4~~~ I i \ k i ~ r I 4. THE MANUF ACTURE OF STEEL, BY M. L. ( RtUNER, Professor of Metallurgy in the School of Mines, Paris, and Inspector- General of Mines. TRANSLATED FROM THE FRENCH BY LENOX SMITH, A.M. E.M., WITH AN APPENDIX ON TIHE BESSEMER PROCESS IN THE UNITED STATES, BY THE TRANSLATOR. ILLUSTRA TED B Y LITHO GRAPHED DRA WINVGS AND WOODCCU"& NEW YORK: D. VAN NOSTRAND, PUBLISHER, 23 MURRAY ST., AND 27 WARREN ST. 1872. Entered according to act of Congress Itl the year 1872, by D. VAN NOSTRAND, In the Office of the Librarian of Congress, at Washington. LANG, LrMTLE & HI.LLUAN, BLBECTWRTYPERM. AND TERUBOTYPRBE, 108 to 114 WoosTER ST., N. Y. CONTENTS. PAGR Translator's Preface................................................... 5 Nature of Steel....................................................... 7 Hardness and tenacity of Steel dependent on Carbon.................. 21 Various methods for the manufacture of Steel......................... 29 I. Fining Pig Iron not melted, producing malleable Cast Iron of a steely nature................................................... 31 II. Fining molten Pig Iron, giving solid steely products.............. 34 III. Fining molten Pig Iron. giving molten steely products............ 37 Bessemer Process in France.......................................... 39 Bessemer Process in England...................................... 49 New rolling appliances used in England.......................... 54 Bessemer Steel in Sweden........................................ 57 Bessemer Steel in Austria.............................................. 59 Bessemer Process in Belgiulm, Prussia, Russia, Italy, etc............ 71 New mode of assay in Bessemer fining used in Germany......... 73 Defects of' the Bessemer Process-means of remedying them....... 74 Berard's Process.............................................. 79 Fining by reaction............................................... 82 Experiments in manufacturing Cast Steel in the Reverberatory Furnace.................................................... 87 The Martin Process at Sireuil.................................... 94 Manufacture of refined Pig Iron.................................. 112 IV. Manufacture of Steel by Cementation............................. 117 Ordinary Cementation............................................ 118 Theory of Cementation........................................... 123 4 CO1VTENTS. PAGEB Cementation and simultaneous fusion in crucibles................. 127 Cementation.and fusion in a Cupola Furnace'(Parry Process)...... 130 Determination of the Carbon in Steels by the Eggertz Method.... 134 Determination of Sulphur by the Eggertz Method................ 141 Additional Note................................................. 146 Explanation of Plates0.............................................. 147 APPENDIX. The Bessemer Process in the United States............... 156 LIST OF ILLUSTRATIONS. PLATE I. Ramsbottom's Roller oscillating by means of connecting rods. II. Ramsbottom's Roller oscillating by means of racks. III. Plan of the Forging Press of Mr. Haswell of Vienna. IV. Forging Press of Mr. Haswell of Vienna-Bessemer's Forging Press. V. IRamsbottom's Double Hammer. VI. Berard's Steel Furnace-Heath's Steel Furnace-Ramsbottoni's Direct-Acting Double Hammer. VII. American Five-ton Bessemer Plant. VIII. Standard English Plant. IX. Standard American Plant TRANSLATOR'S PREFACE. THE constantly increasing importance of the already considerable steel interests of the United States has induced the translator to present this volume to the public, in the hope that the want of a careful, elaborate, and at the same time practical, investigation of the physical properties of steel, as well as of a description of the new processes and mechanical appliances for its manufacture, may thus be supplied. A familiarity with the condition of the steel interests of other countries is of essential importance to those practically engaged in its manufacture who desire to increase their stock of technical knowledge, but who have not access to scientific periodicals, or sufficient time at their disposal to enable them to master their contents in a foreign language. The American manufacturer cannot well afford to disregard information from reliable sources relating to iron and steel, if he desires to successfully encounter foreign competition; and the fact that this work is from the pen of Professor GrUner, who is well known in Europe as one of the most able metallurgical 6 TRANSLATOR'S PREFACE. authors of the present day, should of itself insure for the work a favorable reception. From lack of accurate information, Professor Gruiner has made but little mention of the extent of the Bessemer Process in the United States, and the translator has therefore appended a short memoir upon that subject. NEw YRnK, February, 1872. THE MANUFACTURE OF STEEL. SINCE the publication of Leplay's two excellent memoirs upon Yorkshire and the North of Europe (Ann. des Mines, 1843 and 1846), steel manufacture has undergone considerable modification. Mr. Lan and myself have shown its progress as lately as 1861.