron of Hyanges is smelt- ed from a brown, fossiliferous ore resembling the fossiliferous ore of Eastern Pennsylvania. It is run into large chills, directly from the blast furnace, and cooled off as at a running-out fire. After being properly heated, the furnace is charged with a small wheelbarrowful of hammer cinder, mixed with pounded feldspar. The metal in the stove, previously charged and red hot, is drawn by the puddler upon the cinder. The furnace is then closed, and a good fire prepared. Within a quarter of an hour, the metal will be sufficiently heated for working ; that is, it will be red hot, though not melted. The puddler commences to break up the iron, and mix it with the cinder ; the mass is gradually fused, and the cinder and iron exhibit a tendency to rise. At this stage of the process, the tap hole is opened, and the main body of cinder let out. Only a suffi- cient amount is retained to work the iron. In the mean time, a good fire is prepared ; and the puddler draws the damper as soon as the cinder has flowed out. The refuse cinder is then covered with ashes, and the operations vigorously prosecuted. If well conducted and this consists only in quick work, for the iron comes to nature when the surplus cinder is gone the whole process will be com- pleted in an hour. When the balls are finished, and the door closed up for a final heat, the metal is charged into the stove, after which it is drawn and shingled. The process is then again commenced, and continued as before. At this furnace, but one workman is required at a time. A heat is commenced and finished by one man, without any help; the next heat is worked by another puddler. Some workmen employ a boy for stirring the fire ; but this is not always the case, for the boy must be paid from their own earnings. At the time we 276 MANUFACTURE OF IRON. visited the works at Hyanges (1837), 250 kilogrammes (equal to 550 pounds) formed a charge ; and nine or ten heats were made in twelve hours, the workmen changing, however, at every six heats. With four workmen, a single furnace furnished from twenty to twenty-five tons of iron per week ; a great deal of which time was consumed in shingling the balls. By the use of Burden's squeezer, thirty tons per week could be produced. The manipula- tion at the Hyanges furnace differs from that at common furnaces in the fact that the puddling is done on a red hot iron bottom, as well as in the fact that a feldspar flux is added to the cinder. In another place, we shall investigate the reasons why this process differs so materially from the common puddling operations. n. During a period of three or four years (from 1834 to 1838), we were placed in a position which required the highest degree of perseverance. We engaged in the most difficult enterprises, with the object of improving the puddling operations ; sometimes with success, and at other times failing to accomplish what we had pro- posed to ourselves as the result of our labors. The results of the experience thus acquired, it is our purpose to relate, with the hope that they may prove useful to those engaged in this difficult depart- ment of labor. In the years intervening between 1832 and 1836, great exertions were made by iron manufacturers to improve the quality, and to increase the quantity, of iron, by means of artificial fluxes. It was already a matter of conviction amongst educated metallurgists, that the quality of the metal in the furnace depended upon the accom- panying cinder. The conclusion very naturally followed, that, if we could prepare a cinder of given quality, the desired metal might be obtained with comparative ease. However true the funda- mental premise may be, the sequel proved either that the conclusion was only measurably true, or that a cinder answering, in every respect, our wishes, remained yet a desideratum. In the investiga- tion of this subject, numerous experiments were made, in which we participated. In applying the artificial composition of cinder to the puddling furnace, subsilicates of such remarkable fusibility resulted, that the best fire brick was, after a few heats, entirely destroyed. But a settled conviction was arrived at, that the injurious admixtures of a metal no longer formed an obstacle in furnace operations ; for phosphorus, sulphur, and silex were so completely removed from the iron, that no difference appeared to exist between the best and the worst metal. On the contrary, there was reason to believe that MANUFACTURE OF WROUGHT IRON. 277 the advantage was on the side of the inferior metals. How far the latter conclusion is true, we shall hereafter see. In consequence of the destruction of the hearth, we lined the furnaces with cast iron, wrought iron, and other refractory materials ; but all to no purpose. The uniform result was, that the cinder was either too fusible, or that the iron manufactured was so hard and tough as to require a heat which no lining could withstand. After innumerable experi- ments, we succeeded in constructing a double furnace with water boshes. At first, this answered every purpose ; but how it suc- ceeded where we had to deal with different metals, we shall relate in another place. Nevertheless, from the construction of that fur- nace, the principle was established which, with proper modifica- tions, is applicable in all cases. As this principle was the basis of all subsequent modifications, and as it was extensively adopted throughout the Continent of Europe, we shall present an engraving of the furnace, and notice in another place the alterations which it has since received. Fig. 93 represents a vertical section of the double furnace. The boshes are heavy cast iron plates, ten inches high, five inches thick,, and with a small passage of about an inch or an inch and a half Fig. 93. Double furnace with water boshes. Fig. 94. Ground-plan of a puddling furnace with water boshes. 278 MANUFACTURE OF IRON. bore. They extend all around the hearth ; being coupled at one door. At the other, the water has entrance and exit. But very little water is required to keep these boshes cool. The bottom of the furnace is formed of small cast iron plates, about twelve inches in width ; their length corresponding with the width of the furnace. The grate measures three feet in width by two feet in length ; length of hearth six feet, and width between the doors five feet. Stack forty feet in height, and diameter of lining eighteen inches. "Width of flue twenty-four inches ; height six. Distance between the iron bottom of the furnace and the brick roof twenty-eight inches. The lower parts of the furnace are open, so as to permit a free circulation of air to cool the bottom. This furnace works exceedingly well in all cases in which infe- rior cold or hot-short iron, smelted by heavy burden, is puddled. From any fusible metal, that is, from any metal smelted by heavy burden, or by low temperature, very superior iron may be puddled by the application of artificial fluxes. Iron equal to the best char- coal iron may be manufactured from cold-short, or from any very fusible metal. But for gray metal of small burden, particularly for all coke, stone coal, or hot blast iron, these furnaces are of questionable utility. Eor white metal they are perfectly useless. We failed invariably in our attempts to work white metal of small burden, whether it belonged to the best quality, or whether it was smelted by coke or anthracite, or hot blast. o. The manipulation in this furnace does not differ from that pre- viously described ; but, as the application of artificial fluxes is not practically so well understood, we shall briefly describe it. This furnace is not adapted for puddling, or for the working of white metal, but for boiling alone. It is heated in the same manner as any other furnace. Cinder is filled and melted as described when speaking of the Pittsburgh furnace. At the close of every heat, a portion of cinder is let off, in case too much exists at the bottom. But this is not likely to be the case, if due care is observed. When the cinder at the bottom is cooled off, the metal is charged in the middle of the furnace. This may be taken from the heating stove, in case one is connected with the furnace. Should there be a good fire, it will be ready for work in half an hour, when it may be broken up, and mixed with the cinder. When the pig iron is bad, that is, cold-short or hot-short ; or where it contains sulphur, phosphorus, and silex, besides carbon, the fire should be well stirred, without charging fresh coal, and the temperature raised sufficiently high to MANUFACTURE OF WROUGHT IRON. 279 rnelt the iron perfectly; otherwise we cannot produce a good article. Whether the iron is melted, and not merely mixed with the cinder, may be known by the formation of bright streaks in it. When the mass is thoroughly liquid, the damper may be almost completely shut ; still, the interior of the furnace should be bright, though the flame is not visible. The artificial flux is now thrown into the fur- nace, at intervals of one minute. Assuming this flux to be divided into ten or twelve portions, all of it may be applied in fifteen minutes. During this application, iron of good quality rises ; but that which is bad, or very liquid, rises only by means of hammer- slag or water. Before the cinder rises, blue flames, in many cases, literally cover the surface ; but cease when the iron comes to nature, that is, shows itself at the surface in little specks. One hour is sufficient for this part of the process, that is, from the charging till the appearance of the refined iron. Some metals work slowly ; but this difficulty may be remedied by the construction of the furnace. When the iron is refined, that is, when it boils strongly, and begins to rise, the damper may be raised, and fresh coal applied. The boiling will thus be brought to a stop. By gradually increasing the heat, working fast, and turning the finished iron, which is now in spongy, open lumps, the cinder rapidly sinks, and the iron is left bare, ready for balling. If the depth of the boiling cinder, at its highest point, is five or six inches, one turn- ing will be sufficient; but if ten or twelve inches, the iron gene- rally becomes so cold in the bottom, that a turning back and for- ward several times is required. If this kind of boiled iron is balled up cold, it will break under the hammer or squeezer, of whatever quality it may be. The responsibility of .this department rests upon the puddler. This process differs from other processes principally in the melting-in of the metal. The more inferior the metal, the more carefully should this be performed. If the metal is of poor quality, a charge should never exceed 700 pounds; but if otherwise, it may be increased to 800, and even 900 pounds. Bad pig iron, though inclined to work slowly, may be worked quite as fast as that which is good, if the charges are small. The quality of puddled iron may be made equally good under all circumstances. Pig iron may contain phosphorus, sul- phur, or any injurious admixture except copper. Puddled iron may be completely freed of them. Bar iron manufactured from the most cold-short gray pig iron which contains phosphorus, may 280 MANUFACTURE OF IRON. be made superior, in every respect, to that manufactured from the best metal. VI. G-eneral Remarks on Charcoal Forges. It is undeniable that charcoal forge iron is, in many respects, superior to puddled iron. For all the purposes for which wrought iron is applied, it is more malleable, compact, and durable. The puddling process is conducted on more philosophical principles than the charcoal forge, and in the course of time may be brought to such a state of perfection as to supersede the latter altogether. But this is not the case at present; and the charcoal forge will be needed so long as the puddling process does not furnish a quality of iron equal to it. Another reason why the former will command prece- dence for some time to come, is that it is less expensive than the more complicated puddling establishments, and permits the manu- facture of iron on a small scale without serious disadvantages. Iron works, situated^ at remote places in the country, frequently find a favorable market for a limited quantity of iron ; while an increase of that quantity would not prove profitable. Such cases are very common in the farming districts of the interior of the country which are not easily accessible, as well as in the growing Western States. The same remark is applicable to the new States and territories. It is questionable whether the charcoal forges of the West, and even in the heart of the anthracite and bituminous basin, do not yield larger profits than the puddling forges and rolling mills ; at least, an investment in charcoal establishments may be considered quite as safe as in those of stone coal, at the present time. a. The location of charcoal forges should depend upon the sup- ply of ore and wood. Inferior ore, and the metal smelted from it, are less useful to the charcoal fire than to the puddling furnace. The success of the former depends upon the quality of the metal with which it is supplied. It is thus evident that the best is always the cheapest metal. This rule is not applicable to puddling establish- ments. In addition to this, the charcoal forge requires good coal. But rich ore or excellent metal may counterbalance expensive coal ; while poor metal and expensive coal will yield only unprofitable results. Where the metal is good, a ton of iron requires only 150, and sometimes only 120 bushels of charcoal ; and seven tons can be produced in a week, with but one fire. But where it is poor, a ton will require from 200 to 300 bushels ; while only two or three tons of iron per week will be produced. We refer to blooms, not MANUFACTURE OF WROUGHT IRON. 281 to drawn iron. Consequently, should the iron resulting from the smelting of good or bad metal be equally valuable, which is not a fact, the expenses of manipulation are so decidedly in favor of the former, that the question which to choose will never arise. b. The magnetic ores of the States of New York, Vermont, New Jersey, Missouri, afford an excellent article for the charcoal forge. These ores exist in such immense quantity, and in the north-west part of New York are of such superior quality, that the little inte- rest they excite in the public mind is to us a matter of extreme astonishment. In the magnificent region just mentioned, metal might be made at the ore banks, and sent to the Hudson or the Delaware River to be puddled. The spathic carbonate, the specu- lar ore, and the red clay ores of the transition series, also consti- tute an excellent article for the charcoal forge; but these are not so generally distributed as the magnetic ore ; at least, they are not concentrated in such large masses at any given place. The rich hydrates of Tennessee and Alabama are adapted for the Catalan forge. The same reasons which may be assigned against the working of poor ores in this forge, apply against their use in the charcoal blast furnace. Inferior metal is, at present, employed in the coal regions for the manufacture of charcoal blooms ; but we predict that these efforts will, in a short time, be abandoned, be- cause poor charcoal iron cannot successfully compete against pud- dled iron. Metals which contain phosphorus or sulphur are not adapted for the charcoal forge, because of the inferior iron they produce, and because of the amount of time consumed in convert- ing them into bar iron. Gray metal from rich ores, and mottled or white metal of pure origin, form medium qualities. All metals derived from impure bog ores, sulphurets, silicious ore, and ores containing phosphorus ; all the gray metals smelted from poor ores, particularly those of silicious origin ; and all white metal resulting from small burden, inferior ore, and bad management in the blast furnace, are improper for the charcoal forge. c. The necessity of good metal in the forge is illustrated by the following fact: An instance is recorded in which a ton of blooms, from white metal of excellent quality, was produced, with the con- sumption of only ninety bushels of coal ; while, on the other hand, when gray pig iron was used, 400 bushels of coal were consumed in producing the same amount. d. The site of a forge is generally selected in relation to facili- ties for obtaining water power ; but it is probable that steam may 282 MANUFACTURE OF IRON. prove to be the preferable power, because the waste heat of the forge fire is sufficient to generate it. It is also probable that the first outlay in erecting the works is, at least in a majority of in- stances, in favor of the steam-engine. e. The application of hot blast to the charcoal forge is of ques- tionable advantage. It will save fifteen or twenty per cent, of coal ; but labor is increased, and the iron depreciated. We shall elsewhere make some additional remarks on this subject. VII. General Remarks on Puddling. This method of converting cast iron into malleable iron is de- signed to supersede every other method by which that result is effected. But, thus far, the quality of puddled iron has been such that we have been unable entirely to dispense with the char- coal forge. Still, this quality would be much improved if better metal was generally employed. The nature of the puddling pro- cess is such, as we have elsewhere stated, that we are enabled by it to employ inferior metals to a great degree. Thus, blast furnaces have been erected at places where charcoal forges would not have flourished. Inferior pig iron answers tolerably well for the pud- dling furnace. Metal perfectly useless in the charcoal fire will, in this furnace, produce a very good article. In fact, every kind of pig iron, however bad in quality, may, by the puddling process, be advantageously worked. a. In the Western States, where charcoal pig alone is puddled and boiled in single furnaces, iron of very good quality is made. A great deal of inferior iron is also produced, which, according to the metal used, should be of better quality. The puddling fur- naces of the West work well ; but it is doubtful whether a due amount of labor is spent in working the iron. The puddlers gene- rally finish a heat in less than an hour and a half, including shin- gling; and the boilers in less than two hours. At other places, this is considered an insufficient time to do full justice to the work. At well-regulated Eastern establishments, twelve hours are consumed for five boiling heats, and the same time for six puddling heats. This may be considered fair time for industrious and judicious manipulation. Where the metal is of superior quality, less atten- tion is required. But throughout the United States the tendency of most blast furnaces is to produce gray metal ; consequently, the manufacture of good bar iron requires great industry, however good may be the ore from which it is smelted. MANUFACTURE OF WROUGHT IRON. 283 5. As previously remarked, at Pittsburgh and the Western Works, boiling is carried on in about one-half of the puddling furnaces. Those used exclusively for puddling are regarded as necessary evils, and are employed merely to make cinder for the boiling furnaces. Excellent cinder is produced from metal of good quality, carefully puddled ; but, on account of the refining of the crude iron before it is taken to the furnace, this operation is expensive. All the advan- tages which the process includes are realized at the Western estab- lishments. But, unless other methods are adopted by the Western manufacturers in working pig metal, competition will gradually exhaust all the profits of this business. This, let us observe, is a more important matter than it seems to be, for, if puddling is re- placed altogether by boiling, the question meets us, whence is the necessary supply of cinder to be obtained ? Charcoal forge cinder, at present frequently applied, cannot be obtained in sufficient quan- tity. Artificial fluxes, then, are the only resource of the Western manufacturers. Good iron ore will serve as an excellent flux ; but this cannot be found either in the Western or in the Eastern coal regions. In the State of New York, the magnetic ore from Lake Champlain is employed ; and the furnaces of this State not only produce excellent iron, but furnish a more abundant yield than any we have ever seen. At Saugarties, on the Hudson River, 2000 pounds of rough bars have been made from an amount of pig iron varying from 2075 to 2100 pounds. Loss only from three to five per cent. Amount of anthracite coal consumed from 1600 to 1TOO pounds. Furnaces double, with iron air boshes ; charge 750 pounds, and five heats in twelve hours. The magnetic ores of Missouri, and the red oxides of Arkansas afford a good material for the Western mills ; but ores of the coal formation are not sufficiently pure. The amount of good ore required per ton of inferior pig iron is sometimes from 400 to 500 pounds ; but for excellent metal, rarely beyond 200 pounds. c. We have stated that most puddling furnaces are provided with iron boshes. But in those which work anthracite iron, soapstone is employed for keeping the boshes in order. It is evident that, if iron boshes were proved in all cases to be advantageous, they would be adopted. But, in the present case, they are of doubtful utility, as we shall explain. The necessity of enlarging the hearth, so that a smaller surface of the boshes, in proportion to a given amount of metal, would become cool, originated the double furnace. It was found that tho 284 MANUFACTURE OF IRON. cooling influence of the iron lining, in small or single furnaces, was so great that inferior pig iron could not receive that improvement which otherwise might be effected with comparative ease. The extension of the area of the hearth, to a great extent, removed this difficulty. There is no doubt that the quality of iron might be improved to an inconceivable degree, if a hearth could be con- structed of materials adapted to resist the action of strong alkalies; but the necessity of cooling the boshes is so strong a counteracting element, that the beautiful theory of improving iron by means of artificial cinder is but of limited application. In this respect, double furnaces present greater advantages than single furnaces ; and boshes cooled by air are superior to those cooled by water. In the improvement of bad pig iron, by puddling, our primary object should be to melt it perfectly, and then to remove its impuri- ties by means of cinder. If, therefore, a hearth is so cold as to prevent the melting of the metal, the most essential condition of improvement is not realized. If the iron contains impurities firmly and intimately combined as that from coke, anthracite, or even from charcoal furnaces, smelted with small burden a perfect re- melting is necessary. Such iron requires a strong heat ; and this heat cannot be produced in a furnace with cooled boshes. Hence the failure of experiments made to improve such iron. Anthracite iron contains a large amount of silex, in addition to carbon ; and a furnace with water boshes is unable to produce a heat sufficient to melt it. Fibrous bar iron is preferable, as an article of commerce, to that which is cold-short ; and to prevent it from becoming cold- short, the intimate connection between the impurities and the iron must be destroyed. Therefore, a furnace with soapstone, or, what is still better, good fire brick, will produce a better iron for the market than a furnace with cold boshes. A uniform temperature of the lining and walls is required to produce a thorough solution of the pig iron. The presence of silex in large amount, as in a lining of soapstone or fire brick, affords, by retarding the work, every facility for producing this result. This latter circumstance should be viewed rather as the least of several evils than as a posi- tive advantage. From these considerations, it follows that pig iron from small bur- den, or made by a high temperature in the blast furnace, cannot be improved in a furnace with water boshes ; and that the application of these boshes should be limited to such iron as will thoroughly melt at a medium heat. Consequently, white metal containing a MANUFACTURE OF WROUGHT IRON. 285 large amount of carbon, anthracite, coke, and charcoal iron from small burden and hot blast, as well as all refined metal, are ex- cluded. Pig iron from heavy burden, and from ores containing phosphorus ; gray charcoal pig ; and, in fact, all metal which readily melts, and keeps liquid for a considerable time, are, of all others, the most serviceable. d. Our own experience, which is somewhat extensive in this branch of the business, proves that white metal from the richest ores is unfit to be worked at all in a furnace with a cooled hearth ; and produces far better iron in brick linings. Pig iron from small burden and coke, we never succeeded in improving. With white iron from charcoal furnaces and small burden we were equally un- successful. The most favorable pig iron is that which is made by a small quantity of coal and by low temperature in the blast fur- nace. The lower the temperature, the better the iron. Pig iron smelted from phosphates, is easily converted into the best kind of bar iron, if the temperature of the furnace has been low, or the bur- den heavy ; but if smelted from the same ore, and by a high heat, whether charcoal, anthracite, or coke, it is improved with difficulty ; sometimes total failure results. The same rule is applicable to pig iron smelted from silicious and sulphurous ore. In fact, it may be laid down, as a general rule, that the smaller the amount of coal consumed, or the lower the temperature of the hearth in the blast furnace, the better will be the quality of the metal ; that is, the more fit it will become for improvement in the puddling furnace. We thus see the advantage of heavy burden in the blast furnace, for it not only reduces the first cost of the metal, but makes a far superior article for subsequent operations. We may safely say, that the worst cold-short or sulphurous metal, smelted by a low heat, is quite as good as the best metal from the best ore smelted by a high temperature. We will give a practical illustration. About ten years ago we were engaged in improving cold-short iron; that is, pig iron smelted from bog ore, which, before that time, possessed no value whatever. Our manipulations were conducted in a double furnace, with water boshes. The puddling was carried on by means of artificial fluxes. We succeeded, without difficulty, in producing a beautiful bar iron, in quality equal to the best in the market. With the object of testing its virtues, a portion of it was sent to a distant mill, and converted into wire. So successful was the result, that the puddled iron was preferred to the best charcoal iron. At the wire mills, where an extensive business was done, a large 286 MANUFACTURE OF IRON. quantity of charcoal iron was needed. As this could not be ob- tained in consequence of its expensiveness, puddling works were erected for the purpose of furnishing iron for the inferior qualities of wire. At this establishment, steel metal of the most superior kind was wrought, which of course, puddled in single furnaces, with good fire brick lining, made an excellent bar iron. After using the iron of our cold-short metal, the owners of the rolling mill entered into an engagement with us by which we bound ourselves to fur- nish as good an article from their superior plate metal as we had made from worthless phosphorous pig. A few heats made in one of their own puddling furnaces indicated that improvement was possible ; but, owing to certain peculiarities of the new process, puddling could not be performed in a brick lining. We therefore concluded to erect a double furnace at once, and apply iron boshes. Until that time, our practice had been confined principally to the worst kind of pig iron, and accompanied with more or less success, according to the nature of the metal with which we had to deal. We entered upon the undertaking with great confidence. The idea of failure never entered our mind. This confidence appeared to be justified on account of the insignificance of the improvement re- quired, compared to what we had already arrived at. The metal was the best which the Continent of Europe afforded ; but, after all our exertions, the ultimate result was a total failure. As this is one of the most remarkable as well as interesting cases which ever happened, we shall relate it somewhat in detail, and thus serve a useful purpose. The metal used was smelted from sparry carbon- ates ; it was almost crude steel, that is, white metal containing car- bon in large amount. Being thoroughly acquainted with the most important part of the operation, we took great care to have a fur- nace of good heating capacity. The metal melted in a short time, and at a low temperature; but the least stirring with the tools made it crystalize, and worked it into nature; and sufficient time was not left to enable us to mix it properly with the cinder. The result was a dry, hard iron which broke under the hammer. No effort was left untried to overcome this apparently trifling difficulty; and when we at last succeeded, a very singular circumstance put a stop to the experiments. The breaking of the balls under the hammer is, in all cases, the result of too slow work. The workmen did their best ; but the iron worked too fast. This is generally the case with white iron containing a great deal of carbon. The applica- tion of fluxes retarded the process. At last the metal worked well. MANUFACTURE OF WROUGHT IRON. 287 and became soft and tenacious iron. But, when piled and re-heated, a number of the bars broke in the merchant rollers ; and the iron, commonly of a silvery white appearance, exhibited in its fibres a dark color. On a second re-heating, both in a blacksmith's fire and the re-heating furnace, it broke up into small fragments. In fact, it was iron no longer, but black magnetic oxide. Rolled down to half inch rods, it broke into fragments on the first heat. Bars one and two inches square exhibited on their surface a high degree of oxidation, and appeared, internally, of a fibrous, dull yellowish color. On the application of the slightest heat, this color changed to black. The above experiment is a highly interesting one. It shows clearly the legitimate scope of improvements, and the direction in which experiments should tend. The metal employed was, as we have stated, of the best quality. It furnished excellent steel with the greatest facility. In the charcoal forge, it furnished the strong- est kind of bar iron consuming per ton of iron only from 110 to 130 bushels of charcoal. In the single puddling furnace, with brick lining, it produced a firm, tenacious iron, but of too coarse fibre, and containing too much cinder for the manufacture of wire ; while, in the puddling furnace with iron boshes, it did not work at all, and ultimately returned to its primitive condition that is, became oxidized to ore. The establishment where our first operations were conducted was a very inferior one. The metal used, whether in castings, charcoal forge, or puddled iron, was almost worthless ; at least, it commanded a very low price in the market. The pig iron, smelted from phos- phoric ores, was cold-short in the highest degree, perfectly useless in the charcoal forge, and a poor article in the common puddling furnace. Yet this worthless metal was converted, with the utmost facility, into bar iron superior to any kind in a market where the first quality of charcoal iron was alone saleable. The amount of charcoal consumed in the blast furnace was only from 80 to 100 bushels, notwithstanding the ore yielded but twenty per cent. ; while from 180 to 200 bushels were required in producing a ton of the steel metal on which we experimented. The experiments we have described are extreme cases ; but they exhibit clearly the method by which we can arrive at the most favorable results. We always failed when we attempted to improve white iron from an overheated blast furnace, even though the ore and coal were of the best kind. We failed with the best white 288 MANUFACTURE OF IRON. metal of the Continent of Europe; with the steel metal of Siegen and Styria ; with white Scotch pig ; with the white coke iron of the fossiliferous ore. of France: and with the coke iron of the Mount Savage Iron Works, Maryland. But we always succeeded in im- proving both the quality and yield of pig iron from a tolerably well- conducted blast furnace operation. e. Experience thus shows what is required both for the charcoal forge and the puddling furnace. We will recapitulate the con- clusions arrived at. Gray pig iron of a fusible nature is ill adapt- ed for the former ; but is the best of all kinds for the latter. White metal containing carbon in small quantity, or smelted by heavy burden, is good in either case ; but white metal from poor ores and light burden is, in all cases, inapplicable. White metal from rich ore and light burden is superior to all in the charcoal forge, but in the boiling furnace it is almost useless. We may hence conclude that cold iron boshes are of great advantage where pig iron from a well-regulated blast furnace operation is wrought; but that, where white pig iron from small burden and a high tem- perature as in coke and anthracite furnaces, in which an excess of limestone is used is to be converted into bar iron, they are dis- advantageous. In the latter case, a fire brick or soapstone lining is preferable. /. Thus far we have considered simply the best means of making wrought iron. But if we wish to produce wrought iron for specific purposes, it is not a matter of indifference what kind of apparatus we employ. Merchant iron should be malleable, fibrous, and of good welding properties. This, as well as very cohesive wire iron, is manufactured in great perfection in the charcoal forge, and in the double puddling furnace with iron boshes. But railroad iron should not be made in either of these furnaces. Easily welded iron is made by allowing a small portion of carbon to remain in the metal, and by expelling, as far as possible, all foreign matter from it. But this, by destroying the fibre, will make iron of large dimensions cold-short. By re-heating and rolling small rods, the carbon will evaporate. If, therefore, we want fibrous railroad or any heavy bar iron, we must employ a metal free from carbon, iron from which the carbon is easily expelled, as that from the run-out fires. This is to be puddled in a very warm furnace. A cooled puddling hearth produces a good welding iron ; but the long exposure of large piles of this iron such as are necessary for railroad, heavy bar iron, and boiler plate to a welding heat, occasions great waste. In all such MANUFACTURE OF WROUGHT IRON. 289 cases, a brick lining is preferable to cold boshes. Wire iron should be of the best quality ; but the puddling process by which it is pro- duced would be inapplicable for railroad iron, for the latter would thus become cold-short. Iron designed for small rods, hoops, gas pipes, and wire, ought to exhibit a crystaline fracture, a steel-like grain, which is produced by carbon. But of all other foreign mat- ter it should be free. Silex and phosphorus will not evaporate, like carbon, on repeated exposure to heat ; and iron which contains either in a non-vitrified state, will be cold-short under all circum- stances, and will be useless for wire, or for any purpose for which strength is required. Wire iron, or merchant iron, should be manufactured from gray pig, which, unless improved by artificial cinder, must be of the best quality. By boiling with artificial cinder, any kind of gray pig may be converted into good iron in the puddling furnace with iron boshes. In a cooled hearth, all foreign admixtures can be expelled from the metal, and yet enough carbon retained to preserve its welding properties. This advantage is accompanied with a disad- vantage ; for the carbon, as we have before stated, makes the iron, when in large masses, cold-short, and occasions waste in the re- heating furnace. Piles of 700 or 800 pounds in weight, exposed to a strong heat in the re-heating furnace, will melt at the surface, without becoming, in the interior, sufficiently hot for welding. A specific kind of iron is required for nails, an important article in our iron works. Nails cut from charcoal iron are generally sup- posed to be of good quality; still, this iron, whether from the char- coal forge or the puddling furnace, furnishes an abundance of in- ferior nails. With respect to nails, the intrinsic value of the metal that is, its absolute strength, and welding properties, as in the case of wire iron has no influence whatever upon the value of the manufactured article. All that is desired in a good nail is, that it shall cut smoothly, and bend to a given degree. Iron containing an amount of foreign matter that would make it useless for any other purpose, answers excellently. Such iron may be manufac- tured from any kind of pig metal without difficulty, provided the re-heating and heating are carefully performed. Two different methods of manufacturing nail plates are now practiced. In the Eastern States, plates from five to twelve inches in width, but of no specific length, are drawn ; nails are obtained from cutting these lengthwise. In the Western States, it is customary to roll sheet iron from twenty to twenty-four inches in width, and six or seven 19 290 MANUFACTURE OF IRON. feet in length ; and nails are obtained from strips cut crosswise. Which is the preferable method, it is not easy to decide ; but the immense quantity of nails manufactured will justify us in giving the subject a close examination. To make a nail which cuts smoothly, and does not split, we require an iron of very close grain. For this purpose, cold-short answers better than fibrous, particularly coarse fibrous, iron. Iron is rendered cold-short by carbon, phosphorus, and silex ; the two latter cannot be removed by re-heating the iron. But re-heating will remove carbon ; and, therefore, when we take into considera- tion that iron which contains carbon can be welded with greater facility, and by a lower heat, than that which is free from it, it is evident that a small amount of carbon should exist in the iron be- fore it is placed in the re-heating furnace. From this it follows that, if we re-heat the iron, and reduce the size of the nail-plate, the iron in the rough bar ought to be cold-short : it will be fibrous after it is reduced. So far as the principle of working it is con- cerned, this iron is analogous to wire iron. The latter is best manufactured in the charcoal forge, or in the puddling furnace with iron boshes. Consequently, nail iron should be boiled in this fur- nace, provided it is repeatedly exposed to a welding heat, and drawn out into small sized plates. But if it is our design to make sheet iron, that is, by exposing the plates to the suffocating heat of a warming stove, the iron will not be freed from the carbon, and re- main cold-short. We th^s see that, in one case, small plates are advantageous, and in another case injurious. To make nail iron from white metal, it is necessary to work the latter either in the charcoal forge, or in a puddling furnace without cooled boshes. From this metal good iron can be made in the charcoal forge without the least difficulty ; but, in the puddling furnace with a soap-stone or fire brick hearth, we obtain an iron of coarse fibre, excellent for many purposes, but not adapted for the manufacture of nails. If, in such cases, we attempt to leave a portion of carbon in the iron, silex will remain along with it, should not the pig iron already have been free from it ; of course, such iron and nails will be cold-short, no matter by what method the iron is treated after leaving the pud- dling furnace. Good metal, puddled in iron boshes, will produce fibrous iron ; but the danger is that it will lose all its carbon, and that, by repeated heating, a fibrous, dirty, yellowish-colored, rotten iron, both cold-short and hot-short, will result. To prevent this, and retain the fibrous texture of the puddled bar, it is preferable to heat in stoves, and to roll sheet iron. This is the practice at the MANUFACTURE OF WROUGHT IRON. 291 Western iron works, and is the result of necessity, because iron is puddled principally from white metal ; and small nail plates, ac- cording to the Eastern fashion, would not work so well as sheet iron. Nail iron of satisfactory quality can be easily made, if we are well acquainted with the process of puddling. To work cheaply, we must resort to boiling. We may hence conclude that, in pud- dling for nail iron, we require gray or mottled pig iron, no matter of what quality, provided it is smelted by heavy burden. To secure the presence of carbon, while we remove impurities from the iron, it is absolutely necessary to boil in iron boshes ; and fine fibrous iron cannot be made, unless the pig metal is fusible, and remains fusible sufficiently long for the workman to wash it pro- perly in the cinder. All gray iron of heavy burden whether smelted by charcoal, anthracite, or coke, or whether the ores con- tain phosphorus, sulphur, or any other injurious element is adapted for this purpose. We thus see that the quality of bar iron differs according to the different purposes for which it is employed. The blacksmith needs an iron which can be easily welded, which is neither cold-short nor hot-short. Wire iron must be strong, and very cohesive ; it is of no consequence whether it can be easily welded, or whether it is cold- short or hot-short. Nail iron may be hot-short, but its fibre must be fine. Railroad iron may be anything but cold-short. The pro- perties of the first three are produced by boiling alone. The latter, if manufactured in a cold hearth, will be imperfect. g. The elements of pig iron are seldom of such a nature as to afford the exact quality of wrought iron we require, and it need scarcely be mentioned that, when smelted from different ores, it will contain admixtures according to the nature of the foreign mat- ter contained in each ore. A silicious ore will impart silicon to the iron; a phosphate, phosphorus; and sulphurets, sulphur; but as, under the peculiarities of the blast furnace, silicon and carbon have the greatest affinity for iron, they are most constantly associated with pig metal. One kind of metal exerts more or less influence upon another. The same principle which we have observed in re- lation to the blast furnace, is applicable to the puddling furnace ; that is, metals of different quality, mixed together, work better in the pud- dling furnace than metal smelted from the same kind of ore. A metal from calcareous ore works far better, when mixed with a silicious metal, than when mixed with iron derived from limestone ore; and if to the first two we add a metal smelted from clay ore, the result is still 292 MANUFACTURE OF IRON. better. This peculiarity depends less upon the tendency of the foreign matter to form a fusible cinder, than upon the fusibility im- parted indirectly to the metal by the foreign matter, and occasioned by their mutual affinity. Carbon occasions fusibility ; and silicious and clay ores are more inclined than lime to make a carburet of iron. An excess of lime not only excludes carbon, but it absorbs sulphur and phosphorus. Therefore, the least fusible iron is that smelted by an excess of lime. It frequently happens that a given iron is too fusible ; that it works slowly, and yields badly. If to this we add a metal of a somewhat refractory character, which also works badly by itself, we shall find that a very advantageous mixture results. In this way, we are enabled to work the most un- favorable metals advantageously. For these reasons, it is advisable to work metals from different localities. In our attempts to work, in a puddling furnace with iron boshes, coke or anthracite iron, or even some kinds of charcoal iron, we frequently meet with an un- expected disappointment; and this disappointment results from the imperfect fusibility of the pig iron in the hearth of the furnace. In such cases, artificial cinders are useless, for the best cinder cannot reach the impurities. These are enclosed in the particles of iron, and nothing but a perfect solution of the metal will remove them. This solution may be most easily effected by mixing with the re- fractory iron an iron that is very fusible. Admixtures in themselves injurious cease to be so if the metal can be perfectly dissolved and kept liquid until the cinder has had sufficient time to act upon it. In proof of this, it may be remarked that phosphorus or sulphur may be added as a flux to the half liquid iron ; and, if the cinder of the furnace is of the proper kind, the metal manufactured will be neither hot-short nor cold-short. The application of sulphur or phosphorus as a flux is difficult and expensive; we should, therefore, have recourse to fusible metals. Gray iron from phosphorous, sul- phurous, silicious, and clay ores, is of this kind ; as well as pig iron from the same ores, smelted by heavy burden. Cinder compo- sitions will not improve metal obtained from calcareous ores, or that smelted by an excess of limestone, or by a too light burden, for, though it should melt, and become apparently very liquid, it will crystalize so soon that no time will be afforded for improving it. In a previous chapter, we remarked that the best policy which the iron manufacturer can pursue, is to make cheap pig iron, and leave improvement in quality to the puddling furnace. This is perfectly true within certain limits. But, if we adopt the most economical MANUFACTURE OF WROUGHT IRON. 293 plan of working the blast furnace, that is, by carefully preparing the material, and by carrying as heavy a burden as possible, these limits are very extensive. When these conditions are observed, good iron may be produced with comparative ease. But pig iron, from fur- naces where the manipulations are carried on irregularly, and where a change of ore, coal, burden, and workmen often occurs, is with difficulty improved. In most cases, it is tetter to run this iron through the finery, and make of it coarse bar or railroad iron, than to attempt to improve it in the puddling furnace. Careful manipu- lation in the blast furnace is the. best security of success in the puddling furnace; in fact, success in the one is in exact proportion to the economy observed in relation to the other. The truism that good work is always eventually the cheapest is, in this case, amply confirmed. We have also attempted to explain what kind of iron can be made from a certain kind of pig metal, and to show what kind is neces- sary for specific purposes. So long as the process of puddling is imperfectly understood, the qualities of bar iron may be said to depend on metal, fuel, and labor; for, practically, it is evident that the product will depend on the quality of the materials we possess. At this, if at any stage, in the manufacture of iron, scientific im- provements are available. Industry and attention are sufficient, in most cases, to produce satisfactory results; but in puddling, some- thing else is required. Every experienced iron manufacturer is convinced that the quality and quantity of iron produced depend upon the nature of the cinder employed. That, in blast furnace operations, cinder can be improved only to a very limited degree, we have already shown. But, in the puddling furnace with iron boshes, this improvement may be indefinitely extended. It is of but little use to attempt to make scientific improvements at the charcoal forge. At this fire, the best and most easy method of making excellent iron is by employing white metal of good quality. The same remark is applicable to the ^puddling furnace with brick or soapstone lining. Cinder compositions are, in these cases, unavailable. These are of advantage only in the puddling furnace with iron bottom and boshes; and it may be said that there would be no limit to improvement in the quality of iron, if the iron lining would permit of a heat sufficiently strong to melt the refrac- tory metals. For this reason, we are confined to metals which melt at a given temperature; and for this reason, also, the irregular nature of the metal we employ produces such unsatisfactory results- 294 MANUFACTURE OF IRON. The following statements apply to furnaces with iron boshes; though deductions may be drawn from them relative to refining or puddling. h. In puddling, the most simple method of improving iron is, as we have previously mentioned, by mixing different kinds of metal on. the same principle we have applied at the blast furnace. The obvious deduction is, the more kinds we mix, the better the result, which coincides exactly with experience. For this reason, it is advan- tageous to mix pig iron from the coal regions with iron smelted from primitive or transition ores, and to mix calcareous metal with a silicious or phosphorous pig iron. Stone coal or coke iron is greatly improved by being mixed with charcoal iron. Baltimore pig iron, in itself an excellent iron, would, if mixed with iron from Hanging Rock, Ohio, be made a still better article. The latter may be con- sidered the best metal in the world for castings ; but, associated with the former, it would make a very superior wrought iron. In this matter, it is necessary to guard against the opinion entertained by some, that the mixture of iron from different localities merely is sufficient. This is by no means the case. Mixing is to be performed with due relation to the chemical composition of the ore, to the place at which the metal is smelted, and to the fuel applied in smelting. Magnetic and bog ores work well in the blast furnace, and their respective metals work well in the puddling forge. Calcareous ores and those containing manganese work best in the blast furnace, if smelted along with silicious or clay ores. The metals derived from each of these ores will make, when mixed, far better articles in the forge than each would produce, if wrought singly. Another method of improving iron is by mixing the cinders pro- duced by separate furnaces. This method is extensively practiced at the Western establishments. The kind employed is puddling cinder from furnaces which work refined metal, and cinder from char- coal forge fires. Such cinder is charged along with the pig iron in the boiling furnace. After the iron is melted, hammer-slag or roll scales are employed to excite fermentation, as well as for the pur- pose of accelerating the work, and improving the quality of the iron. The application of cinders, notwithstanding their unquestionable utility, is very limited. Inferior pig iron requires good cinder in large quantity; the use of cinder, therefore, is restricted to charcoal iron; and even here it can be applied only in a very limited degree. In this respect, the Eastern do not enjoy the advantages which the Western works possess, on account of the charcoal forges and char- MANUFACTURE OF WROUGHT IRON. 295 coal iron of the latter, and the extensive use which the former 'make of anthracite pig iron. If the cinder employed is of good quality, and in sufficient quantity, our labors cannot fail to be successful. But in the stone coal regions, good cinder is not abundant ; and that obtained even from the best forges is only of medium quality. Where hot blast iron is refined, it is so inferior as to cease to be of any use. The cinder we require should be obtained from pig iron from the richest ores ; and we may work to the best advantage by observing the same rule in relation to it which we gave relative to the mixing of pig iron and iron ores. A better method of improving iron than the application of cin- ders is by the addition of ore to the iron charges. This is exten- sively practiced at the Eastern works. The ore is either put in large pieces around the inside of the furnace to protect the boshes, or charged in small fragments with the metal, like the additions of cinder. What kind of ore is best adapted for this purpose is a some- what scientific question; but experience shows that none answers so well as magnetic ore ; and this is generally employed. If mag- netic ores cannot be obtained, and if it is necessary to employ oxides, or hydrates, it is advisable to burn the latter hard, and* to convert them into a black magnetic oxide, before we use them. The leading principle which guides us in the selection of an ore is its amount of iron and its purity. Sulphurous, phosphorous, and cal- careous ores will of course be rejected. Unless the amount of iron in the ore is greater than that in the cinder we are making in the furnace, we shall fail to realize the advantage we expect. If there is more foreign matter in the ore than the cinder generally contains, we shall obtain iron in smaller amount, and of worse quality, than though no ore had been added. The best cinder from charcoal forge iron contains scarcely more than eight or ten per cent, of fo- reign or silicious matter; the residue is iron and alkaline substances. The amount of silex varies from ten to thirty per cent, and even more; and its increase beyond ten per cent, is inversely proportional to the quality of the iron. This shows clearly what is required for the improvement of iron ; that is, alkalies and metallic oxides. Al- kaline earths, such as lime, magnesia, and baryta are not adapted for this composition, because the temperature of the puddling fur- nace is so low that they will not combine with the silicious matter; and they injure the cinder, by stiffening it. An ore serviceable for puddling may contain manganese, clay, soda, potash, and silex ; but if it contains lime, magnesia, baryta, sulphur, phosphorus, 296 MANUFACTURE OF IRON. copper, silver, and more than twelve per cent, of silex, it must be rejected. The native magnetic ores are, under all circumstances, preferable. At Lake Champlain, and in Essex county, New York, an abundance of suitable ore exists. New Jersey contains a small quantity of ore which, though very silicious, is well adapted for our purpose. In Missouri and Wisconsin, such an ore appears to exist in abundance. But, in the anthracite and bituminous coal regions, the iron master is placed in a somewhat difficult position. The richest hydrates of Huntingdon or Lebanon county, in Eastern Penn- sylvania, or those from the Cumberland River, Tennessee, may serve as fluxes; but they must be converted, by roasting, into magnetic oxides, before they will serve for the improvement of iron. Magnetic ore of good quality will, of course, serve as well as hammer-slag for boiling, that is, for raising the cinder. Though the application of cinder and iron ore rests upon sound principles, it is still limited to certain qualities of metal, and never produces anything beyond a certain kind of bar iron belonging to the cinder we are able to generate from ore. We are thus some- times left in a difficulty, if we expect a kind of bar iron which it is beyond the capacity of our cinder or ore to furnish us. In such cases, which are not unfrequent, we should apply to chemistry for assistance ; but we must be careful not to waste time in seeking that which it is not the province of chemistry to supply. It is un- questionable that, in Pittsburgh, iron of good quality is produced at puddling furnaces where cinder is applied. But this mode is so expensive and troublesome that it will be abandoned as soon as stone coal iron is puddled, the prospects of which are gradually brightening. The application of good ore, though preferable to cinder, is limited to certain localities, for ore whose price exceeds six dollars is scarcely available. In nearly every instance, arti- ficial fluxes are the safest and cheapest of all fluxes, and, when intelligently applied, produce results which we have yet failed to derive from any cinder or ore. The materials suitable for these artificial compositions are very limited ; and their application requires experience. The kind of material used is of less consequence, in the results obtained, than the manner in which it is employed. Caustic potash, caustic soda, manganese, iron, and clay may be employed with advantage. All other matter, even carbonates of potash, is useless ; lime and mag- nesia are injurious. Soda is preferable to potash. We are, there- fore, reduced to soda, manganese, and clay, as the only available MANUFACTURE OF WROUGHT IRON. 297 substances not already contained in the ore. In many instances, pig iron contains an amount of manganese which renders the appli- cation of any additional quantity superfluous. The materials, then, at our disposal in fact, the only ones we need are soda and clay. In some cases, common salt or borax is useful. But, under all cir- cumstances, whatever matter we employ should be mixed and ground as fine as possible. Our own experience has taught us that, unless this is carefully attended to, success is somewhat problemati- cal. For this purpose, rotary iron barrels, like those employed in foundries for grinding charcoal, are employed. The materials are mixed in definite proportions. A small fire is kept under them to dry the clay, and mix the soda intimately with it. The contents are then ground into an impalpable powder, which must then be placed in a dry, warm place for preservation. When moist, even though under the influence of a dry atmosphere, its virtues are greatly diminished. To illustrate the operation of artificial fluxes, we shall relate our own experience in regard to the different kinds of pig iron for which they were employed, and indicate, as we proceed, the various compositions which we tested. We shall present, at first, the most simple cases, and gradually ascend to those which are more complex. 1. Metals smelted by charcoal from phosphurets. It is imma- terial to what degree this iron may be cold-short, provided it is mottled or gray pig, or the result of heavy burden. By a judi- cious application of Shafhseutl's compound that is, five parts of common salt, three of manganese, and two of clay it will produce an excellent bar iron, equal to any iron from the best charcoal metal. The clay alluded to is not a silicious, sandy, white matter, or common loam ; but the finest white plastic clay, which, when wet, is very tough, and when dry, of smooth appearance ; it forms an impalpable powder. The pig iron is heated as in common ope- rations. It is melted down by a rapid heat ; the damper is closed ; and the cinder and metal diligently stirred. In the mean time, the above mixture, in small parcels of about half a pound, is introduced in the proportion of one per cent, of the iron employed. If, after this, the cinder does not rise, hammer-slag may be applied. Where competent workmen are employed, a good furnace will make a heat in two hours, and furnish highly satisfactory results. Where the operation is well conducted, there will be neither too much nor too little cinder in the furnace. From a rolling-mill of which we have personal knowledge, containing six double furnaces, not even a 298 MANUFACTURE OF IRON. wheelbarrowful of cinder was carried away, while no cinder, in addition to the roll scales, and the cinder supplied by the furnaces, was added. - 2. Pig iron, from sulphurous ore and heavy burden, smelted by charcoal. The appearance of this metal is very black. Under ordinary circumstances, it produces very red-hot iron. It was melted and wrought by the same method as No. 1, with this differ- ence, that, instead of clay, chalk was employed. In the furnace, it worked somewhat faster than the above, but always produced an iron inferior to it. 3. Gray charcoal iron, of any cast or mottled iron, will produce with great facility, by working it in the same way as No. 1, a superior fibrous iron ; but great industry is required to make as fine and strong an article as No. 1. 