* Since that time steel manufacture, far from diminishing, has yearly become more largely developed. At present it has attained colossal proportions. We are now witnessing a complete renovation of the old processes. The Exposition of the Champ de Mars has proven the importance of the products and their great variety, but the methods could not be studied there, although they should especially be understood. It is this motive which has induced me to describe the actual condition of steel manufacture, or, at least, to point out its most salient characteristics. NATURE OF STEEL. What is steel? This point has been much discussed for a considerable time, but no conclusion has been *Etat present de la metallurgy du fer en Angleterre, 1862, part fifth; Manufactare of Steel, p. 711. 8 THEN MA.NUFACTURE OF STEEL. reached, because the meaning of the word was undefined. The province of steel is sometimes enlarged, sometimes unreasonably circumscribed. A certain theoretical idea is taken as a starting point, and all within the limit prescribed is denominated steel. Experience proves that cast and wrought iron can be obtained with any ore of iron, but that the cast and wrought iron produced have various properties because of the greater or less purity of the ores, and because the methods of treatment do not all effect the elimination of foreign substances to an equal degree. According to circumstances, a metal will be obtained more or less tenacious or brittle, hard or soft, pure or impure. But none the less will the names cast and wroughllt iron be given to the extremes produced. For the same reason we should call every intermediate product, which is neither wrought nor cast iron, steel, whatever its degree of purity may be. The crude cast product resulting from the reduction of the ores of iron may be called pigr iron. It is an impure iron which is not malleable, at least when heated, but which may be tempered bS being suddenly cooled. The name wrought iron is given to a metal more or less refined, obtained from pig iron, or in the direct way from iron ores; it is malleable hot and cold, but not capable of being tempered. The practical man will call every intermediate product steel, which may be tempered, but which remains malleable hot and cold when not tempered, and this TIE fMANUFACTURE OF STEEL. 9 metal will be steel, whatever be the method followed to obtain it-direct extraction from the ore, partial finin(r, of pig iron, or recarburation of wrought iron. According to this, in its properties as well as in its manufacture, steel is comprised between the limits of cast and wrouoght iron. It cannot even be said where steel begins or ends. It is a series which begins with the most impure black pig iron, and ends with the softest and purest wrought iron.* Cast iron passes into hard steel in becoming malleable (natural steel for wire-mills, the TVildstaMl of the Germans), and steel, properlS so called, passes into iron, oivingr in succession mild steel, steel of the natureof iron, steely iron, and granular iron. And these transitions are observed, not only when we compare the properties of the products and their mode of manufacture, but also their chemical composition. Doubtless it is very difficult, even impossible, to prescribe exact limits to the:: composition of cast irons, steels, and wrought irons. The elements are so various, and often in such slight proportion, that, in presence of the enormous preponderance of iron, their exact determination becomes impossible. ButD the result of the analyses made, in every case, is that the: same foreign elements are found in pig irons, steels, and wrought irons, and that, after all, that which distinguishes. the three products is solely, as shown long since by Karsten, the relative proportions of carbon, of which a portion is simply mingled with the iron, the remainder *Karsten stated this in these same terms as long ago as 1823. It should. never have been forgotten. (Annrales des Mines, 1824 Vol. IX., p. 657.) 