4. Gray anthracite iron, if free of sulphur, requires the mixture of No. 1 ; but if it contains any trace of sulphur, No. 2 will answer better. This remark also applies to gray coke iron. But coke iron is less easily wrought into a good article than anthracite iron. Neither works so well as charcoal pig. The main difficulty in working them consists in melting-in. But, by careful and industrious manipula- tion, we shall arrive at as satisfactory results as with charcoal iron. 5. From white iron of small burden, or from an excess of lime- stone or manganese, it is useless to attempt to produce a good ar- ticle. A small amount of such iron, containing phosphorus or sul- phur, will make a whole charge cold-short, or hot-short, and it is impossible to remove silex from it. By the addition of a very small portion of soda and clay, the better kinds of such iron may be ad- vantageously puddled. If caustic soda is not too expensive, it may be considered preferable to common salt. One pound of soda and one pound of clay are sufficient for 500 pounds of iron ; or, if caustic soda is not applied, two pounds of common salt. All inferior and irregular metals, whether charcoal, anthracite, or coke iron, should be sent to the refinery, melted into finery metals, and puddled in furnaces with brick or soapstone lining, in which operation a small addition of clay and soda will be found advantageous. From these demonstrations, we see of what importance pig iron, which melts and keeps liquid for a given length of time, is to puddling establish- ments. Such iron is produced only by blast furnaces which carry heavy burden and consume a very limited amount of fuel. i. In puddling manipulations, we must be careful that the fur- nace hearth is kept tight, and that the cinder does not leak through the bottom even in the strongest heat. A heat which loses its cin- MANUFACTURE OF WROUGHT IRON. 299 der is spoiled. The quality and quantity of the metal are inju- riously affected. The amount of cinder required in the furnace depends upon the metal we use, upon the competency of the work- men, and upon the iron we design to make. Good puddlers will work to advantage with a small quantity; but poor workmen re- quire an abundant supply. With a small quantity, the work is ac- celerated. Inferior requires more cinder than good pig iron, and gray more than white. If we desire strong iron, of fine fibre, we must employ gray pig of a fusible nature. By diligent work, with- out adding any scales or hammer-slag to the mixture, No. 1 will furnish an iron of unsurpassable absolute strength. If we desire to make wire iron, it is necessary to employ gray pig containing a large amount of carbon, and flux it by means of caustic soda and clay. If expense is no object, borax may be em- ployed with even greater advantage. In this instance, a fine-grained iron, of steel fracture, but softer than steel and harder than fibrous iron, is required. Fibrous iron is not adapted for the manufac- ture of wire. It does not draw well, and is not so strong as iron of a fine-grained nature. Such iron should be free from impurities and cinder ; for these not only weaken it, but make the wire short and unclean, besides working hard on the draw-plate. To remove these impurities, we require a very alkaline, but at the same time very fusible, cinder. Such a cinder will make a fine compact iron, exhibiting no fibres in the rod or billet, but only in small wire. Jc. The height of the furnace top, or arch, from the bottom, varies, according to circumstances, from eighteen to thirty inches. In puddling furnaces, from eighteen to twenty-four inches ; and in boiling furnaces, from twenty-two to thirty. The latter is the ex- treme, and seldom applied. Inferior pig iron which melts easily, and keeps liquid, requires a higher arch than pig iron of good quality. Gray metal produces better iron by a high than by a low top. White metal of any kind works more favorably by a low arch, for' which reason it is puddled, and not boiled. A high arch works more slowly and consumes more fuel than a low arch, but the yield is superior both in quantity and quality. Wire iron requires a strong heat, but a high top. Good puddlers will work with a low arch ; but such an arch cannot be intrusted to inferior workmen. I. The depth of the bottom, that is, the iron bottom below the door-plate or cinder-plate, is as variable as the roof. From four to six inches is sufficient in a puddling, and from six to twelve in a boil- ing furnace. In some cases, the latter is rather too great a depth; ^300 MANUFACTURE OF IRON. and eleven inches may be considered the extreme. A deeper hearth is required for bad than for good pig iron. A deep bottom con- sumes more fuel, and requires greater attention, than a flat bottom, but it makes better iron, and yields it in larger quantity. A large body of cinder does not make very fibrous, but very clean iron. m. The dimensions of the grate of a puddling furnace depend upon the size of the hearth, and upon the kind of fuel employed. For wood, in small chips, a grate whose size is in the ratio of one foot to twelve feet of hearth, is sufficiently large ; for bituminous coal like that of the Pittsburgh vein, one foot to four. For hard and impure coal, it may be extended to half the size of the hearth. But where blast is applied, as in the case of Pennsylvania anthra- cite, these rules must, in some measure, be modified. However, if we have any doubt about the matter, it is better to make the grate too large than too small. The only disadvantage of a large grate is that it consumes a greater amount of fuel than one of the proper size. n. The influence of fuel upon the quality of the iron manu- factured is not remarkable; but, in some instances, it is important. Sulphur and phosphorus do not appear to have any influence what- ever upon the iron in the furnace, for we have experienced no dif- ficulty in puddling with sulphurous coal or turf, the latter of which generally contains phosphorus in admixture. Wood undoubtedly affects the process very favorably. We have observed very closely two establishments in which the same kind of metal was puddled by wood and by inferior anthracite; the abilities of the workmen about equal. The iron puddled by wood was strong, white, and of fine fibre. That puddled by anthracite was equally strong, but dark in the fracture, and of coarse fibre. In the blacksmith's fire, the superiority of the former was still more apparent. The only reason which can be assigned for this difference is the difference in the composition of the ashes of the wood and anthracite. Wood ashes are of an alkaline, the ashes of anthracite are of an acid, nature. With wood or bituminous coal no difficulty is experienced in pud- dling, on account of the ashes ; but with anthracite, the ashes have at times proved a serious obstacle. The application of blast, and the use of large grates in the modern anthracite furnace, have, in a great measure, removed this obstacle. Anthracite iron, though sufficiently strong and malleable, is frequently of so dark a color as to be unfavorable for blacksmith's use. This color is imparted by the ashes carried over from the grate upon the hearth. These ashes, which are not pure earth and silex, contain a large amount MANUFACTURE OF WROUGHT IRON. 301 of carbon. If they cover the surface of the exposed iron, carbon will be inclosed in the balls. The silex is then absorbed by the protoxide of iron, and a black cinder is formed ; when, to this, black carbon is added, it is not strange that the iron becomes dark in the fracture. The most effectual preventives of this are quick work, and an abundance of very fusible cinder. Still, it may be stated, as a general rule, that iron puddled by a small amount of cinder, and by the application of stone coal, no matter of what kind, is of dark fracture. If more cinder is applied, the same pig iron will exhibit a brighter fracture. o. Heating stoves attached to puddling furnaces are, where fuel is expensive, and where competent workmen are employed, a valuable appendage. By their use, a quarter of an hour or more is saved each heat. If the same wages are paid, with or without stoves, it is advantageous to employ them, because the time saved by their use may be profitably employed in improving the quality of the iron, or, if this is satisfactory, in working inferior metal. Where they are used, it is advisable to give the metal only a cherry red heat, and to keep it in the stove as short a time as possible. The best situation for a stove is between the pillars of the stack. This is common at the New England works, where the flue is made narrower than usual, to keep the heat more in the furnace hearth. In Europe, many varieties of such stoves are in operation. At Pittsburgh, and throughout the West, no use is made of them, for there fuel is cheap beyond comparison, and the metal employed works so fast that hurried manipulation is unnecessary. But when our Western friends shall be obliged to abandon the running-out fire, and shall be compelled to boil their iron, the application of stoves will be necessary ; not for the saving of fuel, but for the saving of time. This will be particularly the case where a large body of hot blast iron is to be wrought, for this generally works very slowly. Stoves designed to heat the pig iron beyond a cherry red heat, or even to melt it, are not advantageous in the way of appendages. p. At the Eastern establishments, wages for boiling vary from three dollars and fifty cents to four dollars per ton ; the latter for anthracite coal and for anthracite pig. At the West, six dollars per ton are paid for boiling and five for puddling helpers' wages included. q. The yield depends very much on the nature of the metal, and upon the mode of working it. In puddling good white metal, there is a loss of four or five per cent. With bad white metal the loss is 302 MANUFACTURE OF IRON. twenty per cent., and even more where we seek to improve its quality. Gray pig metal, of whatever quality, may, by a judicious applica- tion of ore and artificial fluxes, be made to yield from ninety-five to ninety-eight per cent, of rough bars per 100 cent, of metal. VIII. Creneral Remarks on Refining. Our present run-out fire finery is, as we have previously re- marked, an imperfect apparatus. Its design is to improve the quality of pig metal, and to diminish the labor of converting it into wrought iron. This design is partly accomplished ; but the ap- paratus is still far from being complete. It is not our object, at present, to suggest any improvement upon it, but simply to define its purpose, and to exhibit its imperfections. At the time coke iron was first made, a large quantity of bad iron was, as might have been expected, produced in the blast fur- nace. Such metal, of course, worked well neither in the charcoal forge nor in the puddling furnace. In such a case, but little could be expected from the run-out fire ; because then the nature of the charcoal forge and of the puddling and blast furnaces was not well understood. But this offset to imperfect results cannot now be so successfully pleaded by the manufacturer. Gray pig iron from a well-conducted blast furnace operation can be puddled to the best advantage, without refining ; and it is generally admitted that such iron is of superior quality. In addition to this, experience shows that it is cheaper than refined iron. Therefore the only iron left for refining is that which results from badly-conducted blast fur- nace operations. That the run-out fire is not the best apparatus for effecting that result, is sufficiently proved by experience ; but, theoretically, the fact may easily be demonstrated. If ore is once reduced to iron, the result may be a very imperfect metal. Still the largest amount of matter in it is iron. In the crude metal, it is seldom less than ninety per cent. This element is, in all instances, the same ; it is as favorable in bad as in the finest metal, with the exception of being adulterated by some ad- mixtures which, under certain circumstances, are injurious. If such metal has been properly treated in the blast furnace, we find no difficulty in producing from it a good wrought iron. But if badly managed, it is almost impossible to obtain this result. The abun- dance of bad iron in the market is a proof that the finery accom- plishes but a slight improvement. The main difficulty in working pig metal smelted by a high heat in the blast furnace consists, MANUFACTURE OF WROUGHT IRON. 303 as we have before explained, in the impossibility of so completely dissolving it as to enable the cinder to act upon the foreign matter in it to advantage. In most instances, it readily dissolves by a tolerable heat; but the cohesion of its particles is so great, and its affinity for oxygen so strong, that it does not remain liquid sufficiently long for the removal of its impurities. The objections made against cold boshes in the puddling furnace apply with greater force against the finery fire, for in the latter cold boshes not only exist, but are in a less advantageous form than in the double puddling furnace. There- fore, the finery fire eifects scarcely any improvement in the quality of iron, and is, of course, not adapted to produce cheap work. The chief purpose of the run-out fire is the manufacture of a more uniform metal than is produced by the blast furnace. By bringing the metal to a somewhat uniform quality, we are enabled to secure more regular manipulation both in the forge and in the mill. But this advantage can be arrived at in a more perfect man- ner by very different methods. Another advantage it is said to pos- sess is, that it does not consume so much iron, in neutralizing the silex of the pig metal, as the puddling furnace. This is true ; but if the run-out fire works by coke, which is generally the case, all the ashes of the fuel are saturated by iron, and a large quantity of the sand which forms the bottom of the fire. These objections are of a practical nature. We know that it is vain to attempt to improve radically bad metal by running it through the finery. We know, further, that this is not the method of making cheap iron. The finery fire will waste from six to fifteen, and even twenty per cent, of metal. We may say that ten per cent, of this is, on an average, uselessly lost ; for, in the puddling furnace, we can produce a yield equal to that in the finery, whether the metal is refined or not. This loss without considering the wages of workmen, from one dollar to one dollar and fifty cents, and the expense of fuel, which for coke is seventy-five cents, and for charcoal two dollars and fifty cents amounts to four dollars per ton of iron, that is, if we estimate the iron at but two cents per pound. The expenses of refining, there- fore, in the most favorable case, amount to at least five dollars and a half per ton. Pig iron which it is impossible to improve effectually in the pud- dling furnace will always be manufactured. Besides, it is necessary to puddle iron for railroad and heavy bar iron. For this purpose, we need white plate metal. For the manufacture of this article, we require an apparatus superior to any we at present possess an 304 MANUFACTURE OF IRON. apparatus more in accordance with the advancement which science has made in all of its various departments. In the following pages, we shall endeavor to point out still more completely than heretofore the deficiencies of the run-out fire; and for the purpose of assisting inventive genius, we shall add the different methods of refining at present practiced in various parts of the world: a. 1. By charging to excess, in the blast furnace, ores containing phosphorus; these will produce white metal with the least injurious admixtures. 2. By casting the pig iron directly from the blast furnace in iron moulds, and cooling it suddenly by a current of cold water. 3. By running the re-melted iron into a mass of cold water. 4. By the making of rosettes, practiced in Styria,and described in Chapter III. an effectual method, unless the metal is very bad. 5. By tempering the metal; this is done by exposing it for twelve or twenty- four hours to a cherry red heat. 6. By refining the iron in the blast furnace before tapping ; this is effected by turning the blast upon the hot iron, and in this way burning impurities and carbon. 7. By feeding good ore, hammer-slag, or wash iron at the tuyere. Wash iron is the fine grains of iron gathered by pounding the fur- nace cinder, and washing it in a current of water; the water car- ries off the sand of the cinder, and the grains of iron left amount, in many instances, to six or seven, seldom less than three or four, per cent, of the cinder. This method is extensively practiced in Western Germany, where poor silicious ores are smelted. 8. By melting the pig iron in the common charcoal forge, and, by chilling it in water, preparing it for the following operation. A division of labor is thus practiced in the same apparatus : 9. By melting the pig iron in a reverberatory furnace, and by blowing upon it, as practiced in this country and elsewhere ; or by washing the melted metal with ore, hammer-slag, or other ingredients, to make it white. Of all the methods described, of all that are known, none is so well adapted to improve iron as the puddling furnace. The most useful in the above enumeration is the refining of the iron in the blast furnace before it is let out ; but this method is not generally applicable, and it would not answer at all in an anthra- cite furnace. A method of improving iron, commonly employed, is by imme- diately cooling the hot metal in iron moulds, or by the application of water. Generally, both means are resorted to in the same case. As far as the removal of impurities is concerned, no improvement is effected; for the iron contains as much silex, carbon, and sul- MANUFACTURE OF WROUGHT IRON. 305 phur after it is chilled, as before. Still, the manipulation is pro- ductive of great advantage. We shall endeavor, in a future page, to explain the nature of this curious process. Metal designed for the forge should be cast in iron moulds, if for no other purpose than to keep it free from sand. Of late years, attempts have been made to remove impurities by galvanizing iron ; but such experiments are so scientific as to be productive of no practical utility. What may be done, is not always profitable in business. We do not depreciate the motive power of electricity ; but we must be permitted to doubt that it can successfully compete against the steam-engine, so far as economy is concerned. IX. Theory of Refining and Puddling. We now proceed to the examination of a subject which is no less difficult than interesting. It unfolds to us the nature of the material with which we have to deal, and shows us to what extent we can succeed in improving the quality of metal by converting it into wrought iron. We shall probably succeed in conveying a bet- ter understanding of this subject, by pointing out the nature of pig iron and wrought iron. a. The chemical difference between cast iron and wrought iron consists principally in the difference of degree in which foreign matter is present in each ; which is in larger amount in the former than in the latter. 'We should be cautious not to infer that this rule is universally true ; that is, by applying it to iron from different sources. This rule is applicable only to a given cast iron, and to the wrought or bar iron which is made from it. There are many cases in which wrought iron contains a larger amount of impurities than cast iron, and yet is malleable ; while cast iron of the same composition may be very hard and brittle. Berzelius, a celebrated Swedish chemist, tells us that he detected, in a certain kind of bar iron, eighteen per cent, of silex; and that this iron was still mal- leable and useful. One-tenth of that amount of silex will make cast iron brittle. The foreign matters generally combined with pig iron are carbon, silicon, silex, sulphur, phosphorus, arsenic, zinc, manganese, titanium, chrome, aluminium, magnesium, and calcium. Each of these tends to make iron brittle. Therefore, in converting cast into wrought iron, it is necessary, as far as possible, to remove them. Carbon, and, as far as we can judge, all other foreign matter, divide the crude iron into two very distinct classes. 20 306 MANUFACTURE OF IRON. In the one, carbon is only an accidental mechanical admixture; in the other, it is in definite chemical combination with the iron. To the former belong the white iron of heavy burden, and gray iron; to the latter the white iron of small burden, or very fusible ores. Judging from the behavior of the different metals in the refining and puddling process, we are inclined to believe that the presence of silicon and silex has a similar influence, for it is almost impos- sible to remove silex from white metal with which carbon is chemi- cally combined. The silex is present very probably in the form of silicon. This accounts for the great difficulty of improving such metal by any refining process. The same remarks apply to phos- phorus and sulphur. White metal of small burden may contain from five to nearly six per cent, of carbon ; and, if smelted from poor ore, almost an equal amount of other foreign matter, such as silicon. Upon the presence andform of these, its white color and crys- talization, in a great degree, depend. Gray pig iron seldom contains more than 4.75 per cent, of carbon, and generally only from 3.50 to 4 per cent. When carbon is present to the amount of but two to three per cent., it becomes white. We know, from experience, that white iron of heavy burden behaves well, and that it can be greatly improved in the puddling furnace ; but with less facility in the charcoal forge. We also know that it is almost impossible to im- prove white iron of poor origin, and light burden, containing carbon in large amount, by any method of manipulation. If the presence of carbon were the only difficulty to be overcome, we should not de- spair of working it advantageously. In fact, the better kinds of this metal are worked with success in the charcoal forge ; but in this forge the poorer kinds will not work at all. In the puddling fur- nace, the former will produce a good article, though not equal to gray metal from the same ore ; but the latter will yield a very infe- rior iron. Hence we conclude that, in this metal, silex is present in the form of silicon, and that it is chemically combined with the metal as an alloy. This remark also applies to calcium and phosphorus. By these combinations, the difficulties encountered in our attempts to remove impurities from metal, are explained. Were carbon the only difficulty against which we have to contend, the metal could be made to work well in any apparatus. Were the protoxide of iron the only alkali in the cinder, this of itself would absorb any amount of oxidized silicon or silex. But, in consequence of the reduction of silex to silicon, the latter must first absorb oxygen ; and this it can absorb only from the protoxide of iron in the cinder. The MANUFACTURE OF WROUGHT IRON. 307 silex, thus forming, will attract three atoms of iron in absorbing oxygen. The oxygen, in turn, will form very adhesive white iron, which crystalizes rapidly. Its crystals include carbon, silicon, and even silex. In this instance, we require a far larger amount of oxygen to remove impurities than in the case of gray iron ; and this oxygen can be derived only from the oxygen of the cinder, or, what is the same thing, from the protoxide of iron which the cinder contains; or, where the cinder is very alkaline, from mag- netic oxide of iron. Hence it follows that, to saturate the silex formed, an immense quantity of iron is taken from the metal itself, to be oxidized by the atmospheric air. This oxidation raises the temperature of the metal, and separates the carbon before the re- maining impurities can- be effectually removed. The only method of improving such metal is to melt it by a very high heat, and under cover of a very strong alkaline cinder. But this is not feasi- ble either in the charcoal forge or in the puddling furnace. b. In composition, wrought iron is frequently inferior to the metal from which it is made; that is, if we apply the term inferior to the preponderance of foreign matter which it contains. Wrought iron containing a large amount of silex and carbon, especially the first, and even a given proportion of phosphorus, may still be a good bar iron. The main difference between pig and wrought iron con- sists in their mechanical structure or aggregate form. Pig iron is a homogeneous mixture of impurities and metal, in which, by affinity, atom is brought close to atom, and in which a transforma- tion from the mechanical to a chemical admixture is easily effected, as in the case of gray and white metal. Wrought or bar iron is a mixture of iron more or less pure with a mass of homogeneous im- purities, or cinder the latter filling the crevices between the crys- tals of the iron. If we remember that iron is fusible in proportion to the carbon it contains, we shall arrive at a very comprehensive con- clusion. If we melt metal or pig iron, and expose the cinder which surrounds it to the influence of oxygen, the carbon will evaporate, and iron of greater or less purity will remain. This iron, to be kept liquid, requires a higher temperature than at first; conse- quently, unless the temperature is raised, it will crystalize. In this state of metamorphosis, its infusibility will increase, and after the expulsion of the carbon, it will contract into a solid mass by the highest possible heat. By stirring and mixing the pasty iron, small crystals are formed ; at first, on account of the partial fusing of the iron, in small particles ; but, as the fusibility diminishes, 308 MANUFACTURE OF IRON. these particles unite by force of cohesion; and the bodies thus formed may, by exposure to a higher heat, be welded together. The mixing of cinder and iron will prevent the latter from forming large crystals. This result, of course, will be more easily prevented by diligent than by tardy manipulation. Where the pig iron is of such a nature as to keep liquid while the work goes on slowly, still better results will be afforded. This process is analogous to that of salt-boiling, in which, by stirring the brine, the formation of large crystals is prevented. If the crystals of iron thus formed cohere, they form, under the influence of motion, a porous, spongy mass, whose crevices are, if not filled, at least coated, with cinder. If these masses, which are the loups or balls at the charcoal forge and puddling furnaces, are shingled or squeezed, the crystals of iron will not unite, but form coated cells with a film of cinder, of greater or less thickness, according to the fusibility of the cinder. Iron in a connected form, and cinder in separate cells, are thus blended in a homogeneous mass. The more this iron is stretched, the more it forms fibres. Fibrous bar iron resembles hickory wood, in the fact that it is a combination of fibres and spaces. In bar iron, these spaces are filled with cinder. When other circumstances are equal, the strength of the iron will be proportional to the fineness of the fibres. That portion of the iron which is not melted, which crystalizes too fast, or whose premature crystalization the workman cannot pre- vent, is in the condition of cast metal, and cannot be converted into fibrous wrought iron. In the puddling furnace, it is necessary to prevent crystalization by manual labor. This result, whether in the Catalan forge, the wiilfs oven, or the German forge, is partly accomplished by the blast. If the characteristic difference between wrought and pig iron consists in nothing else than such a well-regulated mechanical mix- ture of cinder and iron, we ought to be enabled to produce fibrous wrought iron from any cast iron, whether it is or is not purified by the process we have described. This is actually the case. Very fibrous bar iron, which is strong and malleable, is made from very inferior metal in which no removal of its impurities is effected. Among other instances, which, we may observe, are very frequent, in which this result is accomplished, we may mention that, at Hyanges, France, very inferior metal is converted, by a cheap and skillful puddling process, into a very fibrous bar iron, of great strength and ductility. But this iron is puddled and re-heated by the lowest possible heat ; it is then rolled, and ready for market. For MANUFACTURE OF WROUGHT IRON. 309 hoops, rails, and nails, it is a very useful article; but it is of no use to the blacksmith. Heated by any temperature beyond that of the puddling and re-heating furnaces, it returns to its primitive state, in which condition it becomes worse than the cast iron from which it was originally made. None but a very skillful blacksmith can weld it ; for, when slightly re-heated, it falls to coarse, sandy pieces, or melts like pig iron. That which thus loses its fibrous texture in heating, the smith calls burnt iron. c. Fibrous iron returns in this manner to its original condi- tion because of the impurities which exist in it; these absorb oxygen. It is of little consequence whether such impurities are carbon, silicon, or calcium, for each of these will reduce the cinder. Let us assume that carbon is present in fibrous bar iron. If we heat this iron to a certain temperature, the carbon which it contains will combine with the protoxide of iron in the cinder, and form iron and carbonic oxide. The latter escapes, and leaves silex in the pores of the iron. The silex, thus enclosed, will not prevent the cohesion of the crystals into an aggregated mass. This mass is then in the same form as crude iron. Silicon acts in the same way as carbon, and so does any element which has more affinity for oxygen than iron. Therefore, the destruction by heat of the fibres in iron is nothing else than the result of the destruction of the cin- der, which, by its vitreous nature, prevented the formation of larger crystals in the iron. The sudden cooling of such iron will produce the same result. Wrought iron of a white color, fine fibre, and yielding when struck a dead sound, is not liable to these alterations. It remains fibrous under all conditions, and is altered neither by heat nor water ; that is, provided the heat is not excessive, or of too long duration. Such iron must be free of all carbon or elementary im- purities, or its cinder must be of such a nature as not to be altered by carbon or silicon. In Styria and Carinthia, iron of good weld- ing properties, very fibrous, and remaining fibrous, however often it is heated or cooled, and very tenacious in fact, a perfectsample of excellent iron is manufactured from carbonaceous, spathic ores, which contain a large amount of manganese. Bar iron, from what- ever source, if manufactured from metal smelted by a well-con- ducted process from ore containing manganese, is generally of the same character ; at least, it is always the best for blacksmith's use. These facts show conclusively that manganese has a favorable influ- ence upon iron. Manganese has greater affinity for oxygen than 310 MANUFACTURE OF IRON. iron; and its oxides are powerful bases, stronger even than the oxides of iron. If cinder, composed of manganese and silex, is that which produces the fibre in the iron, carbon will have but little influence upon the cinder, for manganese is reduced to metal only under very favorable conditions. It is with still more difficulty separated from silex. If the destruction of fibres in bar iron be thus prevented, it is evident that a stronger alkali would be still more favorable. d. The removal of carbon from pig iron is of less consequence than the removal of silex and other foreign matter from it. The first may be effected with comparative facility, as is proved by the case we have related in which iron from the purest plate metal was converted into black magnetic oxide. In that case, the plate iron contained at least five per cent, of carbon; but it was so totally destroyed that the heat of the re-heating furnace converted the iron into an oxide of iron. This shows clearly that the very exist- ence of iron depends upon the presence of a given amount of car- bon; otherwise, the iron, when exposed to heat, will absorb oxygen into its minutest particles. Analyses prove that the best kinds of wrought iron contain from 1.2 to 1.4 per cent, of carbon, in addi- tion to sulphur, phosphorus, silicon, manganese, arsenic, and tita- nium. The two last elements are frequently found in Swedish iron, and generally in iron manufactured from magnetic ore. Of all these substances, silex, phosphorus, and sulphur are with the greatest difficulty removed; and, when present in too large quan- tity, those only which so injure the metal as to make their removal necessary. If they are combined in an oxidized state with the pig iron, they may be removed with but little difficulty. But this is not the case with phosphorus and sulphur, for, in the presence of the large amount of carbon which exists in the blast furnace, phos- phurets and sulphurets are formed; and these are mixed with the metal. We know positively that silicon may exist in an oxidized form in the metal. Silex, generally in admixture with ore, is re- duced to silicon only by a very favorable heat. Very strong cohe- sion alone will form a chemical compound of iron, silicon, and car- bon. This state of things appears to exist at the blast furnace, where fusible ores are smelted by an excess of coal, want of pres- sure in the blast, blast of too great pressure, hearth of too great width, and imperfectly prepared ores. From such operations, we obtain pig iron which is with difficulty improved; whence we con- clude that the injurious impurities so difficult of removal are chemi- cally combined with the iron. MANUFACTURE OF WROUGHT IRON. 311 e. Carbon can exist in iron in two distinct combinations. It is mechanically mixed with the gray pig, and chemically with the white iron of small burden ; the former contains it in smaller amount than the latter. We know less of silex. We know that the metal contains silex, but of the form in which it is present, whether as silicon, or its oxide, silex, we are ignorant. Iron, carbon, and sili- con have a great affinity for each other, but, so far as we can judge, it requires a given temperature and certain conditions to develop a polarity sufficiently strong to blend them so intimately together that the specific quality of each is lost in that of the other. This is actu- ally the case between carbon and iron ; and it is rational to expect that it is the case between silicon and iron, as well as between iron, phosphorus, and sulphur. The two latter combine with iron at almost any temperature. If such a chemical connection between silicon and carbon happens to exist in metal, it is evident that their separation must be a matter of extreme difficulty. Silicon has a stronger affinity for oxygen than either carbon or iron ; for this rea- son, as well as for the large amount of oxygen required to oxidize the silicon, it is so difficult to remove it from the iron. The amount of carbon may be very large in some kinds of metal, particularly those smelted from poor silicious ore, or those smelted by coke and anthracite, or the result of light burden. If such metal is re-melted in a puddling furnace, or in any apparatus ia which there is access of atmospheric air, the silicon will absorb oxygen, should it exist. If, perchance, atmospheric oxygen, or, what is still worse, watery vapors have access to the metal, the temperature of the particles of iron is raised to such a height that a portion of carbon will evaporate. The metal, thus transformed, will, by its infusibility, enclose portions of metal which are brittle. The fibres of such iron we cannot expect to retain. If present in the rough bars, they will disappear when the iron is re-heated. This result, theoretically deduced, coincides with practical observa- tions. The fast working of such metal, when it is to be wrought in the charcoal forge, may be retarded by throwing sand on the par- tially refined iron. Sand dissolves a part of the iron which encloses the injurious particles, by forming with it cinder ; and gives the en- closed iron a fusible protection. It also retards, in the mean time, the too rapid absorption of oxygen. Such iron, of course, yields badly in the forge fire ; the wrought iron which is produced from it will be inferior. Metal so impure should never be taken to the charcoal forge. In the puddling furnace, it is worked with but 312 MANUFACTURE OF IRON. little advantage, though with greater success than in the charcoal forge. As previously remarked, puddling furnaces with cooled boshes are, in this case, of no use ; for this metal requires a high heat, and a large quantity of cinder, to make it work slowly, to protect the iron and carbon, and gradually to oxidize the silicon. When it is worked by the addition of oxidizing cinders, or of water; or when it is melted slowly ; or where oxygen has access to it, which happens if there is too small an amount of coal in the grate, the result is a bad article, and a poor yield. In our investigations, we invariably fall back upon the white metal containing carbon in large amount. But when we consider its frequent appearance in the forges, its bad qualities, the fact that it originates from an imperfect blast furnace manipulation, and, finally, its relation to hot blast, we hope we shall be justified in this course by the intelligent reader. In fact, this subject affords the best illustration of the theory of refining. Besides the white metal, composed of iron, carbon, and silicon, which may result from the very best ores, as is the case in the mag- netic ore regions, there is white iron, with an admixture of phos- phorus and sulphur. The latter is inferior even to the former, if the result of light burden. But the worst of all metals is that smelted from bad ores, and from an excess of limestone, and hot blast. In addition to carbon, silicon, phosphorus, and sulphur, which may be removed, this metal contains calcium and magnesium, the elements of alkalies, which destroy every prospect of improvement with the means at present at our disposal. As we cannot entirely get rid of this metal ; as its quality is of such a nature as, thus far, to have baffled the most acute ingenuity ; and as it contains at least eighty- five or ninety per cent, of good iron, we trust that the time will not be considered entirely lost which we shall consume in a some- what close examination of the subject. This may serve to suggest a method of improving it, or of preventing the production of so large an amount of it as heretofore. /. All metal smelted beyond a certain temperature, and produced under specific conditions, is white ; often of a bluish color, if it con- tains the elements of the alkaline earths. We suppose that the earthy matters, silex, lime, and magnesia, are reduced to metals, and chemically combined with the iron. If we rnelt such iron with access of oxygen, it will of course be transformed into a pasty mass, because a part of it is not fusible, that is, already deprived of carbon, for this is the only means which can effectually secure fusi- MANUFACTURE OF WROUGHT IRON. 313 bilitj. Silicon and calcium increase the fusibility of metal ; but these are oxidized by the slightest exposure, and thus serve to diminish its fusibility, and in this way, as well as by the increase of temperature resulting from oxidation in the metal itself, destroy the carbon. If the carbon could be retained, it would tend to keep the metal liquid ; it would thus offer a chance of acting upon the impurities. For these reasons such metal will work fast, unless we are desirous of improving its quality, which, with the better kinds of charcoal metal from rich ore and judicious blast furnace manipulations, is generally our object. But when we attempt to improve its quality, the metal works slowly, yields very poorly, and is seldom or never made to produce an article of tolerable value. We know, practically, the difficulties connected with such metal, and from facts we have presented, have deduced a rational theory, which, if judiciously applied, ought to show in what direction our attempts at improvement should tend. If, in consequence of oxi- dation, the metal becomes less fusible, and thus encloses impurities, a mechanical separation that is, fast work and the application of strong cinder, the latter of which will keep the iron separate ought to be a means of improvement. Such is actually the case. By the application of strong cinder where silex predominates, a dark fibrous bar iron is produced, which, when re-heated, returns to a more or less strong cast iron, that is, a cold-short bar iron of a crystaline or granular fracture. This kind of puddling is very frequently practiced, especially in our anthracite region ; but that which most clearly illustrates such work is the puddling process at Hyanges, to 'which we have so often referred. At that place, the pig iron is melted-in with a mixture of feldspar and squeezer cinder; in this the iron keeps pasty for a long time. By means of the feldspar, cinder of a silicious composition is formed which is very fusible ; before the feldspar is melted, it is almost sandy. Its fusibility in- creases as the furnace becomes warmer, and as the work progresses. "When its fusibility arrives at a point at which its utility appears questionable, the cinder is let off, and the iron, by this time ready for shingling, is quickly balled up, and taken to the hammer. In- ferior pig iron may thus be worked to advantage ; but its quality cannot be much improved. In this case, we see by what means a fibrous iron can be produced from a metal which, to all appear- ances, it is impossible to improve. But, at the same time, it sug- gests the method by which improvements may be effected. It shows that an alkaline cinder is of no advantage in working such 314 MANUFACTURE OF IRON. iron. As the removal of silicon or silex is the principal object at which we aim, the correctness of the method pursued in working white iron is somewhat doubtful ; nevertheless, it is the best with which we are acquainted. g. Metal whose quality is such as either to produce, if worked on a cheap plan, bad iron, or to produce it in such limited quantity as to make its application unprofitable, is said to be the result of too much heat in the blast furnace, or the effect of hot blast. Low temperature and cold blast do not make such iron, or, at least, not very frequently. Therefore, the cause of the difficulty lies chiefly in the conditions of the blast furnace. Before hot blast was intro- duced, such pig iron was generally the consequence of too wide a hearth. When it happened to be smelted, the furnace was blown out, and a new hearth put in. In this case, the cause of bad iron must be attributed to the absence of free oxygen in the hearth, as well as to an excess of fuel. Where cold blast is employed, gray iron is made in a high, narrow hearth, and flat boshes; but where hot blast is used, it can be made in a low hearth, and steep boshes. Consequently, in the latter case, the heat is high up in the stack, for which reason the pig iron will contain a large amount of chemically combined impurities, and will become hard and brittle;, Therefore, hot blast does not affect the iron in any other way than it is affected by cold blast and too wide a hearth. It is thus evident that the conclusion which most writers on this subject have arrived at, that a too high temperature is the only cause of the inferiority of the metal, needs qualification. That is to say, if, where the hearth is too wide, and the boshes too sloping, the same kind of metal is produced by cold blast which, under other conditions, is produced by hot blast, the conclusion that an excessive temperature occasions the bad metal referred to, is obviously erroneous. From this argument, which holds good in many other respects, we infer that chemical composi- tions of iron and impurities are the result of conditions in which free oxygen is excluded from acting upon the metal. This explains at once the necessity of the differences in the construction of blast furnaces for different ore and fuel. For gray pig iron, or pig iron in which impurities and iron are mechanically mixed, a narrow hearth is required. On what hypothesis can we explain the fact that silex, in this case, is not reduced, and that carbon is not more closely connected with the rnetal ? Not on that of a want of heat, for the heat is more concentrated in a narrow than in a wide hearth. The only way of accounting for so singular a fact is, either that the MANUFACTURE OF WROUGHT IRON. 315 melted metal is prevented, for want of time, from chemically com- bining with silex and carbon, or that such a combination, already existing, is destroyed by free oxygen. Against the first hypothesis, we may urge the fact that a too wide hearth forms chemical combi- nations ; for in a wide hearth less time is allowed for the iron to come down than in a narrow hearth. In support of the latter hypo- thesis, we may adduce the result of our experience in the manufac- ture of steel. German steel is manufactured from white metal simply by melting it down in a charcoal forge. It may be made with the greatest facility, though of inferior quality, in a very hot puddling furnace. In this case, the very best kind of metal is re- quired; but this will not affect our argument. The steel metal is iron and carbon chemically combined, and the steel itself iron and carbon mechanically combined. By re-melting the metal, with the observance of certain practical rules, the chemical combination is destroyed. In this instance, we see that it is not an excessive temperature which produces inferior white iron. It is a heat rela- tively too high, under given circumstances. These circumstances appear to be the too rapid conversion of the oxygen of the blast into carbonic oxide, or the passing of the reduced ore or metal before the tuyere under conditions in which it is either not touched at all, or touched in very slight degree, by the blast. Too high a tuyere in the blast furnace is also very apt to produce such metal. In the chapter on hot blast, we shall make some additional remarks on this subject. From the foregoing, we may draw the conclusion that white metal, chemically combined with impurities, is produced under circumstances imperfectly understood; and that this chemical com- bination, by re-melting the metal, and carefully observing certain practical rules, can be dissolved, and changed into a mechanical admixture. How the metal is made in the blast furnace, is a matter about which we may be indifferent. The only question which concerns us is, how to improve it. That such metal can be converted into steel, is an evidence that it can be improved; but this conversion involves a greater expense of fuel than bar iron which is made from the same metal will justify. In the former case, a high heat, and the presence of carbon, are all that is neces- sary to dissolve the chemical connection of the compounds. In such metal, designed for steel, but little silex is generally found, for it is made from the richest ores. Where a large amount of silex is to be removed, the case is different. 316 MANUFACTURE OF IRON. We would not devote so much space to the consideration of this metal, were it not so common an article. If there existed a method of improving it, it would be still more common ; but, until lately, this has been difficult. It is not the object of this work to propose improvements, the results of which may be doubtful. All such matters we consign to the enterprise of the industrious manufac- turer, who does not despair after the failure of an experiment. But we may be permitted to point out here the leading principles which such improvements must necessarily embody. A high heat is re- quired to dissolve the chemical combination of iron and foreign matter. A limited amount of heat and oxygen is sufficient to re- move carbon from this combination. In the removal of silicon, carbon and oxygen are needed; otherwise the iron becomes in- fusible. If calcium and magnesium are mixed with the iron which, however, is seldom the case the difficulties which we encounter are so grave that we may safely say, with the means at present at our service, that the improvement of the metal is an impossibility. h. The rule, that cinders are the criterion of the quality of the iron made, is in no instance more correct than in the present case. The removal of impurities, of which silex and silicon are the most injurious, is the main object in the refining of iron. Inasmuch as the principles of refining are the same, whatever is the substance "with which we may be engaged, we shall confine our attention principally to silex. To remove silex, which is an acid, we re- quire an alkali with which to combine it, and thus to form a vitrified, fusible slag. The greater the affinity in the slag for silex, the greater the amount of silex which will be removed. Therefore, in a forge cinder, we need as much alkali as we can possibly obtain, for upon the quantity and quality of this will de- pend the quality of the iron. Forge cinder from a charcoal fire contains more alkali than cinder from the puddling furnace. This accounts, in some measure, for the difference in the quality of iron which each produces. On close examination, we shall find, in this circumstance, the reason why the charcoal forge will not work in- ferior metals to advantage, and why the puddling furnace does not produce from good metal an article equal to that from the charco-al forge. The reason is evident. Metals containing a large amount of silex must, if we desire a good article, necessarily be partially converted into cinder; because good cinder requires a given amount of protoxide of iron to neutralize the silex which it contains. If MANUFACTURE OF WROUGHT IRON. 317 the silex is not thus neutralized, the iron will be worse than the same metal from the puddling furnace ; because, in addition to silex, we leave more carbon in the iron, on account of the presence and contact of charcoal. This makes the iron in the charcoal forge more cold-short than that in the puddling furnace. Intrinsically, it may be purer, but it is not generally more useful. In conse- quence of its bricks and coal ashes, consisting almost exclusively of silex, the puddling furnace cannot produce so good a cinder as the charcoal forge, in which everything can be kept free of silex. Therefore, the puddling furnace makes better iron from poor metal than the charcoal forge ; but the latter makes better iron than the former from good metal. To what extent the qualities of iron are connected with the com- position of cinder may be understood by comparing one cinder with another. From a rolling-mill in Firmy, France, a puddling cinder contained 31.2 silex, 60.5 protoxide of iron and manganese, and 1.7 phosphorus. In this case, we observe an immense quantity of protoxide of iron compared to that contained in blast furnace cinder. We also observe a diminution of silex, besides a large portion of phosphorus. Puddling cinder from Dowlais, Wales, consisted of 36.8 silex, 61.0 protoxide of iron, and 1.5 clay or alumina. The first cinder is from charcoal pig iron ; the latter from coke iron. Re- heating furnace cinder contains about 40 or 50 per cent, of silex, and often clay, in proportion to the sand used in making the hearth. Charcoal forge cinder from a forge where good iron, though on a cheap plan, is manufactured Hartz Mountains, Germany con- tained 32.3 silex, 62.0 protoxide of iron, 1.4 magnesia, 2.6 prot- oxide of manganese, and 0.28 potash. The amount of protoxide of iron increases, and that of silex de- creases, in proportion to the quality of the iron. A cinder from a good iron in France consisted of 16.4 silex, 79.0 protoxide of iron, 3.0 lime, 1.2 alumina, and 0.6 protoxide of manganese. A Swedish cinder, from the softest kind of iron, was composed of 7.60 silex, 82.10 protoxide of iron, 2.80 magnesia, 1.10 alumina, and 6.80 protoxide of manganese. And a cinder from a strong kind of Swedish bar iron was composed of 38.5 silex, 44.4 protoxide of iron, 3.1 lime, 3.1 alumina, and 11.0 protoxide of manganese. From these analyses of cinder, we deduce a leading principle which will guide us in directing the refining operations. The in- crease of alkalies in the cinder shows the method by which we must arrive at a good result. But it is necessary to guard against the 318 MANUFACTURE OF IRON. very natural conclusion that the iron will be improved in proportion to the increase of alkali in the cinder. This is not the case, at least so far as the strength of the iron is concerned. As before explained, wrought iron is nothing else than cast iron with a judicious admix- ture of cinder. If that cinder is of such a nature as to be decom- posed by the remaining carbon, or other unoxidized impurities, the fibrous bar iron will return to cast iron, on being re-heated. Therefore, the absolute cohesion, or strength, of wrought iron, is not dependent upon the degree of purity of the metal, but upon a given mixture of cinder and iron. Pure iron, .which is always soft, may be required for various purposes as, for example, in the manufacture of cast steel ; but, in most cases, an impure, but fibrous iron is preferable. In making wrought iron, the main diffi- culty consists, not in producing fibres in the first stages of the ope- ration, for this may be accomplished by almost every experienced manufacturer, but in retaining these fibres through every subse- quent stage of the operation. We do not think it necessary to enter upon an investigation relative to the proof of our statement that the absolute strength of iron does not consist in its purity, but in its aggregate form. Nobody will doubt that a fine thread of the worst iron is stronger than an equally fine thread of glass. Yet how elastic is such a thread of glass ! Who doubts that, if all the fibres of a hickory stick were united in one solid mass, without pores, the absolute strength of the stick would be greatly diminished ? at least that its strength would be far inferior to a bar of iron of poor quality ? Yet how far superior is the fibrous hickory to cold-short iron, with respect to relative strength and elasticity ! But it is unnecessary to dwell upon so plain a subject. i. To form bar iron of a permanent fibrous structure, the cinder employed must be of such a nature as to resist the reducing influence of carbon and other unoxidized compounds of iron. Forge cinder is chiefly composed of protoxide of iron. One reason of this is that it is the nearest alkali at hand. Another is that the protoxide of iron, and the silex separated from the iron, are approximate at the moment of liberation, and, according to general laws, combine more readily than they would if once separated. Protoxide of iron is, under ordinary circumstances, with difficulty reduced to iron, particularly if once in chemical connection with silex ; but its re- duction is possible, as is proved by the reduction of forge cinder in the charcoal forge, and puddling cinder in the blast furnace. We may expect that carbon, and, in a still greater degree, silicon, con- MANUFACTURE OF WROUGHT IRON. 319 tained in bar iron, will reduce the protoxide of iron in the cinder, and leave pure silex, or other matter ; that it will, at any rate, de- stroy the vitrified texture of the cinder in the pores of the iron. A good cinder, of general application, would be one whose vitreous nature could not be destroyed by the influence of the impurities of the iron. A gray, glassy, tough blast furnace cinder would be the best of all materials for this purpose ; but it may be impossible to mix such matter properly with iron, or to mix with iron any glass which contains no metallic oxides, and which is not acted upon by reducing agents. The development of the nature of cinder will thus be the surest means of arriving at a correct understanding of the nature of bar iron. Carbon will reduce protoxide of iron from its combination. Therefore, for poor metal, the latter will not form a satisfactory cinder until all the carbon of the metal is destroyed. Where carbon remains in wrought iron, re-heating will restore the granular texture of the metal. Any metallic oxide, forming glasses, which can be reduced to metal by carbon, is useless in the forma- tion of forge cinders, for it would not serve to retain the vitrified character of the enclosed cinder. Protoxide of manganese is bet- ter than that of iron ; it is only slightly affected by carbon ; but silicon will reduce one part, and combine with another part of it. It forms an excellent glass, which resists the destructive agency of carbon. Iron manufactured by means of a rich manganese cinder is very strong, of good welding properties, and retains its fibres in al- most any heat, and even when suddenly cooled in cold water. Be- sides manganese, the alkaline earths and the alkalies proper are the only substances at our service. Alkaline earths are of no use in the forge, for the temperature of the hearth is low, and sufficient time is not afforded for their combination with silex; they serve merely to stiffen the cinder, and add impurities that are injurious to the iron. When present in large quantity, they frequently prevent the for- mation of fibrous iron. The alkalies proper, such as soda and pot- ash, ought to be the best agents in forming a good cinder ; still, experiments have not confirmed the conclusions which, theoretically, have been arrived at. Until the present time, no benefit has been derived from so apparently practicable a theory. Bad pig iron contains carbon, silicon, and calcium, which should be partially removed ; or, if not removed which, in some instances, is unnecessary we should employ a cinder which, mixed with any metal, is not affected by the reviving properties of the impurities. If we melt such pig iron, and add to it carbonate of potash or of soda, 320 MANUFACTURE OF IRON. the carbonate will not combine with silex; if once combined, it can- not effectually remove protoxide of iron from the cinder; it serves the purpose of simply making the cinder more fusible ; it dissolves the oxides of metals, but it does not dissolve silex, lime, and mag- nesia; it will augment the fusibility of a strong alkaline cinder, and to this extent promote vitrification, but it cannot prevent the for- mation of protoxide of iron. Caustic potash or caustic soda is evapo- rated in the heat of a puddling furnace, before any union with silex can be effected. Potash and soda, mixed, are of greater advantage, for they offer a stronger resistance to the action of the heat ; but our own experiments have convinced us that even these are inappli- cable, because the greater part of the alkalies is lost in evaporation. Were it practically