10 THE ffANUFACTURE OF STEEL. intimately combined with it, or rather in a state of solution. We find it strange that sonime tenths of a per cent. of carbon can modify iron so far as to cause it to pass into steel. Rivot in his Docimnasie appears to admit that the two substances are chemically identical, and only differ from each other in their molecular constitution-a constitution pre-existent in the ores in such a manner, that certain ores are in some sort predestined to produce steel.* Such is the case with those which have been long known under the name of Ores of Steel. Without wishing to discuss at present this ingenious hypothesis, I must observe that these same steel ores yield excellent irons; to obtain them it is only necessary to prolong the fining. These irons again give steel of the most superior quality, when surrounded with charcoal and heated. It appears, then, most natural to suppose, and this seems to me a result of the facts I am about to recall, that modifications so various as iron undergoes are due to the presence of carbon and divers other foreign substances. The simple changes of density, texture and tenacity, through which all metals pass, must, of course, be excepted when their crystalline condition is favored or destroyed by heat and mechanical action. The modifications which most metals undergo through the influence of very small proportions of foreign substances have long been known, and it is not apparent why it should be otherwise for wrought iron. It is a' Rivot, Docimasie, Vol. III., p. 545. THE MAITAFACTURE OF STEEL. ]li general property of metals which we cannot explain, but which must be received as a fact. We know that copper: is greatly modified by traces of oxygen, sulphur, or lead; that some tenths of a per cent. of iron affect the malleabil-: ity of zinc and tin. And on this head Fremy calls to our: attention that 0.01 per cent. of bismuth or lead renders& gold as brittle as antimony.* We also know that chromium, nickel, tungsten, titanium, etc., render wrought iron hard as well as carbon. In short, the considerable. influence of. these very small proportions of foreign substances cannot be denied. Now,, these substances which we find combined with iron, are very much more numerous in pig irons than is generally believed. Fremy mentions carbon, silicon, phosphorus, arsenic, sulphur, potassium, sodium, calcium, magnesium, aluminium, manganese, nickel, chromium, titanium, vanadium, copper, nitrogen, etc.,t and the analyses which I shall quote will contain, in addition, some other elements, such as cobalt, molybdenum, etc. It may be said, in general, that pig irons, like all crude metals, contain a fraction of almost all the elements found in the charges which produced them. Fremy has thought proper lately to attach very great importance to the presence of nitrogen. According to this savant, steel and the purest pig irons are combinations, not of carbon and iron alone, but nitro-carburets, and it is this complex radical which imparts to steel its special properties. Fremy afterward became aware that wrought irons* Comptes rendus, Vol. LII., p. 1004. t Comptesrendus, Vol. LII., p..1001. TEHE MAANUFACTURE OF STEEL. also contained nitrogen. In consequence, the radical nitro-carburet, if it exists, would not be a distinctive characteristic of steels. The series-pig iron, steel, wrought' iron-would, nevertheless, exist. But nitrogen does not, appear to play in these compounds the part attributed to it- by Fremy. The proportion of nitrogen is very much less than was believed at first to exist. Boussingault, to -whom we owe the most exact analyses on this subject, found, in a first series of assays, 0.057 per cent. of nitrogen in cast steel, 0.124 " " " wrought iron,* 0.007 " " " "only" cast steel, and the same proportion in irons, and 0.022 per cent. in Krupp's steel.t' It follows from this that no fixed relation exists be-, tween the proportions of nitrogen and carbon, consequently they would not be united under the form of a. special radical. We even see that in steel, properly speaking, the proportion of nitrogen does not reach onete~nth that of the carbon, while this relation is always greater in wrought irons. The presence of nitrogen ap. pears then; somewhat accidental. It appears to be owing