possible to make a cinder composed of potash, soda, and silex, and mix it with any kind of metal, however bad it may be in our estimation, the bar iron resulting would be strong, its fibres durable, and it could be welded with ease. But it would lose its strength, in proportion to the oxidation of its foreign matter, that is, the matter originally combined with the metal. From these statements, we infer that the application of manganese is the best means of improving the quality of wrought iron. To what extent this improvement may be applied, has been already explained. It will, we hope, be more clearly understood from the following considerations deduced from experience : k. When a heat is drawn and shingled, the furnace must be so uniformly charged as to prevent the re-melting of any portion before the melting of the rest. This is accomplished by a repeated turning and moving of the iron. The melting of the cinder before the iron becomes soft is a disadvantage, for when the cinder covers the frag- ments of iron, the difficulties of breaking the iron and mixing it with the cinder are augmented. The most favorable results follow when the iron and cinder melt together, that is, when both become pasty at the same time. It is of less consequence when pig iron is the more fusible than when the reverse is the case. To produce this state of things is sometimes a difficult matter ; still, upon its ac- complishment, success mainly depends. This difficulty is augment- ed by the fact that the composition of the cinder is not a matter of indifference. We can increase or diminish the fusibility of a cinder by adding to it an alkali, or silex ; but this may injuriously affect the iron in the furnace. White metal is very apt to make a strong alkaline cinder, rich in protoxide of iron ; this cinder will not melt sooner than the iron melts. In such cinder, poor white metal works MANUFACTURE OF WROUGHT IRON. 321 too fast ; sufficient time is not afforded for the metal to dissolve. Besides, it contains too large an amount of oxygen, or protoxide of iron ; and the metal, by means of silicon and carbon, is converted into an infusible white iron. In such cases, a cinder which contains just so much silex that its fusibility will be increased by the appli- cation of an alkali is preferable, and therefore it is advisable the lining of the furnaces should be of fire brick or stone. Such cinder will increase in toughness, at first, because the first matter liberated from the metal is silex; and the addition of silex to a silicious cinder will retard its fusibility, and afford more time to work the metal. In this, also, we perceive the utility of charging the iron along with the cold cinder. Any rich cinder will afford oxygen to the melting metal, but the application of it in too large amount will accelerate the work beyond the limits of prudence. White metal, especially that which is not to be improved, ought to be melted-in without any cinder ; but the grayer the metal is, the longer it remains fusible, and the greater may be the amount of cinder which can be charged along with the cold iron. Thus far, the melting-in of the iron is the most important part of the operation ; but it is evident that the true cause of the difference in work, or the difference in cinder, will not be found out by practical manipulation. The management of this part of the operation is the duty of the manager of the establishment ; that is, the manager should give the general directions, in accordance with which the iron should be worked. If we are in doubt as to the propriety of charging cinder along with the metal, it is better to forbear, because we are more sure of obtaining good work by melting the metal in without any oxidizing agent. When the iron is properly melted down, that is, when it does not rise, or exhibit crystalized particles, we may accelerate the work by throwing in good iron ore finely powdered, roll-scales, hammer-slag, chemical compounds, water, or, in fact, anything which experience has shown tends to consummate the desired result. The matter thrown into the furnace, at this stage of the operation, will deter- mine the quality of the iron which is made. If we desire to make hard iron, we should leave a certain amount of silicon and carbon in the iron; and if, at the same time, we desire to produce fibrous iron, we must be very cautious in relation to applying anything of an oxidizing nature. Neither hammer-slag, manganese, water, nor any kind of iron ore is applicable. Active manipulation, and a fur- nace that is not too warm, will produce a hard, strong, fibrous iron 21 322 MANUFACTURE OF IRON. of a dark color, but not adapted for blacksmith's use. By the in- troduction of hammer-slag, roll-scales, or magnetic iron ore, a purer iron may be made. Of these, magnetic ore is the preferable. Hammer-slag and cinder always contain the greater part of the im- purities of the iron from which they are derived, especially sulphur and phosphorus. Cinder is the very element which removes impu- rities ; hence, if we introduce an impure material, we cannot ex- pect a pure article of iron. Cinder has only a limited capacity for sulphur or phosphorus ; but this increases proportionally to the amount of alkali added to it. Therefore, even though we introduce a very alkaline cinder, such as hammer-slag, roll-scales, and forge cinder, which may have been obtained from a very good iron, still, in a puddling furnace, its capacity for impurities is diminished, be- cause a large portion of the alkali is absorbed by silex. In this way, we may spoil our iron in the furnace with the very material we employ for its improvement. This unfortunately too often happens. Mag- netic iron ore serves all the purposes of hammer-slag; and if it can be obtained in purity, free from sulphurets, it is by far the safest of all means for improving iron. To the manufacturers of the East, vast quantities of magnetic ore from Lake Champlain, or New Jer- sey, are at all times available. The Western establishments are less advantageously situated ; for the only serviceable ore which they possess, so far as our knowledge extends, is the compact mag- netic ore of Missouri. Still, it may be possible to obtain useful ores from the head waters of the Alleghany River. By the application of good ore, and by the observance of sound principles of manipulation, we may obtain iron applicable to all our common wants. In this way, merchant iron of the finest quality may be made. If it is our intention to make a very superior iron, neither mag- netic ore nor hammer-slag will be of muck service; in fact, they are, in a greater or less degree, injurious. The reason of this is that, in a very alkaline cinder, the solvent power, for the magnetic oxide, is rather small ; hence, a given portion of the flux is left un- dissolved; this, as an isolated matter, will be visible in black spots and grains, darkening the fracture, and adding nothing at all to the strength of the iron. In addition to this, it produces coarse pores, which, for fine polished work, are injurious. A fine, superior iron, of great cohesive properties, requires a very alkaline, well fluxed, but not too fusible cinder a cinder free from all mechanical ad- mixture, or imperfectly dissolved matter. An alkaline cinder is re- quired, to remove impurities from the iron, and a well fluxed cinder MANUFACTURE OF WROUGHT IRON. 323 to form fine fibres. A strong, tenacious cinder is needed to resist the union of the iron fibres, as well as the deoxidizing influence of the carbon in the iron. These conditions are fulfilled in the char- coal forge only with the best kinds of metal ; in the puddling fur- nace only where the lining is of iron, and where very fusible gray pig is employed. Gray pig iron, smelted by charcoal in small amount, is best adapted for the manufacture of a superior, very strong, pud- dled iron. To make a good cinder, we require strong alkalies, and if even in excess, the cinder must be perfectly vitrified. The latter result may be produced by applying those alkaline salts which have the pjpwer of dissolving a surplus of alkali, or metallic oxides, as, for example, protoxide of iron ; but if silex is to be removed, the acid of the introduced salt must not be too strong to resist the influence of the silex upon its alkali. Such salts are the carbonates, borates, phosphates, chlorides, and a few others which are not practicable in our manipulation. Carbonates of the alkaline earths require the heat and time which a blast furnace affords to be of any service ; and the carbonates of potash and soda are not only expensive, but they are weak solvents. Borates are, of all fluxes, the most perfect, but of a carbonizing, reducing character; therefore, borax is a most powerful reducing agent. In the puddling furnaces, it produces a very pure, but carbonized iron. Phosphates may be considered the most perfect of fluxes, in puddling. The fear which prevents some iron manufacturers from placing phosphorus in connection with iron, is without foundation in theory or practice. Our own experience and that of others prove that there is not the least difficulty in re- moving phosphorus from iron, or even in smelting iron ore contain- ing phosphorus, without injury to the metal, that is to say, provided carbon has not been present in so large an amount as to leave phosphoric acid undecomposed. Much carbon is required to de- compose phosphoric acid. In the puddling furnace, phosphates cannot be decomposed, and they remain as a solvent for cinder of a very alkaline nature. Any free alkali, in the presence of carbon, will abstract phosphorus from iron ; so, also, will the stronger alkalies, such as barytas, soda, and potash, without the presence of carbon. Phosphates are soluble in any silicate ; and if silex increases, so that no alkali is left with which the phosphoric acid can combine, the latter, unless carbon and a portion of the protoxide of iron in the cinder are present to reduce it, will evaporate, or form phosphuret of iron, which, of course, will then remain in combination with the iron. Where there is a large 324 MANUFACTURE OF IRON. amount of phosphorus in pig iron, we require an equivalent amount of free alkali for its removal. The latter, in the course of the puddling process, which is an oxidizing process, will become a phosphate, even though its first compound was a phosphuret. As a phosphate, it cannot be of injury to the iron, if the latter is free from carbon or silicon. Therefore, the wish to exclude phosphorus containing ores is an unfounded prejudice. In the charcoal forge, iron containing phosphorus cannot be wrought to advantage. As fluxes, phosphates occupy a position between borates and chlo- rides ; they are not so much of a reducing agent as the former, nor so much of an oxidizing agent as the latter. Chlorides, sipch as common salt, are very powerful oxidizing agents; therefore, in the blast furnace, they are not in their legitimate place ; for, to make gray iron, when a considerable amount of chlorine or a chloride is present, is impossible. But under certain conditions, chlorides are the best materials to improve iron in the refining process ; they are far superior to the strongest alkalies. They are more permanent than borates, phosphates, or sulphates. But, in the presence of carbon in excess, they are very volatile. When in the condition of neutral compounds, they are but little inclined to associate with other salts; if the latter are well heated, they evaporate. Chlorides possess great power of dissolving alkalies in a heated condition ; and before chlorine is moved by silex, it will drive off every other acid. The employment of common salt in the refining of iron, thus shown to be unquestionably useful, is of very limited application, owing to the difficulties involved in its use. Where carbon is present, chlo- rides are useless, for, in a given heat, they will evaporate without leaving a trace behind. Where the heat of a puddling furnace is quite high, as in the melting of some kinds of pig iron is neces- sary, any chloride introduced into it will immediately evaporate ; and thus time is not afforded for its combination with the alkali of a cinder, even though an abundance of it is present. We now perceive the causes of the divergence between science and practice. No element is so well adapted to improve iron as common salt. But, in consequence of imperfect knowledge, its application, thus far, has been extremely limited. In fact, because it has failed to produce certain results, which a knowledge of its nature and con- stitution ought to teach us we have no reason to expect, many igno- rant persons have refused to employ it at all. In puddling, the fur- nace ought to be as cold as possible, if salt, or any chlorides, are applied. Therefore, iron which melts at a low heat is preferable ; MANUFACTURE OF WROUGHT IRON. 325 the chlorides must be very dry, and finely ground; and a large quan- tity of cinder is required to prevent, as far as possible, the immediate contact of the chlorides and iron, before the solution of the latter takes place. If chlorides are brought in direct contact with the iron, the chlorine and a portion of the iron will evaporate, and consequently, no benefit will result from the oxidizing nature of the cinder. A cinder containing chlorine is a powerful oxidizing agent. Neither silex, phosphorus, sulphur, manganese, nor iron can withstand its influence, even though it exists in small amount. It will oxidize phosphorus, sulphur, and carbon, and cause their destruction; and of these elements it aids to form acids, which are either combined with the alkali of the cinder, if any are free, or are ejected in the form of sulphurous and phosphorous acids, and carbonic oxide. For these reasons, an excess of salt is very injurious, for it will evaporate, and oxidize a large portion of iron, and the iron which is produced will be dark and weak ; but by applying it in proper quantity, we shall obtain an iron which, in strength and color, cannot be surpassed. The above are the only elements which are serviceable for the improvement of iron. But, as this statement may seem to require some explanation, we shall enter upon a brief consideration of those materials which might possibly be made to serve the desired pur- pose. We shall only enumerate oxidizing agents, for these alone at present interest us. The sulphates are almost superior to chlorides as oxidizing agents ; but the danger of decomposition, by which sulphurets would be left in the iron, precludes their use. The decomposition of the acid would at once deprive the cinder of its oxidizing power, and the sulphuret left in the metal will remain in the bar iron, unless the cinder contains a great excess of alkali. Black manganese gives off its oxygen too soon; it does not serve a much better purpose than any other alkali. It possesses greater strength than the protoxide of iron, but it is inferior to soda or pot- ash. Iron refined under its influence is generally hard, fibrous, and strong. Soda and potash are excellent alkalies ; but, when applied in such quantity as to remove sulphur or phosphorus, they are too fusible to make strong iron. In small quantities, they are serviceable where the iron is of good quality ; but in large quantities, they increase the fusibility of the cinder to such a degree that the iron cannot become strong and fibrous. Besides, so strong is the heat of a puddling 326 MANUFACTURE OF IRON. furnace, that the greatest part even of the carbonates will evaporate before a combination of the alkali and the cinder is effected. Oxide of lead is perfectly useless, because of the shortness of the time in which it loses its oxygen, but we have applied the basic chloride of lead with success. This chloride is obtained from a mixture of litharge and common salt, in the proportion of four pounds of the former to one of the latter. The mixture is moistened with water, and left to stand for twenty-four hours. We were in- duced to make this experiment from having observed a metal which produced the most beautiful bar iron we had ever seen. The metal was white, of a reddish flesh-colored cast; it was smelted from an iron ore containing lead in admixture. The lead separated from the iron in the lower part of the crucible. This metal, white as any metal can possibly be, though with a reddish cast, melted as thin as any gray iron containing phosphorus. It kept liquid for an unusual length of time. From an inferior pig iron, we obtained, by the application of the chloride of lead, an excellent quality of iron, though not equal to the white metal smelted from the lead containing ores. To the following material we wish to call special attention ; not on account of its quality as a flux, but because of the facility with which it can be applied. Its chemical composition is remarkable; and we are therefore induced to reflect upon the primary condition of those materials designed for the improvement of iron in the pud- dling furnace. We refer to the magnetic oxide of iron. Expe- rience has proved that hammer-slag, roll-scales, and finely powdered magnetic oxide are the best means of promoting the process of puddling. They do not produce the best iron, nor the fastest work ; still, it is unquestionable that they are the most available materials at our service. Black magnetic ore, hammer-slag, and roll-scales constitute the magnetic oxide of iron. This is a combination of one atom of protoxide and one atom of peroxide of iron, forming a neutral compound which is less easily decomposed than the peroxide, and which resists the influence of carbon for a shorter time than the protoxide. If such a compound is brought in contact with cinder, it will be neutral, because it melts only at a very high temperature ; and, unless carbon or some other reducing agent is present, it will remain in the cinder in its original integrity: at least, it will resist decomposition for a great length of time. At first, it stiffens the cinder. This is just what is required ; for a strong cinder enables us to separate the iron properly; at least, this is the object which we MANUFACTURE OF WROUGHT IRON. 327 always aim to realize. In the progress of the work, the uncom- bined particles of the magnetic oxide will come in contact with the melted iron. If the iron is of good quality, and carbon alone re- quires removal, the compound oxide will be decomposed by the carbon ; carbonic oxide will escape with a blue flame on the surface of the cinder ; and the protoxide which results will combine, should it come in contact with silex, or divide the silex of the cinder with the other alkalies. With respect to the removal of carbon, many other materials may be employed with even greater advantage; but, with regard to silex, this is not the case. Any alkali will remove silex from the iron ; but for the removal of silicon, the black magnetic oxide is preferable to even the best alkalies. If silicon exists in the pig iron in which case, oxygen is required to form an acid and the melted iron is brought in contact with the magnetic oxide, the peroxide of the compound is decomposed; oxygen is thus imparted to the silicon ; the newly-formed silex and the newly-formed pro- toxide of iron will then combine instantly. According to the laws of chemistry, this is the most favorable condition under which the combination of bodies will take place. In this case, three atoms of magnetic oxide impart oxygen to one atom of silicon, by which means a single silicate is formed. A more perfect compound than the sesquioxide of iron can scarcely be imagined; still, in the pro- gress of knowledge, some more advantageous method may yet be found for the removal of silicon. Sulphurets and phosphurets are decomposed with facility by the magnetic oxide ; but the resulting sulphates and phosphates are, in turn, decomposed by carbon ; and as no wrought iron is entirely free from carbon, the magnetic oxide is not the best material at our service for the removal of sulphur and phosphorus. Great attention is therefore required so to arrange matters in the puddling furnace that we shall have a cinder slightly less fusible than the iron. With white metal, it should be less fusible for want of alkali ; and with gray iron, by virtue of the absence of silex or acid. But in most cases, it is advantageous to have a somewhat alkaline cinder, because a better yield is produced. After the iron is melted, and fluxes are applied, the cinder begins to thicken, long blue flames escape at the surface, and the whole mass begins to fer- ment. Blue flames appear only when the work goes on well ; in which case carbonic oxide is formed below the cinder. If the cin- der is too alkaline, that is, if it contains too much iron, no such flames appear; but a lively ebullition is visible on the surface of the 328 MANUFACTURE OF IRON. cinder. The same result occurs, if we throw in oxygen in too large amount, or that which is too loosely fixed. The reason of this is the formation of carbonic acid, instead of carbonic oxide, below the cinder ; and it requires very good metal indeed to make valuable iron by such manipulation. To produce blue flames, and to pre- vent the formation of carbonic acid, the most effective means we possess are a cold hearth at the commencement of the operation, diligent manipulation, and the application of the smallest quantity of fluxes that is commensurate with success. After the iron has risen, and the mass has begun to ferment, the quality of the iron is fixed. Nothing but industry is now required, to obtain as large a yield as possible, and to make the iron work well at the squeezer. To an unphilosophical mind, the fermenta- tion of the metal appears to be a singular phenomenon. In fact, few sights are more beautiful than that of a mass of iron from 700 to 1000 pounds' weight, occupying at first scarcely the depth of an inch in the hearth of the furnace, gradually rising, fermenting, and boiling, while small particles of iron, in apparently spontaneous motion, suddenly appear on the surface in small clusters like brilliant stars, and then as suddenly disappear. This fermentation happens only with metal that contains a given amount of carbon. White metal, which contains little or no carbon, does not ferment. Not only the amount of carbon, but the composition of the cinder, in- fluences fermentation. The cause of the boiling is nothing else than the evolution of gas, generated by the combination of oxygen and carbon. If the cinder which covers the liquid iron is very fusible, the gas escapes in bubbles on its surface, and the metal does not rise. If the cinder possess a certain tenacity, if it is slimy like soap water, it will resist the escape of the gas, and rise until its surface isc lose to the flame of the furnace, when, becoming warmer, and more liquid, its power of resistance is diminished. The slimy consistency of the cinder is produced by silex, but more perfectly by clay ; the latter may be derived from the pig iron, or it can be charged with the fluxes. If we reflect upon this quality of clay, which is the same under all circumstances, we shall arrive at the cause of its beneficial influence upon iron. Our exer- tions are chiefly directed towards obtaining a well-defined cinder ; neither too acid nor too alkaline. Clay fulfils these conditions. It serves both as an acid and as an alkali. It fluxes, and in the mean time, will strengthen the body of the cinder. It is very serviceable in removing phosphorus, for with phosphoric acid it forms a fusible MANUFACTURE OF WROUGHT IRON. 329 compound of great solvent power. Though experience were not in its favor, a consideration of its quality ought to convince us of its great utility in the puddling furnace. If the above explanation of the causes of fermentation is correct, it follows that the process de- pends upon the quality of the metal, and upon the nature of the cinder. Nevertheless, an experienced and skillful workman will make almost any metal boil, provided it contains but little carbon. As the fermentation proceeds, the iron coagulates, that is, crystal- izes below the cinder. As the small particles formed still contain a portion of carbon, which combines with oxygen derived from the cinder, the newly-formed carbonic oxide rises, and, in its ascent, draws along with it a small crystal of iron, which, coming to the surface, burns for a moment with a vivid light, and then disappears, because, after it loses its bubble of gas, there is nothing to coun- teract its gravity. This motion of the particles of the iron continues until the carbon of the metal is exhausted, or until the oxygen of the cinder is so diminished that no more gas can be formed ; after which the cinder gradually contracts, and sinks to the bottom of the fur- nace, leaving the iron, to a greater or less degree, unprotected, and exposed to the heat. Metal containing silex, silicon, phosphorus, sulphur, but no carbon, will not ferment, for no gas is liberated to cause fermentation. Therefore, such metal is greatly exposed to the heat and oxygen of the furnace, and works too quickly; hence the difficulty of improving it, even though it is very fusible. We thus see the advantage of fermentation in working inferior pig iron. As boiled iron is preferable for small iron rods, wire iron, black- smiths' iron, and nails, we should always seek to obtain gray pig for the boiling process. After the fermentation is finished, the oxidation of the iron com- mences; for, if the process has been properly conducted, all the previous operations will have tended only to remove impurities. The time at which this takes place depends upon the time occupied in finishing the heat, and upon the amount of silex the cinder con- tains. The oxidation of the iron serves to flux the slag, which be- comes more and more liquid, as the temperature of the hearth increases. At this stage of the process, the utility of the iron boshes is evident, for, should the furnace have been lined with bricks or stones, all the alkalies we have applied, and all the iron which has been burnt, would have been wasted in their destruc- tion. Besides, the main object of our skill and industry is to in- 330 MANUFACTURE OF IRON. crease the amount of alkali in the cinder; but this object is directly counteracted by brick and stone boshes. I. Having delineated, though by no means having exhausted, the various matters which relate to puddling, we shall take a critical view of the present mode of refining. We shall also investigate the cause of the improvement which results from the sudden cool- ing of metal, and shall conclude the chapter by a few general re- marks on wrought iron. The run-out fire, which is generally employed for refining iron, is based upon principles derived from the charcoal forge. Before hot blast was introduced into blast furnace operations, this was doubt- less a useful apparatus. Pig metal which, fifteen years ago, would have been considered worthless for the forge, is now employed in the manufacture of iron. The run-out fire labors under the same difficulties which exist in relation to the puddling furnace with iron boshes. For iron which contains carbon in small amount, or in chemical combination, its hearth is too cold. From gray charcoal pig iron, of good quality, the run-out produces a tolerably useful article. But we do not need it for this purpose. Cold blast gray pig may be worked to advantage in the puddling furnace with- out Difficulty. Since the introduction of hot blast that is, since the use of anthracite and stone coal quite a revolution has taken place in the chemical constitution of pig iron : the amount of chemi- cally combined carbon has increased; silicon and other reduced matter are more generally present ; and even the grayest specimens of metal are not free from unoxidized elements. To these causes, the difficulty of refining hot blast iron may be mainly attributed. Our previous investigations have proved that a high heat is required for the removal of silicon: but a still more necessary element is a cin- der which does not too freely yield its oxygen. To what extent does the run-out fire fulfil these conditions ? With respect to heat, it is but little better than the puddling furnace ; and with respect to cinder, it answers scarcely a better purpose. Analysis has shown that the cinder of the run-out fire contains as much protoxide of iron as the cinder of a puddling furnace. A finery cinder from Dudley, England, contained silex 27.6, protoxide of iron 61.2, alu- mina 0.4, and phosphoric acid. If we melt very impure pig iron in such a cinder, we cannot produce iron of good quality; this is espe- cially the case should the iron have been smelted by hot blast. For the melting of such iron, we require a cinder containing less alkali. Less alkali is required to make good iron in puddling. From the MANUFACTURE OF WROUGHT IRON. 331 amount of iron in this cinder, it is evident that the run-out fire can- not improve bad pig iron in any high degree, unless there is a serious loss in metal. That this loss occurs is shown not only from con- clusions theoretically arrived at, but from observation. Whatever advantage the run-out fire, in this case, possesses, is that of a division of labor, which, of course, we are not disposed to rate very highly. We believe that, in the construction of the run-out fire, but little science and philosophy have been embodied. At all events, but one principle governs the present case; that is, bringing the inferior qualities of metal, in the cheapest possible way, to a higher standard. The idea of making white metal was undoubtedly derived from the ancient method by which such metal was made in the blast furnace. The latter metal, when smelted from good ore, was, and still is, a prime article in the charcoal forge. Since the introduction of coke, anthracite, poor ores, and hot blast, the iron business has undergone a change. At the present time, we cannot avoid producing pig iron which, a few years ago, would have been considered worthless. This metal it is now our object to bring to as high a standard as the best iron of the ancients. In the accomplishment of this object, it is evident that cautious manipulation and scientific knowledge are required. We cannot believe that any one doubts that the quality of our worst pig iron is equal to that of the ores from which steel metal is made. If such is the case, it should not be deemed an impossibility to make steel metal from our most inferior pig iron. There would be no necessity for making white metal were it not for the railroad, boiler-plate, and heavy bar iron which is needed. For these purposes, boiled does not answer so well as puddled iron. But, if such iron is necessary, it should be well made. The run- out fire is imperfectly adapted to accomplish this result. By destroy- ing the carbon in pig iron, without removing its impurities, it fails to produce a metal fit for boiling. Therefore, the run-out fire de- stroys the element necessary to make metal boil, without producing a metal profitable for puddling. Hot blast iron would be an excel- lent metal for boiling, were it possible to remove its impurities without destroying its carbon. Still, it is not impossible to remove impurities and carbon together, and thus make a useful metal for puddling. The finery is considered a link between the blast furnace and the puddling furnace ; that is, it is believed to occupy the same rela- tive position between these furnaces that the blast furnace occupies 332 MANUFACTURE OF IRON. between the ore and the finery. The blast furnace is an apparatus designed for the removal of impurities with the least possible loss of iron. It answers the purpose of its construction excellently. But what is the fact with respect to the finery? Simply this: the cinder from the finery contains more iron than that from the puddling fur- nace; and, when we consider that the contact of coke or anthracite increases the amount of silex in the former, we find that there is a far greater loss of metal in the run-out fire than would result from the same pig iron in the puddling furnace. If this is the only ad- vantage derivable from the finery, it is surely far preferable to take the worst kind of pig iron directly to the puddling furnace. We trust that some practical men will be sufficiently interested in this subject to endeavor to construct something better adapted to our wants than this exceedingly imperfect apparatus. If heat, and a cinder to protect the iron can be obtained, all the conditions of a good finery will be fulfilled. m. The philosophy of the improvement of metal consists in the circumstance that a part of its impurities, which are originally in chemical combination, are converted into mechanical admixtures. Iron containing a small amount of carbon, silicon, or phosphorus, is always more hard and strong than pure iron. Pure iron is quite soft. Impure iron has the property of crystalizing on being sud- denly cooled. The size of these crystals is proportional to the amount of carbon in chemical combination the iron contains, in proportion to other matter. Between the crystals, minute spaces are left, which serve for the absorption of oxygen. By this means, silicon and calcium may be oxidized; but such is not the case with carbon, phosphorus, and sulphur. Therefore, the metal improves in quality in proportion as oxygen finds access to its impurities. For this reason, the habit of running metal, or any kind of pig iron designed for the forge, into iron chills, is a good one, and is worthy of imita- tion wherever it is applicable. By this means, the absence of sand and the cleanliness of the metal are secured. For the same reason, the metal is tempered; that is, the plates of metal, or, as in some parts of Austria, rosettes of metal, are piled up with small charcoal, braise, and exposed to a lower temperature than a cherry- red heat, for twenty-four or forty-eight hours, in a kind of large bake oven. By this method, the value of the metal is improved for the manufacture of soft and fibrous iron. It is not applicable to plates from which steel is to be made. n. Wrought iron, if of good quality, is silvery white, and fibrous ; MANUFACTURE OF WROUGHT IRON. 333 carbon imparts to it a bluish, and often a gray color ; sulphur a dark dead color, without a tinge of blue ; silicon, phosphorus, and carbon a bright color, which is the more beautiful the more the first two elements preponderate. The lustre of iron does not depend principally upon its color ; for pure iron, though silvery white, re- flects little light. A small quantity of carbon in chemical combi- nation, phosphorus, or silicon increases the brilliancy of its lustre. Its lustre is diminished by silex, carbon in mechanical admixture, cinder, lime, sulphur, or magnesia. Good iron should appear fresh, somewhat reflex in its fibres, and silky. A dead color indicates a weak iron, even though it is perfectly white. Dark, but very lus- trous iron is always superior to that which has a bright color and feeble lustre. Coarse fibres indicate a strong, but, if the iron is dark, an inferior article, unfit for the merchant or the blacksmith. But, where the iron is of a white, bright color, they indicate an ar- ticle of superior quality for sheet iron and boiler-plate, though too soft for railroad iron. For the latter purpose, a coarse, fibrous, slightly bluish iron is required. Iron of short fibre is too pure ; it is generally hot-short, and, when cold, not strong. This kind of iron is apt to result from the application of an excess of lime. Its weakness is the result of the absence of all impurities. The best qualities of bar iron always contain a small amount of impurities. Steel ceases to be hard and strong if we deprive it of the small amount of silicon it contains, or if, by repeated heating, that sili- con is oxidized. This is the case with bar iron. If we deprive it of all foreign admixtures, it ceases to be a strong, tenacious, and beautiful iron, and becomes a pale, soft metal, of feeble strength and lustre. Good bar or wrought iron is always fibrous ; it loses its fibres neither by heat nor cold. Time may change its aggregate form, but its fibrous quality should always be considered the guarantee of its strength. Iron of good quality will bear cold hammering to any extent. A bar an inch square, which cannot be hammered down to a quarter of an inch on a cold anvil without showing any traces of splitting, is an inferior iron. 334 MANUFACTURE OF IRON. CHAPTER V. FORGING AND ROLLING. THE machines adapted for forging and condensing wrought iron vary both in principle and in form. This department of the labors of the iron master is very extensive. But, as our treatise must necessarily be restricted within certain limits, and as this branch of iron manufacture is already highly cultivated in this country our establishments excelling in finish those of Europe generally, and in some respects particularly we shall devote the present chapter to a mere enumeration of the machines required in an iron factory, and explain the principles upon which each is con- structed. We are satisfied that, if a higher degree of perfection is needed in this department, it will be realized by the intellect, skill, and industry of our practical engineers. I. Forge Hammers. a. The most simple machine by which iron is forged is the Ger- man forge-hammer, often called the tilt-hammer. This machine, often of a fanciful form, is very extensively employed. The lead- ing principle which we seek to secure in its construction is solidity; and every variety of form has been invented simply to give per- manency to the structure, which is mainly endangered by the action and reaction of the strokes. The common form of a forge-hammer with a wooden frame is represented by Fig. 95. a The cast iron hammer, which varies in weight, according to the purposes for which it is designed, from 50 to 400 pounds. For drawing small iron and nail rods, a hammer of the former size is sufficiently heavy ; but for forging blooms of from 60 to 100 pounds in weight, a hammer weighing 300 or 400 pounds is employed. Such a ham- mer is represented by Fig. 96, in detail. It should be cast from the strongest gray iron, and secured by wooden wedges to the helve b. The fastening of the hammer to its helve is, in many cases, effected FORGING AND ROLLING. 335 with difficulty, especially if tlie cast is weak ; and to this weakness attention must be paid. If the hammer is of good cast iron, or, as Fig. 95. Tilt-hammer. in many instances, of wrought iron, there is no difficulty in wedgi. , it. If wooden wedges are properly applied, and well tightened . Fit Hammer. long iron wedges may be driven in between them ; but these must be so placed as not to injure the helve, or lie too close to the iron 336 MANUFACTURE OF IRON. of the hammer. These iron wedges are then secured by a sledge, weighing thirty or forty pounds, with a long handle. This is sus- pended on a rope, or, what is better, a small iron rod or chain, ad- justed upon the head of the wedge, by which means a horizontal stroke is secured. The face of the hammer is polished, and in case long bar iron is to be drawn, it is frequently twisted with the helve. c The anvil, a cast iron block, about the weight of the hammer ; but it may be of less weight if the iron stock d is employed, e A log of wood from six to eight feet in length, and frequently four feet in diameter. This log is secured at its base and top by iron hoops. It rests upon timber laid across piles very stoutly driven in the ground. Such a foundation for the log, or hammer-stock, is re- quisite, because rock, or the most solid ground, forms at best but an insufficient base. The two standards sometimes of wood, some- times of iron in which the hammer has its fulcrum, need no de- scription; neither does the mode of fastening them. All that is required here is strength, no amount of which is superfluous. The helve b is of sound, dry hickory, or, more commonly, of white oak. The fulcrum #, a cast iron ring, is represented by Fig. 97 ; this Fig. 97. Helve-ring. must be tightly wedged upon the helve, i A wrought iron ring, fastened to the helve ; on its upper side, it receives the taps of the cams ; on the lower side, it strikes against a vibrating piece of tim- ber for the purpose of increasing, by recoil, the force of the ham- mer. It is easily understood that, if the hammer is thrown up with force, the reaction upon the fulcrum and framework must be im- mense. This is especially the case where a high stroke of the ham- mer is required, as in forging blooms. The destructive power of this reaction increases with the ratio of the weight, and according to the square of the speed. That is to say, if the hammer strikes with 100 pounds force against A, when seventy strokes per minute are made, it will, when 140 strokes per minute are made, strike FORGING AND ROLLING. 337 with a force of 400 pounds. The same rule is applicable in rela- tion to the space described by the hammer. If the hammer, lifted ten inches, strikes with a force of 1000 pounds, it will, when lifted twenty inches, strike with a force of 4000 pounds. This shows the great increase of power which follows that of speed, and imparts some idea of the reaction which machinery of this kind sustains. If the length of the shorter part of the helve, from i to ; of course, produces a heavy shock all through the machinery. To avoid this shock, the top ought not to touch the bottom roller: ln.it Fig. 120. Rollers and pinions for sheet iron. then the pinions are necessary; without them the top roller will not move, and unless this moves, the rollers will not bite. In cases where the top moves independently of the bottom roller, the first is gene- rally balanced by counter weights, applied either below or above the rollers; these weights keep the top and bottom rollers apart. We think that the arrangement indicated in Fig. 109, for keeping the top roller up, is far preferable to any other. The wrenches on the top screws form a cross, so as at any time to expose a handle to the workmen before the rollers. The distance between the rollers must be perfectly controlled by the foreman, because he regulates the thickness of the sheet by these screws. Of all the improvements made relative to the regulation of the distance be- tween the rollers, none is preferable to the above simple mode. c. For making very thin and polished sheet iron, cast iron hous- ings are not sufficiently strong, unless very heavy, and of the best kind of iron. In this case, wrought iron standards are prefer- able; and, as there is no difficulty in obtaining heavy and good wrought iron, at reasonable prices, in Eastern Pennsylvania, where it is manufactured up to seven inches in diameter, it may, in many instances, be advantageous to employ such standards. In 368 MANUFACTURE OF IRON. Fig. 121, a, a represent wrought iron pillars; these are fastened by being cast into the bottom-plate. Each of these pillars is Fig. 121. provided with a screw and nut ; the advantage of taking small, very minute grades of pressure, decrement, upon the top roller is thus secured. For the making of thin sheet iron, this is a very con- venient and essential arrangement. The aprons are broader than at bar iron rollers, which is indispensable. If heavy plates are to be rolled, even small friction rollers in the apron are to be added on the work side. The friction of heavy iron upon the apron is great, and the employment of additional hands would be necessary, if this friction were not diminished by the above friction roller. Sheet rollers move with various speed, and the foreman ought to have it in his power to give to them just the degree of speed re- quired. The speed necessary for these rollers is from twenty to forty revolutions. In a well-conducted establishment, there are roughing-rollers, finishing rollers, and hard or chilled rollers. We generally find only the two first, and in very few establishments the last. d. For making boiler-plate, but one pair of rollers is needed, and the slab rolled down in one heat. The slab, as received from the T FORGING AND ROLLING. 369 hammer, is generally from twelve to eighteen inches long, from seven to ten inches wide, and from two to three inches thick. It is heated, in a re-heating furnace, to a bright red, but not welding heat. The dimensions of the sheet to be rolled from a given slab are produced by turning the slab more or less, and increasing in one direction. The surface of the iron is repeatedly chilled by sprinkling cold water on it by means of a broom; this loosens the adhering scales, which may then be removed by turning the plate, or by the broom. This operation must be particularly attended to when the plate is nearly finished. Polish and great smoothness are not required for boiler-plate. Uniform thickness and good quality of iron are the main requisites. e. Sheet iron thinner than boiler-plate is generally rolled from platines, or from cuttings of flat merchant bars. That which is heavier may be made from one length of the flat mill bar ; and two, or even three thicknesses, when sufficiently heated, are welded to- gether in the sheet rollers. Common sheet iron, as No. 15 and higher numbers, is made from one thickness of the mill bars, which r heated to a cherry-red heat, is run through the rollers in single sheets. At subsequent heats, two, or even three may be rolled together. When heated, the mill bars or platines are brought from the oven in pairs, which are pushed singly through the rollers. This keeps the workmen actively employed; for, while one plate is be- tween the rollers, the other is returned over the top roller, the one thus closely following the other. If three plates are at once in motion, still more active manipulation is required, for, while one plate is between the rollers, the two other plates are in the tongs on each side of them. In the first heat, the iron is reduced as much as possible; and to what extent it may be brought to the desired form, depends on the power of the engine, and the dexterity of the workmen. In this heat, the breadth of the sheet is determined, in case the platines are not already cut to the proper length. After this operation, the iron, which already assumes the appear- ance of sheet iron, is returned to the heating oven, or, as in well- conducted establishments, it is heated anew in a more advantageous oven. From this second heat, two sheets are taken and rolled together, with the caution that, after passing them two or three times through the rollers, they are separated, and their sides re- versed, partly to prevent the adhesion of the plates, and partly to impart a smooth surface to both sides of the sheets. Sheet iron for 24 370 MANUFACTURE OF IRON. the manufacture of nails, and other common purposes, is in this heat generally finished; but a deficiency of power, or want of skill on the part of the workmen, frequently makes it necessary to give it an additional heat. Sheet iron of less thickness and of higher polish, such as that used for stove pipes, requires another heat, and sometimes several additional heats. An ordinary, smooth surface will be produced by passing sheets, two by two, through rollers of tolerable hardness. But if, in this heat, the sheets are passed singly through the rollers, before which a scraper is put to clean the surface from the coarsest part of the adhering scales, a finer surface is produced. For this purpose, common rollers of good close castings are sufficient. But if a still finer surface is required, hard and highly polished rollers are necessary. To such thin sheet iron a high degree of power must be applied, because the iron, when passing through the roll- ers, is very nearly cold. For this reason the rollers are made only from twenty-two to twenty-four inches in length, while their diame- ter is sixteen or eighteen inches ; the housings of great strength, and the power applied greater than in any other case. Highly polished sheet iron of larger size than twenty inches in width, and four or five feet in length, is seldom made. In making sheet iron, it is sometimes difficult to obtain the pre- cise color which the manufacturer desires. Such a color is a bright, light blue, or that of Russian sheet iron. Experience proves that from impure iron we cannot obtain a bright, silvery-looking sur- face. But the color of the best and purest iron may be destroyed by the influence of the fuel, To give a bright surface to sheet iron, we require, in addition to hard and well-polished rollers, the removal of the scales, as far as possible, from the surface of the white or pure iron, which ought to shine through the thin coating of magnetic oxide. The brightest colors are received from the whitest iron. It is thus seen that the color has no relation to the purity of the metal. We have seen very beautiful sheet iron made from very cold-short iron containing phosphorus, and very cloudy- looking black sheets from the best and toughest charcoal iron. If we wish to make a light, fine-looking sheet iron, a portion of carbon, or even of phosphorus and silicon, will be advantageous. In very thin sheets, the most cold-short iron is malleable; and, therefore, in this instance, it is useful. White iron whether the whiteness arises from impurities, or from remarkable purity separates easily from its scales, and is on that account preferable to metal of any other color. FORGING AND ROLLING. 371 /. The color of sheet iron is affected not only by the quality of the iron, but also by fuel, and by the construction of the heating ovens. Sulphur imparts a black color to iron, if present only in very minute quantity; and it may be regarded as an impossibility to make a fine-looking sheet iron in cases in which sulphurous coal is employed. Though the iron, in such cases, may be of the best quality, the sheets will appear of a cloudy, black, or of a dirty, dark blue color. Pure carbon will not injure the color; but when present in connection with sulphur, the color of the iron will be en- tirely spoiled. Therefore, if the sheets are well cleaned in the third heat, all our attention should be concentrated upon the endeavor to protect them against the influence of sulphur, pure air, and against the silicious dust which is thrown out by anthracite coal. This can be effected by high arched ovens, which will prevent the flame from playing on the iron. We should select fuel free from sulphur ; and, if we employ anthracite, we should secure so weak a draft in the oven that no silicious dust shall be carried over from the grate to the furnace. Charcoal is the best fuel. In fine, by employing charcoal, clean iron, a high oven, well-polished rollers, and a suffi- ciently strong power, we shall experience no difficulty in making the finest kind of sheet iron. Cleaning the iron by means of acids is a waste of time, and an unnecessary expense. It may be cleaned by means of a scraper, on the principle applicable to the cleaning of hoops, without difficulty. VII. Re-heating Furnaces. Re-heating furnaces are those which serve to give a welding heat to the iron. In these furnaces, either piles of flat rough bars, or single billets are heated, scraps are welded, and the first heat to sheet iron is given. A re-heating differs but little from a puddling furnace. The same kind of chimney, with the same dimensions, is employed, and the outward form of the furnace is the same as that of the puddling apparatus. Fig. 122 exhibits a re-heating furnace, with the exception of the chimney, which it is not necessary to re- present. The whole interior, with the exception of the hearth a, is made of fire brick. The hearth is constructed of sand. For this pur- pose, a purely silicious sand is required; the coarser the better. Pebbles of about half an inch in size are the very best article we can select. If no sand of sufficiently good quality can be conve- niently obtained, white river-pebbles, or white sandstones, burnt and pounded into a coarse sand, will answer for making the bottom 372 MANUFACTURE OF IRON. of the hearth. The hearth slopes very much towards the flue ; and this inclination tends to keep it dry and hard. Provided Fig. 122. Section of a re-heating furnace. the sand is not carried off by the flowing cinder, the slope can- not be excessive. The iron wasted in re-heating is combined with the silex of the hearth, and forms a very fusible cinder, which flows off through the opening 5, at which there is a small fire to keep the cinder liquid. The sand bottom, from six to twelve inches thick, rests on fire brick. After two or three heats, it is generally injured by the melting cinder, when some additional sand is required to fill up the cavities that are made. The height of the fire brick arch, or its distance from the sand bottom, is seldom more than twelve inches ; and, for common purposes, it can be reduced to eight inches, without injurious results. a. In these furnaces, the grate is very large in proportion to the size of the hearth ; and, with respect to the rules laid down for the construction of puddling furnaces, extremely large. The grate is frequently as large as the hearth, and seldom of less size than half its area. The mean for general practical application lies be- tween these dimensions. Nevertheless, we should be guided by local circumstances, for a size that would be appropriate in one case would not be suitable in another. The rules which govern us in proportioning the size of the grate and hearth depend, as in the case of the puddling furnace, upon the quality of fuel we employ. That is to say, a larger grate is required for anthracite than for bitu- minous coal, and a larger one for the latter than for wood. We FORGING AND ROLLING. 373 should also be influenced by considerations of economy. A large grate produces a greater yield than a small one, provided the roll- ers take the iron fast. A large grate works faster than a small one, and consumes less fuel. Hence, the advantages appear to be in favor of the former. But, in cases of slow work, and iron of small dimensions, the reverse is the fact. The hearth of a re-heating furnace ought never to be longer than five feet, and it may, with advantage, be reduced to three or three and a half feet. A long hearth will produce but an imperfect welding heat ; it works too cold either at the bridge or at the flue. The flue should be as wide as the hearth, and contract gradually towards the chimney. This produces a uniform heat throughout the hearth. Where a hearth is to be made larger for special purposes, such as for heat- ing rail piles, or any other heavy piles, it is more advantageous to extend its breadth than its length. The width of the furnace is generally five feet, but these dimensions may be extended to eight and even more feet, without inconvenience. b. The quantity of iron re-heated in a good furnace depends on circumstances. Much depends on the character of the mill, and upon the kind of iron we design to produce. A good furnace will produce, in twenty-four hours, from eight to ten tons of iron em- ployed for coarse bars arid hoops, and an equal amount of railroad and other heavy iron. But it will not produce more than from two to four tons of iron designed for small rods, hoops, and wire. Where small iron requiring no welding heat is made from mill bars, large furnaces may be employed. By this means, we may obtain twice the amount just stated, if such an extension is deemed ad- vantageous. By so managing the furnace as to make a heat in the shortest possible time, we shall, in all instances, arrive at the most favorable results. The time required will, of course, vary according to the size of the iron. Still, it is evident that both iron and fuel will be economized in proportion to the shortness of the time consumed in welding a given amount of iron. Another good rule is, to work slowly from the commencement of each heat, to charge the exact amount of fuel required to finish a heat, to keep the temperature as long as possible below a welding heat, and then, after suddenly raising the temperature to that standard, to draw as fast as the rollers will receive the iron. Small charges of iron generally produce a better yield than large charges, but consume more fuel. Where fuel is cheap, and iron expensive, it is better 374 MANUFACTURE OF IRON. to charge only a small amount of iron at a time, and to make a proportionate increase in the number of heats. c. Re-heating furnaces are employed for welding wrought iron scraps. For this purpose, a variation in the height of the arch from the bottom of the furnace, and in the form of the hearth, is required. The height of the arch, in such cases, is generally from eighteen to twenty inches ; and the hearth is somewhat more level than usual. In some establishments, scraps are assorted, and put up in bundles of from forty to fifty pounds each. Due care is taken to have the pieces of iron in each bundle of equal size ; that is, sheet iron and bar iron scraps should not be bound up together. These bundles, well secured by binders, after receiving a welding heat, are either shingled or rolled. This course is also pursued at charcoal forge fires. At these fires the bundles are re-heated singly, and drawn out into bar iron, according to the method commonly practiced. At some places, another method is pursued. This consists in charging the re-heating furnace with loose scraps, applying to them a welding heat, and forming balls in the same manner that they are formed in the puddling furnace. These balls are brought to the T hammer, or squeezer, formed into blooms, and roughened down into bars, as puddled iron. Scraps make a very fine bar iron, particularly in the charcoal fire, and such iron is highly valued by the blacksmith. Where good iron is generally manufac- tured, there is no special demand for scrap iron rods. Iron made from these scraps cannot be cheap ; therefore, there is no advan- tage in seeking to make specific qualities of it. Where scraps can be bought at reasonable prices, the most profitable way of using them is to cut them into small pieces, about the size of one's hand, and to charge the puddling furnace with them. This should be done at the time the iron in the furnace is so far worked as to be nearly ready for balling. From fifty to seventy-five pounds may be thrown in at one time. By this application, the puddled iron, instead of being injured, will be benefited. Thus, large quantities of scraps may be worked to advantage. Wages at the puddling furnace are not only economized, but the excessive waste of iron in the re-heating furnace and the forge fire is obviated. The cinder of the puddling furnace protects the scrap iron. VIII. Heating Ovens. Iron which is sufficiently soft and malleable to be wrought into any shape by the hammer or roller does not require a welding heat. FORGING AND ROLLING. 