to the property, so common to solid bodies, of condensing gas in its pores. At all events, it is now thoroughly' established that nitrogen is only met with in steel in very small proportions, and if this gas is necessary to: the existence of steel, it is still more necessary to wrought iron and pig metal. * Comptes rondus, Vol, LII., p. 1251. t Comptes rendus, Vol. LIII., p., 9. THE.MANUFACTURE OF STEEL. is Among the other foreign.elements, there are several which are not nearly as unobjectionable. Besides this, the various properties of -the irons of commerce cannot really be explained, except by the presence of these elements. Their number is always considerable, and when the energetic reactions in the working of blast furnaces are taken into account, it becomes evident that even the purest pig irons are quite complex compounds. This is the result of the several analyses which I think proper to quote. It has been long known that pig irons contain carbon and silicon, and, usually, some sulphur, phosphorus, and manganese. Karsten, in his Metallurgy, mentions also calcium, magnesium, and chromium, but does not venture to state anything regarding aluminium.* The frequent occurrence of this latter body was first noticed by Schafhaiitl, about 1840; at present the fact is well established, for we even find aluminium in pig metals obtained from charges containing but little alumina. The tri-silicate slags of the cannon foundry of? Finspong,, in Sweden, only contain 3 per cent. of alumina,, and these cast irons are of the following composition," according to Eckmann and Eggertz: t Metallic Iron............. 93.660 Aluminium..... 0.173 Manganese.............. 0.190 Calcium and Magnesium... traces * French edition, 1830, Vol. I., p. 252. { Pamphlet published upon Finspong on the occasion of the Universal Ex,. position in 1867. 14..THE MANUFACTURE OF STEEL. Copper................. 0.005 Silicon.............. 0.946 Phosphorus...... 0.050 Sulphur........ 0.120 Carbon... 3.920 of wbicl 2.17 is graphite, 99.064 According to the annals of the Bureau of Iron, of Stockholm, Durocher had already published, in 1856, several analyses of these same pig irons from Finspong, all of which mention some aluminium, calcium, and magnesium.* A Swedish pig iron was exhibited at the Exposition, containing, according to Eggertz, 0.26 per cent. of calcium and 0.16 per cent. of aluminium.t If the numerous analyses published by Bertbier do not show any of these earthy metals in pig irons it is because no importance is attached to them, and that even the search for these elements was purposely neglected. But it is evident that, by smelting any ore containing, alumina, a pig iron more or less rich in aluminium would be obtained. Under this head may be quoted the gray pig irons produced from our granular ore, and the black pig irons of Scotland, smelted from argillaceous black band ores, in a very hot working of the blast furnace. *Annales des Mines, 3d series, Vol. IX, p. 420. Durocher was wrong, in thinking that the conclusion could be drawn from one of these analyses that sulphur increases tile tenacity of gray irons, and that irons containing from 1 per cent. to 2 per cent. of sulphur, are made purposely at Finspong. They really seek to reduce the proportion to less than 0.2 per cent. (Page 11 of. Rinman's memoir above quoted.) t Some information concerning the wrought irons and steels of Sweden, by L. Rinman, 1867, p. 25. THE MIANUFACTURE OF STEEL. 15 The presence of so great a number of foreign elements renders a complete analysis of pig irons one of the most complicated operations of inorganic chemnistry, and notwithstanding this, the search for any of these substances cannot be omitted under the pretext that it exerts no influence upon the quality of the irons and steels, resultinrg from finingt these pig metals. The last volume of the Archives de Karsten (vol. xxv, p. 235, 1853), mentions two analyses which suffice to prove the complicated composition of pig irons. The irons are from Vekerhagen and Holtzhausen, in Hesse Cassel. They were smelted With hot blast and charcoal from Tertiary limonites. The first answered for foundry purposes; it was very fluid, filled the moulds well, but whitened in thin layers, and then became brittle. Its specific gravity was only 6.668. The second, that of Holtzhausen, was slightly mottled, softer, more tenacious than the former, but like it of a dull shade in fresh fractures, which, as is known, denotes a great proportion of foreign substances. For the same reason its specific gravity was also low, beingr only 6.799. Both irons sustained great waste in fining. Solution was effected for the principal analysis by a mixture of hydrochloric acid and chlorate of potash, and the residue was separately: examined. The results are as follows: 16 THE MIANUFACTURE OF SIEEL. VEKERIIAGIN PIG. HOLTZHAUSEN PIG. Total Carbon.............. 