375 For such iron, a cherry-red heat will suffice. This heat is produced by ovens or stoves. In these ovens, all sheet iron, and rods which require an extra polish, or tempering, are heated. Charcoal billets, from the forge, are also heated in them, to be rolled into rod iron of small size. a. These ovens may be heated by a variety of methods, and with almost any kind of fuel ; still, every caution is requisite to prevent, as far as possible, the access of free oxygen and steam, for both steam and oxygen occasion waste of iron. For these reasons, our statement must be received with some qualification. Wood, turf, and brown coal are, so far as their capacity for generating heat is concerned, an excellent fuel; but, unless they are very dry, the steam generated from their hygroscopic water will oxidize, and thus destroy an amount of iron whose expense will not be counter- balanced by the entire profits derived from the fuel employed. Therefore, instead of using the raw material, it will be found ad- vantageous to use only the charcoal derived from charring it. The same objections which apply against any fuel containing water which of course excludes the use of all kinds of heating, re-heating, and puddling furnaces apply against waste flame, for this contains a large amount of steam, or free oxygen. b. The ancient form of a heating oven was that of a common bake oven, with this difference, that the bottom was formed of iron bars. Upon these bars, placed in the form of a grate, coal from ten to fifteen inches in depth was laid. The iron was placed upon the coal, after the oven was heated. In some establishments, such ovens are still employed. It is evident that a portion of the iron in con- tact with the fuel particularly raw fuel, such as turf, brown coal, and bituminous and anthracite coal must be wasted. The only instance in which such an arrangement may be considered profita- ble is where wood charcoal is employed. Even the best coke or turf coal is not sufficiently pure to guarantee success. Stone coal, coke, and turf are never free from sulphur, and this sulphur will of course combine with the iron. A waste of iron is thus occasioned exactly proportionate to the amount of sulphur the fuel contains. In addition to this, sulphur blackens the metal, which, in the case of sheet, iron and nail plates, gives rise to very disagreeable conse- quences. Fuel burned in this way, even though spread in a high column upon the grate, never combines with all the oxygen which passes through the coal; the result is a waste of iron. Therefore, there is every reason why such ovens are not serviceable. 376 MANUFACTURE OF IRON. c. Heating ovens of a superior kind are at present constructed on the principle of the reverberatory furnace. In these, the fuel and iron are properly separated, and all contact between them obviated. Fig. 123 represents a vertical section of a heating oven for sheet Fig. 123. Heating oven for sheet iron. iron; a is the hearth, b the fire grate, and c the chimney. The height of the furnace is often thirty inches. The object of this is partly to prevent the contact of the flame and iron, but principally to gain room for setting the sheet edgewise; they are thus set on both sides of the furnace; besides, in the middle of the hearth, sufficient room is left for laying a sheet or two flatwise, c? is a cast iron plate, forming a sliding door. The chimney has two flues, the one inside, the other outside of the oven. Its draft is weak, and the smoke or flame frequently issues from the mouth, in which case it is carried off by the second or outside flue. Fig. 124 represents a vertical section across the furnace and the flues ; and Fig. 125 a ground-plan of the furnace, hearth, and fire-place. The cast iron plate e is here shown more distinctly. Its object is to protect the bricks or stones from the destructive agency of the tongs and iron. Like puddling and re-heating furnaces, these ovens are built of fire bricks, enclosed with cast iron plates, and ^preserved from the effects of expansion and contraction by wrought iron cross binders. A slight variation from the form of the oven we have described, occasioned as well by individual taste as by locality, is sometimes FORGING AND ROLLING. 377 observed; still, the one we have presented is the one generally Fig. 124. Front elevation of a heating oven. employed for the manufacture of sheet iron. If it is desirable that the surface of the iron should be kept very clean, the fire bridge Ground-plan of a heating oven. and the inside flue may be raised ; but, in all such cases, pure fuel is our safest reliance. 378 MANUFACTURE OF IRON. IX. Shears, and Turning Machines. These are of much importance in a rolling mill. The first we shall describe somewhat minutely; but a brief description of the latter must suffice. a. When rollers are cast, and ready for turning, they are placed upon a strong and heavy turning lathe, and the gudgeons and couplings turned between points. They are then put into cast iron standards, into which brass pans are inserted. In the latter, the gudgeons revolve. At first, the rollers are turned into smooth cylinders. After a set is thus far completed, the grooves are cut in, according to a design previously drawn on a board. Sheet iron or sheet brass patterns are made for each groove in every roller. These should be preserved, in case a roller is injured, or fails to answer its purpose. Rollers for sheet iron are of course smooth cylinders, but it is not necessary that the bottom and top of the roller should be of the same diameter. Those for thin sheet iron should be turned one upon the other, that their surfaces may be perfectly parallel. Unless there is too great a variation in the sur- faces, this may be done in the housings. After using the rollers for a time, their surface is apt to become rough. Its smoothness may be restored by cutting it with one edge of a square piece of cast steel, from three to four inches in length. This operation is gene- rally performed in the housings, for the moving of the rollers to the turning lathe is attended with great expense. Good hard rollers are turned with difficulty by common methods. A steady turning ma- chine, of slow motion, excellent cast steel chisels, and patience, are the conditions of success. Hard rollers are required for making thin and polished sheet iron. They are polished by means of emery and leaden pans, which extend almost quite around the roller. b. The shears required in a mill are the movable hand-shears, for cutting small rod and hoop iron, and force-shears connected with a waterwheel or steam-engine, for cutting common bar, rough bar, and sheet iron. The first are small lever shears, fastened upon a two inch plank, as represented in Fig. 126. The length of the whole is about two feet or thirty inches. The shears are placed at each end of a pile where small bar or hoops are deposited. The boys, who catch behind the rollers, cut off the bad ends, before a rod or hoop is laid down. c. Fig. 127 represents the common force-shear. It is a power- ful cast iron lever, varying, according to locality and purpose, from FORGING AND ROLLING. 379 seven to twelve feet in length. The excentric a is generally fast- ened upon the main shaft, or, if such is not accessible, upon any Fig. 126. Portable hand-shear. other strong and well-supported shaft. The foundation must be very firm, and not inferior in solidity to the roller trains. The steel Fig. 127. Shear moved by an excentric. blades are made of good shear or cast steel, tightly fitted into the cast iron lever and standard, and screwed on with screw bolts. For the cutting of heavy bar and rough bar, the standard block is generally placed very low, about a foot above ground ; but for cut- ting common bar and sheet iron, it is raised from two feet to thirty inches above ground. If sheet iron is principally brought to the shears, an iron frame #, 5, as high as the lower cutter, is to be fast- ened to the standard. Upon this frame, the sheet is moved. In work- ing sheet iron, shears of this construction are attended with some disadvantage. The acute angle at the points, and the obtuse angle close to the fulcrum which they form, in addition to the difficulty of adjusting them accurately, make them somewhat objectionable. 380 MANUFACTURE OF IRON. To obviate these disadvantages, various plans have been devised, of which the following, Fig. 128, appears to be the most prac- Fig. 128. Shear moved by a crank. ticable. These shears are generally used for cutting nail plates, and for trimming sheet iron. The cutters 6, b are shown in section, and are frequently from sixteen to twenty inches in length, so as to cut over the entire breadth of a sheet ; the same length is required where sheet iron is used for making nails. In case small nail plates are used, shorter cutters can be employed. The lower or fixed cut- ter is horizontal, but the upper is screwed to the cast iron lever in such a manner as to form an angle with the lower cutter seldom greater than fifteen degrees. The motion of the lever can be produced by an excentric, as in Fig. 127, or by a crank, as in the present case. It frequently happens that shears are wanted where we cannot reach directly to an excentric with a lever, nor in a short way with a crank. In such cases, a crank motion from some shaft is con- ducted below ground to the desired point by means of an iron con- necting rod. This arrangement, which may be modified accord- ing to circumstances, is exhibited by Fig. 129. The tail may be turned above or below ground, forward or back ; but care should be taken that the connecting rod is always on the pull side, as shown in the drawing, for a long connecting rod is not adapted to push the lever. This arrangement in which the shears are directly connected with the elementary power is necessary where heavy bar iron and boiler-plate are to be cut, for these require a strong foundation. Portable shears, with their own independent flywheel, propelled by means of a belt and pulley, are preferable for light iron, sheet iron, or nail plates. Bar iron more than two inches square cannot be conveniently cut by shears; like railroad iron, this is to be trimmed by circular saws. FORGING AND ROLLING. Fig. 129. ^mfflllllllM 381 Shear moved by a crank. d. The method of fastening the steel cutters to the castings by means of screw bolts passing through the steel blades, is exposed to several serious objections. In this way, the blades are weakened, and screws and cutters frequently broken. Strong blades would meet the emergency, but cannot be successfully fastened by screw bolts. If designed for heavy work, the cutter should be very strong, and formed of a solid piece of steel, from an inch and a quarter to an inch and a half thick, four inches wide, and eight inches long. By fitting such a piece of steel carefully into the cast iron, and hold- Fig. 130. Screwing in the cutters. ing it by means of steel screw bolts, as represented by Fig. 130, it may be conveniently and securely fastened. 3-82 MANUFACTURE OF IRON. X. Tools. The tools required in a rolling mill are of some importance, and absorb considerable means in their outfit and repair. The principal tools are tongs, which, in an extensive establishment, are very nume- rous. A set of tongs belongs to each re-heating furnace, compris- ing long tongs for the catching of small billets, and tongs for heavy piles. A set is placed before and behind each set of rollers, which are shorter than those at the re-heating furnace. A different kind from the above, from three to four feet long, belong to the heating ovens, and short, narrow tongs are necessary at the sheet rollers. The tongs of a rolling mill, and the manner in which they are kept, show in a great measure the character of the establish- ment. Light, well-made tongs, calculated to answer their different purposes as well as possible, kept in good repair, and in sufficient number, indicate a well-managed mill. Clumsy, heavy tongs roughly made, in bad repair and in insufficient number, are a sure sign that something is wrong in the management of the mill; and, by close investigation, we shall find similar deficiencies in all the other departments of the establishment. To make good and light tongs, sound and strong iron is required. Strong shear steel is the best material for making light and elegant tools. But where the steel is short, it is advisable to pile steel and good charcoal iron alternately. The resulting metal, drawn into bars, will make a kind of Damascus steel, of great cohesiveness, from which light tongs of the most elegant form may be wrought. A number of hammers, sledges, and wrenches are also required. At each re- heating furnace, there must be a water trough, supplied constantly by a current of cold water, for the purpose of cooling the tools. From three to four hooks, five or six feet long, and lighter than those at the puddling furnaces, belong to a re-heating furnace; also a couple of pointed bars for turning the piles. Piles composed of bars are to be turned as soon as the heat of the furnace is suffi- ciently strong to make them adhere. Small rod iron, hoops, and some kind of wire, are put up in bundles of fifty or one hundred pounds' weight each, on a bench constructed for the purpose. The rods of small iron are first weighed; a parcel of the desired weight is then put upon the bench, into iron standards, and there tied together by iron hoops. The bench is a table eighteen or twenty feet long, two or three feet wide, and made of plank three inches thick. Wrought iron FORGING AND ROLLING. 383 stands, in the form of a V, are fastened at such distances from each other as to suit the convenience of the binders, say from two to four feet. To one leg of the stand, a small lever about a foot or a foot and a half long is fastened; it is pressed upon the bundle of iron, forced down, and locked. A binder of small square or round iron is then fastened in that place. These binders are generally tied when heated ; but, in many establishments, they are tied cold around the bundle. Generally three binders, seldom four, are fast- ened to one bundle. Hoops, and small, well-polished rod iron ought to be handled by the workman in gloves, particularly in summer; for clean iron has a fine appearance. XI. General Remarks. a. Where the country is settled and densely populated, there is generally a large demand for iron, and competition and large estab- lishments have so reduced its price that a profitable employment of Catalan fires cannot be expected. But, in less settled coun- tries, where wrought iron for farming purposes is in demand, and where rich ores are accessible, the Catalan forge is profitable from the fact that but a small capital is required to start it. The machinery can be made very cheaply, if a water power is at our disposal; and even a steam-engine costs but little. A hammer 150 or 200 pounds in weight is sufficient to do a great deal of work, at least work for two fires, and will draw, every week, from five to six tons of iron, if the rods are not smaller than horse-shoe bars or wagon-tires. From good ore, the iron made in the Catalan forge is preferable, for agricultural purposes, to any other kind of iron. The expenses of erecting this forge cannot be stated with precision ; but they depend on locality, industry, and the kind of iron wanted. A Catalan forge may be put in operation at* an ex- pense of 500 dollars ; but ten times that amount may be laid out to advantage. As the main body of materials required are timber and stones, its erection, in a new country, where timber is abun- dant, cannot cost a great deal. Hardly any stock of coal or ore is needed ; and the ore and coal worked to-day may be turned into iron and cash to-morrow. b. Very little bar iron is drawn in the charcoal forges, and that which is made is horse-shoe, tire, or heavy bar and pattern iron. For all these purposes, cast iron hammers and anvils, if made of good cast metal, are sufficiently hard and smooth. But if we want to draw fine iron or steel, wrought iron hammers and anvils, 384 MANUFACTURE OF IRON. furnished with well-hardened and polished steel faces, are neces- sary. It is difficult to weld steel to a large piece of iron, with- out spoiling the steel ; it is, therefore, more profitable to insert the steel, in a separate piece, into a groove made in the hammer face, and in the anvil, and to fasten it by means of wedges. Such a piece may be of the best kind of cast steel, and very highly polished, without running the risk of breakage or loss of lustre. The more strokes a hammer makes in a minute, and the lighter the hammer and the more polished the faces, the greater lustre the surface of the rod will show. The exclusion of oxygen from the heating-oven, and water from the anvil, will impart a uniform blue, or dark violet color to the iron. If we want a fine, 'polished surface, and a good color, the iron must be heated in an oven, upon charcoal, by a low heat, with the utmost exclusion possible of atmospheric air; the coals, burning in a very suffocated fire, will form only carbonic oxide; the surface of the iron or steel will thus be prevented from oxidizing. c. In rolling-mills, iron is rolled and drawn to one-fourth of an inch in diameter, whether round, or square. Smaller iron is drawn from the above in the wire mills, by passing the rods through holes made in steel plates. Hoop iron is made as small as five-eighths of an inch wide, and one thirty-sixth of an inch thick. In former pages, we have spoken of the quality of iron required to manufac- ture small iron to advantage; it is only necessary to remark here, that the rollers and re-heating furnaces are to work in such perfect harmony, that the least loss of time in taking out the iron from the furnace must be obviated. The quantity of the iron not only suffers, but its quality also, if it is too long exposed to the influence of the flame in the re-heating furnace. The iron loses, with its car- bon, its strength and lustre ; it looks dull, and is more or less rotten, or cold-short. We would repeat here the remark formerly made, that, if we want the iron to have finely polished surfaces, we must employ hard and well-polished rollers ; for the lustre of the manu- factured article cannot be greater than the polish of the rollers from which it receives its impression. A train of five-inch rollers ought to make, in a week, from twenty to twenty-five tons of small iron, with a loss of not more than ten per cent, of iron if made directly from rough billets, and of not more than seven or eight per cent, if rolled from merchant bars. The amount of coal consumed, whe- ther anthracite or bituminous, does not exceed half a ton to each ton of iron. FORGING AND ROLLING. 385 d. Iron of first rate quality is not required for the manufacture of hoop and small rod iron; any cold-short iron will answer. For particular purposes, such as the making of small chain rods, a fine- grained, yet cold-short, iron is preferable, unless we design the rods to be fibrous. Cold-short iron is to be preferred for small ar- ticles made by the blacksmith, on account of its welding properties. We do not mean to say that an excess of silicon or phosphorus is not hurtful to small iron ; on the contrary, we wish it to be under- stood that such matter is very injurious. Iron containing a small amount of silicon or phosphorus, and sufficient carbon to make it cold-short, is the most profitable for the manufacture of small iron, as well as for blacksmiths' use. The kind of iron needed, in this case, may be considered the connecting link between the quality necessary for making heavy iron and that for making wire. e. Wire iron should be strong, hard, and tough. That which is fibrous, but not strong, is of no use; the wire made from it will be rotten, and without lustre. Where fibrous iron is employed for this purpose, the fibres must be very fine, and the color of the iron white. If the iron is bright, coarse fibres make rough, scaly wire; but if it is fibrous, and of a dull color, the wire will be both rough and rotten. Iron rendered cold-short by silex or phosphorus is not adapted for wire manufacture; but, when rendered cold-short by carbon, it will form, for this purpose, the most advantageous material available. Iron which is hot-short on account of sulphur is very apt to break in drawing. That which is hot-short in consequence of a deficiency of impurities is a still worse article. Very pure iron is so weak that wire cannot be made from it. The strongest kind of wire is made from an iron which contains just sufiicient copper to make it de- cidedly hot-short. Wire iron must be very uniform in its aggregate form; that is, the billets from which wire is to be drawn must have a uniform texture throughout. It must not be a conglomeration of fibrous iron, cast iron, and steel. It requires dexterous workmen to make a good article even from excellent metal. The Catalan forge is not adapted for this purpose, even though the iron should be the product of the best of all materials; it does not work with sufiicient regularity. White coke or anthracite metal will not make wire iron, even with the greatest attention, and in the best charcoal forge. Very nearly the same may be said of hot blast and white charcoal metal smelted from the poorer ores of the coal formation, and bog ores. White metal from the rich magnetic ores of New 25 386 MANUFACTURE OF IRON. York, New Jersey, and Missouri, answers excellently, if carefully worked in a charcoal forge. The best wire iron is made, in the pud- dling furnace, from cold-short gray charcoal pig, or from any gray pig iron, whether charcoal or anthracite, provided the metal is of such a quality that no difficulty results from boiling it. Neither the best nor the inferior qualities of white metal will make good wire iron in the puddling furnace. Iron of fine grain, or fine fibre, of uniform texture, great lustre, and of a somewhat bluish color, and soft as well as very tenacious, is that upon which our principal reliance must be placed. From this, it is evident that wire iron may be most profitably made in puddling works where boiling is well managed. In such establishments, the finest and strongest rough bars may be reserved for wire iron. A selection may be more safely made from billets than flat bars ; and cold-short is to be pre- ferred to fibrous iron. /. Iron for the manufacture of railroad, coarse bar, boiler-plate, and common sheet iron, is to be of a different texture from wire iron, merchant bar, and thin sheet iron. That which is designed for these purposes is exposed to the heat of the re-heating furnace in large piles, blooms, or slabs, to prevent the oxygen of the flame from abstracting the carbon from the interior of the pile so soon as would otherwise be the case. It should be white and fibrous in the rough bars ; otherwise, it will be very cold-short by the time it is trans- formed into the proper shape. Another reason why such iron should be fibrous and white is that cold-short iron, re-heated in heavy piles, is liable to great loss in the furnace. Before the interior of the pile can be heated to a welding heat, the exterior is melted and wasted. The iron most liable to loss in the re-heating furnace is the very best and finest. To avoid this waste, and, at the same time, to secure a perfect heat, heavy piles are composed of different qualities of iron. The top and bottom bars are generally refined mill bars, or a superior quality of rough bars. Mill bars will resist a higher and a more prolonged heat than rough or puddled bars, and are, therefore, commonly employed. A pile thus formed that is, composed inside of iron which can be welded by a lower, and outside of iron which requires a higher heat will answer best with respect to quantity. Of quality we shall speak hereafter. For heavy work, the iron must resist a high heat, and often a heat of long duration. For these reasons, we require white, fibrous iron. Iron which, in consequence of the absence of carbon or other matter, is not cold-short, is advantageous. Such iron requires, and will bear, FORGING AND ROLLING. 387 a higher heat for welding than that which is less white and less fibrous. At the same time, it will yield better, and not waste away so fast. Therefore, in selecting qualities of rough bars for the different purposes of the merchant mill, iron of fine grain or fine fibre must be chosen for the lightest kind of rod iron, such as wire and small rod ; coarser, though not fibrous, iron, for hoops and all kinds of flat iron ; fibrous iron of a dark color for square and round merchant bar ; and white fibrous iron for heavy bar, boiler-plate, and railroad iron. g. The piling of iron for the re-heating furnace is, as far as quality and quantity are concerned, an object of the utmost im- portance, and too much attention cannot be paid to it. In a well- managed rolling mill, every rough bar is tested, and classified ac- cording to its texture. All the iron may be equally good, the grained equal to the fibrous ; but the texture of the rough bars has an influence upon the quality and quantity of the article made from them. These bars must be straight, to admit of close packing, and they are to be sound and smooth on all sides, to offer a com- pact and close surface to the influence of the flame. Cold-short is more easily welded than fibrous iron ; and dark iron will adhere together at a lower temperature than white iron. This, of course, is to be applied to iron made from the same metal, and in the same forge. Accordingly, the puddled iron, or rough bars, are to be classified into grained or cold-short, dull or dark fibrous, and bright or white fibrous iron. The second is the worst class ; the first is designed for small, and the latter for heavy iron. These classes ought to be kept separate, and mixed at the re-heating furnace ac- cording to the kind of pile made, or the commercial iron which is to be manufactured from it. It is easily understood that, on the piling of the rough bars, suc- cess in a great measure depends. If it should happen that cold- short or dark iron were put at the outside, or bottom, or top of a pile, it is very clear that the outside would be melted and wasted before the inside could be welded. The same happens if, acci- dentally, all the bad iron is on one side of the pile, and all the good on the other. This matter ought never to be left to accident, but it should be regulated with reference to the quality of the iron. Heavy piles of rough bars for re-heating are to be composed, in the centre, of the weakest iron ; at the bottom, of that next in quality ; and on the top, of the best. Or, if there are but two classes, then the weaker at the bottom, and the stronger on the top. Where bar 388 MANUFACTURE OF IRON. iron, weighing from 50 to 150 pounds, is to be made, it is most advantageous to mix the metal alternately, that is, to change with each single bar, so that not more than one bar of each class is joined by that of the next class. A uniform mixture will thus be made, and the best iron produced which can be made from the quality in question. If we reflect upon the difference which exists in iron of the same quality, but of different form of aggregation, with respect to welding at a high or a low heat ; that cold-short iron, made from the same metal, will waste away, while the fibrous iron is not yet sufficiently hot for welding; if we further reflect upon the accidental running together of one or the other class of iron, in the same pile or the same heat, irregularly, which cannot be avoided where the rough bars are not tested and classified, we must not wonder when we see one pile yield ninety-five per cent., while the other makes but eighty ; or when we see the splitting of blooms or the falling off of pieces in the roughing rollers. If each pile, as we have stated, is thus composed of a systematical range and quality of rough bars, the yield is uniform in the single piles, and of course better in the whole. There is then no splitting of piles, or waste in the rollers. This is always true, whether the manufac- tured iron be of the worst or the best kind, the weakest or the strongest. All iron, if uniform, no matter what its quality may be, yields better and works better than it would under other circum- stances. A pile of dull, weak, fibrous iron will be wasted away in a temperature scarcely strong enough to heat a good charcoal bloom sufficient for drawing, much less for welding. How imprudent would it be to put such different kinds of iron together in the same heat ! We can very easily conceive the result of an attempt to weld a flat rough bar of weak puddled iron to a bloom of good white charcoal iron, without making the experiment. The weak iron will, of course, be wasted before a heat sufficient for welding is raised. It is unprofitable to weld weak iron to strong iron. It is a generally known fact that we can unite iron of the same texture, whether weak or strong, with the smallest possible loss. These remarks we deem sufficient to awaken more general attention to the classification of rough bars than is paid to it at present. Where iron is puddled from white or refined plate iron, the difference is not so striking as in establishments where gray pig is boiled. The difference is most apparent where boiling and puddling are carried on in the same mill, as at Pittsburgh, and at almost all the Western puddling es- tablishments. FORGING AND ROLLING. 389 h. An object of considerable importance is the manufacture of sheet iron; this, as well as bar iron, is generally made in the same establishment. So long as charcoal iron is used in the manufacture of sheet iron, which is commonly the case, it does not make much difference where the latter is made. But, if it is to be manufac- tured from puddled iron, it is of some consequence to use iron of a particular quality. A fine, fibrous, tenacious iron, of a bluish color, though not too white, is required. Hot-short iron is, of all kinds, the very worst, because it splits, and gets porous and scaly in the progress of the manipulation. Cold-short is not good for heavy, but answers well for very light sheet iron, such as stove pipe ; it makes a smooth and polished surface. Here, as well as in the manufac- ture of bar iron, the iron is to be classified according to the purposes for which it is designed. Classification is, in fact, of more im- portance in the one case than in the other. The price of sheet iron justifies the application of better pig metal, and more careful and expensive work in the forge, than would be prudent with bar iron. The price of the latter is, generally, so low, that a scrupulous ad- herence to quality is scarcely admissible. Badly worked iron, whe- ther from the charcoal forge or the puddling furnace, is not adapted for the manufacture of sheet iron. For these, and many other rea- sons, sheet and bar iron ought not to be made in the same estab- lishment, and from the same kind of iron. That which is most profitable for bar iron will make neither good nor cheap sheet iron : and that which makes the best and cheapest sheet iron is too expensive to be used for making bar iron. Blooms for sheet iron should be of the purest and best kind of iron, and of that which is most free from sulphur, silex, and phosphorus, and rendered cold- short by nothing else than carbon. For light sheet iron, cold-short iron is preferable, because it works easily, and takes a brilliant and durable polish. It does not oxidize to the same extent as a more pure and fibrous iron, and for this reason takes a more uniform color, and does not scale. On light sheet iron, we always look for a rich, bluish color and high polish. Well-polished, hard rollers and clean iron secure any degree of polish; but, to succeed well in this case, a particular kind of iron is required, which separates easily from the scales, or hammer-slag. The purest iron is that made cold-short by combination with a little carbon. White, fibrous iron is apt to form thick and heavy scales of hammer-slag, which ad- here strongly to the iron, and are deeply impressed in its surface. These scales are removed with difficulty ; acids are our only reli- 390 MANUFACTURE OF IRON. ance, because many small particles of hammer-slag are so deeply squeezed into the iron, that nothing short of solution can remove them. After their removal, such sheet iron looks rugged and un- even. Good iron, of a steel-like grain, if properly treated in the heating oven, does not form these thick and uneven scales. To make heavy sheet from cold-short iron is an unsafe policy, for, if we leave the quality of the former out of the question, its quantity is unfavorable. Cold-short, or weak iron, if rolled from heavy slabs or piles into boiler-plate, is very apt to split or open in the centre; this is a great loss, for, in most cases, the whole sheet is turned into scraps. Besides the loss caused in the rollers, we always find the central part of such plates very weak, scarcely better than cast iron, though the edges may be good. Heavy sheet iron and boiler-plate should be made from a coarse, fibrous, white iron, which will stand a high heat in welding. Boiled iron, and iron from hot blast worked in the charcoal forge, are unprofitable. Such iron will either spoil the sheet, or make it of such bad quality that it would be dangerous to put it in steam boilers, and thus run the risk of destroying human life and property by explosion. Heavy sheet iron should be made from blooms which are manufac- tured, in the charcoal forge or the puddling furnace, from white, and the very best kind of plate iron ; or, if not made from plate iron, the blooms must be rolled into flat bars, piled, re-heated, and then rolled into flat mill bars, to be again re-heated ; this is repeated until the iron assumes a strong, fibrous texture and a bright color. If the latter cannot be produced, it is necessary to reject the iron altogether, at least for the manufacture of boiler-plate. This method of refining iron, which is frequently practiced in England, is too expensive in this country; and, in addition to being expensive, it is ill adapted to make iron of proper quality, that is, strong in every direction. It will be strong in the longitudinal direction of the fibre and bars, but very weak in the transverse, that is, across the fibres. Coarse, weak, fibrous iron is to be considered in the same light with respect to the making of sheet iron, as to the mak- ing of wire. The sheet from such material will be porous, scaly, and weak, and will not answer the purposes for which it is designed. In the erection of works for the manufacture of sheet iron, too much attention cannot be paid to the solidity and strength of the machinery, and to a surplus of power. Success depends as much on an unfailing power and well-constructed machinery as on the well-trained intellect which manages the daily operations. FORGING AND ROLLING. 391 . Nails are an important article of commerce, and an item in our factories which must be regarded with due attention. For these, good iron is unnecessary ; they require neither strength nor the quality of welding, which are two requisites of iron the most dif- ficult, at least the most expensive, to produce. Nail iron must be malleable, that is, in one direction, and as cheap as possible. Sheet iron, which is now most generally employed in the making of nail plates, is ill adapted to make a cheap, and, at the same time, a good nail. So long as good charcoal iron is used, the quality of sheet iron is secured: but, if an inferior quality of charcoal blooms* or puddled iron, is employed, the form of the sheet iron does not answer so well; the nails will be cold-short, though made from an article which, for the manufacture of bar iron, would be highly useful. In New England, nail plates from four to six inches wide, and from ten to thirty feet long, are made. These are preferable to sheet iron; that is, where the quality of iron is the same, a far better nail can be made from these plates than from broad plates or sheet iron. As previously stated, we are enabled to make fibrous and very malleable iron from almost any kind of pig iron, whatever its quality may be, provided there is no necessity of making the iron v,ery cohesive, or of giving it the property of welding in the black- smith's forge. Such iron is made in the puddling furnace by a very low heat, and it must be re-heated at the lowest possible temperature; otherwise, it will lose its fibres and malleability, and become cold-short. It cannot be transformed into sheet iron at all ; but it will make a hoop very malleable lengthwise, though not transversely to the fibres. It cannot resist a long-continued red heat, which is frequently applied at the nail machine^, with- out altering its fibrous into a cold-short texture. Such iron, if well worked, is generally very soft; it may be cut when cold, and then tempered so as to give color after the cutting of the nails. This is a proposition, however, not based upon practice^ and requires confirmation. Iron of this kind can be made from pig iron, no matter how bad its quality may be. Puddling does not at all improve its purity; it only alters its texture from a granulated into a fibrous aggregation. Of all kinds of iron it is the cheap- est, for it is worked fast, with but little loss, and little fuel ; no skill is required to manufacture it. Industrious work and the lowest possible heat are the best means of success in puddling. Such iron is of but little use for other purposes, but we verily believe it will make a nail far superior to most of the nails at present in market. 392 MANUFACTURE OF IRON. To make nails from white, coarse, fibrous iron, however strong it may be, is unprofitable, for the nails will split, and cut badly. Such iron is of too good a quality, and is better adapted for making coarse bar or heavy sheet iron. Whatever may be the kind of iron used for making nails, it is always better to draw it into long and small nail plates; these are to be cut into strips crosswise, so that the nail can be cut parallel with the length of the plate and fibres. The piles for making nail plates are to be put together with due regard to the production of the most perfect fibres. All cross piling is to be avoided ; and if cold-short iron is to be worked at all, it must be mixed regularly in alternate courses with fibrous iron. Such piles may be very heavy ; the greater the number of cuttings of rough bar, the better will be the result. The rolling of these piles is to be performed in such a manner as to make the welding joints parallel with the surface of the nail plate. In the re-heating fur- nace, the lowest heat commensurate with the performance of the operations is the most profitable. Any heat relatively too high will transform most kinds of fibrous iron, particularly this, into cold-short iron, or iron of a crystaline texture. In the technical management of a rolling mill, we cannot pay too much attention to the classification of the puddled bars, and the composition of the piles, before they enter the re-heating furnace. BLAST MACHINES. 393 CHAPTER VI. BLAST MACHINES. THE principles involved in the construction and application of blast machines are based rather upon the chemical effect which a strong peculiar draught produces in burning fuel, than on any me- chanical or chemieo-physical effect. The latter effect merely in- creases the consumption of fuel in a given space, and increases the heat in that space to a limited degree ; but the former causes a union of the oxygen in the blast with the fuel or carbon in a manner more or less favorable to the reviving of iron from the ore, and the pro- tection of the iron against oxidation. The means to effect a favorable result in the application of fuel for the purpose of augmenting temperature, as in puddling and re- heating furnaces, and heating stoves, are various. This result can be accomplished by the simple application of chimneys, or of blast, or by using both together. Fuel is most perfectly used where it is oxidized in the highest degree; this oxidation takes place in the re-heating furnace, where, generally, all the hydrogen is converted into water, and all the carbon into carbonic acid. "We cannot say the same of any other apparatus, for we generally find a mixture of carbonic oxide, carbonic acid, and free oxygen, which is an evi- dence of imperfect combustion. Increased draught, or the concen- tration of more heat in the fire chamber, will lessen such an evil; but there is frequently a deficiency of draught in cases in which heat is necessary, as in that of the puddling furnace. If, in such instances, it is impossible to produce sufficient heat by the draught of the chimney, we are compelled to make use of blast machines. This is the case with anthracite coal and coke. How chimneys act in producing draught, and what are the rules to be applied in con- structing them, are matters which require scientific demonstra- tion not included in our investigations. We have described the practical workings and dimensions of apparatus, which may be deemed sufficient for all practical purposes. An explanation as to 394 MANUFACTURE OF IRON. the chemical effect of blast under different pressures, we shall give at the close of the chapter. There are many forms of blast machines; but in our own country we are very fortunately reduced to the most simple and practical. We shall notice some blast machines in operation in Europe, which are frequently recommended by writers on metallurgy, principally for the purpose of showing their imperfections. The most simple blast machine is the smith's bellows, a description of which it is unnecessary to give. I. Wooden Bellows of the Common Form. A kind of blast machine, called Widholm's bellows, is very ex- tensively used in Sweden, Russia, Germany, and France. We do not know that any are employed in the United States. As it works well, as the expense of its construction is small, and its applica- tion to the Catalan forge very simple, we shall furnish a drawing and description of it. Fig. 131 shows it in section; a is the Fig. 131. Swedish bellows. movable part or piston ; b an iron rod connected with a crank of the waterwheel, or the steam-engine ; c, c are the valves, and d the nozzle. The latter is fastened to the permanent top /, which is again fastened to some wooden framework. The whole has the ap- pearance of a common smith's bellows, with the only difference that it is made entirely of wood. From ten to twelve strokes may be made in a minute, and two bellows are required for one fire. The whole is from six to seven feet long and thirty inches wide the piston having a motion of twelve inches. This kind of machine is applied to no other apparatus than the charcoal forge, and we allude to it merely because it is simple and cheap, fulfilling its purpose excellently. BLAST MACHINES. 395 II. Wooden Cylinder Bellows. These are of various forms. We have seen square cylinders and round ones : the piston playing from the top, or from below ; or the piston working in both directions. There are vertical and horizon- tal cylinders, and machines working with one, two, or three cylin- ders, with a dry receiver, water receiver, or with no receiver. In our own country, we are almost entirely confined to one prin- cipal form, that is, the machine with two round tubs or bellows the piston working from below, and a dry receiver placed on the top of the tubs. This may be considered the best form of the wooden blast machine, if but a single stroke is desired. Fig. 132 is a re- presentation of a blast machine of this kind; a, a are the bellows; Fig. 132. Wooden cylinder bellows. b the receiver from which the sheet iron pipe c leads the blast to the furnace; d, d are the pistons, moved alternately by the beam 396 MANUFACTURE OF IRON. /, which is set in motion by the crank and wheel e. The wheel may be moved either by a waterwheel or a steam-engine. From eight to ten strokes is generally the speed required to supply a charcoal furnace. The tubs or cylinders, as well as the receiver, are generally from four to four and a half feet wide, and four feet high, making the stroke of the piston three feet. To the piston g, in the receiver, an iron rod is fastened, playing in a stuffing-box at the bottom, which carries a box h filled with iron or stones, to counterbalance the pressure of the blast, and to regulate it by play- ing up and down as the pressure from the tubs increases or di- minishes. The valves are made of wood, lined with leather. The beam is generally laid below, and the tubs raised a few feet above, ground. The whole machine is made of dry, well-seasoned wood the cylinders glued : that is, composed of small segments of dry pine or ash, an inch or an inch and a half thick ; the woody fibre thus runs around the cylinder, i. e., horizontally instead of verti- cally. This construction of the tubs secures greater permanency to their form. Their interior is, in some instances, covered with a thin coating of a mixture of glue and plumbago, which gives it the appearance of iron, diminishes the friction, and secures a closer fit of the piston. This kind of blast machine works admirably, if properly con- structed ; it is very durable. In every respect, this apparatus is preferable to the wooden bellows of the common form, such as that represented by Fig. 131. It can be erected at an expense of from $250 to $350. It will work one blast furnace for charcoal, or from four to five forge, or Catalan fires. A steam-engine or waterwheel of from twelve to sixteen horse power is required to put it in operation, and furnish the necessary blast for a blast furnace. a. There are double working wooden tubs also in use, but not very frequently. These, in particular cases, may be of advantage ; in cases, for instance, where room or expense is to be saved, or where wooden are very shortly to be replaced by iron cylinders. The wooden tubs are but a temporary arrangement, to gain time and means after the works are just started. The double working tubs, that is, those which make blast at each motion, like iron cylinders, offer no real advantages over the single ; in fact, in ordi- nary cases, the tub with single stroke is preferable to the double tub. Among the advantages of the former, is the facility with which we can attend to the interior ; in case damage is done to the BLAST MACHINES. 397 surface of the tub, it can be instantly mended. This is not v the case with double stroke cylinders ; here the top and bottom are closed, and the interior is not accessible without stopping the blast machine, and the operations which depend upon it. For these reasons, tubs which open at the top are preferable to those which open from below. The principal objection against wooden cylin- ders is that they are frequently severely rubbed by the packing of the piston ; this diminishes the pressure of the blast in consequence of the leaking between the piston and tub. The disadvantages resulting from single stroke tubs, open from below, are more than counterbalanced by the greater simplicity of the piston rod, and the facility with which the valves can be adjusted. A stuffing-box is required, which, if the tubs are to be opened from -above, must be made of iron. The expense of erecting a solid and strong frame to carry the crank and beam, is also comparatively great. b. A good mechanic, and a thinking one, is required to construct a wooden blast machine. To put the wood well together is not sufficient ; it is necessary to select it with due relation to its liability to twist, warp, and crack. All curly, knotty wood, and wood from the heart of the tree, must be rejected. The circumference of the tree, or both seams of the heart plank alone, are to be used for the tubs and receiver. The tops as well as the tubs are generally three inches thick. The latter are glued together from segments one foot or more in length, and not more than one and a half inch thick, as before stated. The tops and pistons are composed of strips of plank not more than three or four inches wide, grooved and feathered, and well glued. Ash may be considered the best wood for making the tubs ; but good dry pine will answer. Other kinds of wood, such as maple and walnut, are too apt to warp, and therefore ought not to be used. To keep the interior slippery and sound, the sur- face of the tub is frequently brushed over with plumbago, or soap- stone powder, or with a mixture of both. These ingredients are moistened with water, to which a little glue may be added. Fat or oil is an improper material with which to lubricate the surface of a wooden tub, for both are very soon destroyed; the destruction of the piston and the wood of the cylinder then follows, to the injury of the machine, and the loss of blast. Square tubs, and horizontal tubs of double stroke, have been tried ; but, it appears, with no good advantage, for nobody now thinks of such forms. It is unnecessary to speak of these machines 398 MANUFACTURE OF IRON. in this place, as they belong to antiquity, and are, at the present time, of no practical importance. III. Iron Cylinder Blast Machines . a. There are various forms of these machines. The smallest, but not the most simple, apparatus, is a double stroke cylinder that is, composed of two beams and two cylinders which is frequently met with at the Western establishments. In rolling mills, it is used to blow the finery ; we find it also at blast furnaces. Fig. 133 Fig. 133. Iron cylinder bellows. exhibits it so plainly, that a particular description of it is unne- cessary. This machine makes an excellent blast. Its cost is the main objection to its use ; this objection is valid, as far as the first outlay is concerned; but its expensiveness is counterbalanced by the excellent manner in which it works. It does not make quite a regu- lar blast, if worked without a receiver ; but even in this case, it may be made to work better frhan others differently constructed. In this machine, the cylinders, pistons, pipes, valves, wheels, and cranks are all of iron, except the beams and pitmans, which are of wood; but BLAST MACHINES. 399 the latter would be better if made also of iron. This machine is constructed on an excellent principle, and is superior to the hori- zontal cylinder, very much used in the Eastern States. This is finding its way to the Western States, which does not augur well for the speedy and successful application of stone coal in the blast furnace of that section of our country. b. The desire of constructing a cheap apparatus has led to the making of an iron cylinder blast machine with a horizontal instead of a vertical motion of the piston, as shown in Fig. 134. There is no doubt that such a machine is far cheaper than one of vertical stroke; but, when we consider the difficulty of keeping the packing tight, and the loss of blast which thence ensues, and the frequent disturb- ances which originate from very hard rubbing of the piston on one part of the cylinder alone, it may be doubted whether it should be considered a useful apparatus; in fact, experience rather bears against than confirms its utility. Fig. 134 exhibits such a cylinder. Fig. 134. Horizontal cylinder blast machine. The piston rod runs through both heads, to carry the weight of the piston, and prevent its rubbing with all its strength on the lower part of the cylinder. The valves are generally made of sheet iron lined with leather. Machines of this construction have their ad- vantages, besides the great simplicity in their entire arrangement which they afford. There is no difficulty in procuring a solid foundation for the whole. The weight of the piston, piston rod, and pitman, which is objectionable in vertical machines propelled by a waterwheel, particularly in those where but one cylinder is em- ployed, is, in this case, almost balanced. The crank and a small portion of the pitman form the only weight which is not equipoised. The application of the valves is very simple, and very correct. They must be suspended vertically in a good blast machine. 400 MANUFACTURE OF IRON. The foregoing are the two leading arrangements involved in the construction of blast machines; both have their advantages and disadvantages. Where the propelling power is a waterwheel, and where it is contemplated to use but one or two cylinders, the hori- zontal cylinder may be considered to present many advantages; for, in such cases, it is always more or less troublesome to balance the weight of the piston, rod, &c. If the motive power is steam, the vertical position of the cylinder is decidedly preferable ; for, in this case, the weight of the piston and its accompaniments can be balanced by the weight of the steam piston and its associated parts. A machine consisting of one blast cylinder and a receiver may here be considered the most simple and advantageous. Where a water- wheel is the propeller, it is less advantageous to employ a single cylinder, for, as the stroke is made by a crank, a great irregularity in the blast ensues, and a comparatively large receiver is there- fore required to regulate the inequalities of the pressure. This is one of those instances in which a crank works to the disadvantage of the power applied, which is seldom the case. For these reasons, various forms of blast machines, propelled by waterwheels, have been tried. In this country, however, only those with two cylinders and double stroke are used. This makes a useful, but not an ex- cellent blast, even though the cranks work at right angles to each other. A receiver is almost indispensable, in this case, to equalize the blast, and to make the best possible use of the water power. Blast machines with three cylinders and double stroke have been applied; and this arrangement may be considered the most advan- tageous where water is the moving power. Such a machine pro- duces a very steady blast, without a receiver, and gives the best effect of the waterwheel. c. There is no reason whatever for employing water power in the propelling of blast machines at blast furnaces. There is abund- ance of waste heat for the generation of steam. The expense of erecting a steam-engine will be found less, in most cases, than that incurred in the erection of a waterwheel. For these reasons, we shall not dwell any longer upon the application of water power to blast machines, and shall confine our subsequent remarks to those propelled by steam alone. IV. General Remarks on Cylinder Blast Machines. There is no doubt that the application of one cylinder to a blast machine is accompanied with great advantages. Such an arrange- BLAST MACHINES. 401 ment is in conformity with sound principles of mechanics, because, by this means, the least friction commensurate with the same effect is produced. Weight and surface, the two most important causes of friction, are very greatly reduced. If the blast cylinder is on one end of a balance beam, and the steam cylinder on the other, the regularity of the blast is much greater. But this is no reason why a balance beam should be applied : because any inequality in the pressure of the blast can be regulated by applying a large re- ceiver. If, therefore, it is found advantageous to abandon the balance beam, and still to retain the vertical position of the cylin- ders, the unbalanced weight of the pistons and piston rods is no obstacle. A piston rod, to connect blast and steam cylinders, has been applied where horizontal cylinders have been used in cases, however, in which the blunder of making the piston rod too short was committed. The hot part of the piston rod, playing in the steam cylinder, is thus cooled when in the blast cylinder, and the adherent oil dried when playing there. By this means, the hemp of the stuffing box of the former is very soon worn out. Besides this disadvantage, the close proximity of the steam apparatus and the blast cylinder is very injurious to the operations in the blast fur- nace. It is impossible to keep the steam out of the blast cylinder, if the latter is too close to the steam cylinder or the steam boilers, or even if it is in a very warm place. We know that moisture in- troduced into the hearth of a blast furnace is very injurious. a. The size of a blast cylinder depends partly on the amount of air needed, and the number of strokes made, and partly upon the purposes for which it is designed. A charcoal forge requires from 400 to 500 cubic feet per minute ; a finery from 800 to 1000 ; a charcoal furnace from 1000 to 2000 ; and an anthracite or coke furnace from 3000 to 5000. The number of strokes that can be made by a machine depends chiefly on the length of the stroke and the construction of the valves. In cylinders of four feet diameter, the piston can move with the speed of three feet ; in smaller cylin- ders with greater, and in larger ones with less speed. If the motion is regulated by a flywheel and crank, more speed can be given than where a flywheel is not employed. b. The size, form, and weight of the valves have a highly im- portant influence upon the speed of the piston, loss of power, and quality of blast. The smaller the valves are made, the greater is the increase in the velocity of the air which is to pass through them. 26 402 MANUFACTURE OF IRON. Friction of the air and valves, besides a direct loss of pressure and air in proportion to the pressure in the valve, is thus occasioned. One-twelfth of the surface of the piston is sufficient for the passage of the blast ; but no disadvantage results if the valves are larger. The form of the latter has an influence upon the effect of the ma- chine. Trap valves are the most practicable. Semicircular valves, with the hinges in the diameter, deserve to be more extensively em- ployed than they are at the present time. The semicircle has less outline in proportion to the same surface than the square or paral- lelogram, the usual form of valves, and for this reason diminishes the friction of the air. Quadrilateral valves are seldom used. In general, the oblong shape is preferred, in which case, the hinges are put to one of the longest sides. It is obvious, from reasons which will be subsequently given, that the longer the valve, the more perfect will be its form. The weight of the valve is an important object, for, if neglected, it may seriously injure the effect of a blast machine. It is easily understood that this weight may be so in- creased, that the effect of a blast apparatus amounts scarcely to anything. The weight of the valve causes an expansion of the air in suction ; consequently, the pressure on the suction side of the piston in proportion to this weight will be less than that of the atmo- sphere. A loss of power and blast on the compressing side of the piston, proportional to the weight of the valve, is also occasioned. The air which remains in the dead space of the cylinder is of greater pressure than that in the blast pipes or receiver, in the ratio of the weight of the valve. To diminish the influence of this weight, the valves are generally placed in a vertical position, and are made en- tirely of a light material, such as wood and leather ; they are also made as oblong as circumstances will admit. In this respect, the horizontal blast cylinder possesses great advantages. The location of the valves is best secured by vertical heads ; and if the friction, or rather the weight, of the piston and piston rod could be balanced, the horizontal cylinder would be the best form of the blast machine. Their position and weight, also, have considerable influence upon the effect of a blast machine ; but of still more consequence is the dead space left at the heads of a cylinder. Dead space is that which is not filled by the piston head, in its alternate motions, and from which the air that is compressed is not forced by the piston. In the best blast machines, the loss which this occasions amounts to at least ten per cent., and in some cylinders is as great as twenty-five per cent. The loss in power and blast increases with the size of the BLAST MACHINES. 403 dead space. In this respect, the horizontal has the advantage over the vertical cylinder.