2.876 2.215 Silicon................... 2.748 1.981 Sulphur................... 0.207 0.083 Arlsenic................... 0.139 traces Phospllhorus............... 0.421 0.068 Calcium.................. traces 0.352 Magnesium............... 0.146 0.895 Aluminium.............. 0.032 0.272 lMolybdenum.............. 0.184 0.014 Chromiul................ 0.080 0.051 Vanadium............ 0.004 traces M[anganese................ 8.953 2.814 Ietallic Iron.... 83.383 30.718 99.173 99.463 In both irons the insoluble residue afforded the greater part of the manganese and silicon, viz.: Iron from Vekerhagen iSilicon.............. 2.668 Manganese...... 6.871 Iron from Holtzhausen Silicon.1.345 Manganese.......... 2.814 The insoluble residue of the latter contained in addi? tion 0.167 per cent. of aluminium, and 0.133 per cent. of metallic iron. It is apparent that in gray pig irons, rich in manganese, the silicon seems to be chiefly united with the latter substance. And we know, in fact, from the investigations of Brunner and W3hller, that manganese combines readily with 11-12 per cent. of silicon by fusion. These analyses prove that gray pig irons often contain above 10 per cent. of foreign substances, and that almost always their number is considerable. It is certainly true of most black irons possessing little tenacity, which result THE iMANUFACTU' RE OF STEEL. 1I from a very hot working of the blast furnace. But even white pig irons, obtained from spathic ores, classed generally among the pure products, have also a very complex composition. Below are two analyses by the skillful chemist Fresenius. The first is of spiegel pig iron (Spiegeleisen) from Miisen, smelted from the ores of Stahlberg, in Siegen,* with charcoal. The specific gravity of the metal varies from 7.60 to 7.66. Total Carbon.................... 4.323 Silicon............................. 0.997 Nitrogen....................... 0.014 Sulphur....... 0.014 Phosphorus...... 0.059 Arsenic........................... 0.007 Antimony................... 0.004 Sodium and Lithium............. traces Potassium.......................... 0.063 Calcium............... 0.091 Magnesium................. 0.045 Titanium....... 0.006 Aluminium...................... 0.077 Copper....... 0.066 Cobalt....... traces Nickel.......... 0.016 Manganese........................ 10.707 Metallic Iron..............82.860 Intermiingled Slags..0.665 Total........ 100.014 The second analysis is of a coke spiegel pig iron,. * Etat acuel de la mtallurgie du fer dans le pays de Siegen, by Jordan, p. 36. 2 18 THE MAN2UFACTURE OF STEEL. smelted in 1864 at the works of St. Louis (MNarseilles). from a mixture of iron and manganese ores. The analysis was made at the request of Mr. Jordan, then the manager of the works in question, and we are indebted to him for communicating the results obtained. The iron was white, but showed some gray specks: Combined Carbon................... 4.040 Graphite................ 0.126 Silicon............................. 0.584 Sulphur................. 0.035 Phosphorus...... 0.090 Arsenic.................... 0.032 Antimony.................. 0.026 Magnesium.................. 0.058 Aluminium.................... 0.068 Copper.................. 0.046 Manganese............,............ 5.920 Metallic Iron...... 88.781 Total........... 99.806 Traces of calcium, cobalt, nickel, and nitrogen were found besides. Among the purest pig irons we may mention that of Eisenertz in Styria, smelted from spathic ores, with charcoal in a cold working of the blast furnace. The white cavernous pig metal obtained under these conditions is composed, according to Dr. Percy,* of Combined Carbon..................... 3.79 Silicon............ 0.34 Sulphur................... 0.02 * Percy's Mietallurgy, Iron and Steel, p. 536. TEE MANUFACTURE OF STEEL. 19 Phosphorus................ 0.07 Manganese....................... 1.06 Calcium.......................... 0.05 Magnesium.................. 0.02 Metallic Iron........................ 94.57 Total.......99.92 It is apparent, even in the case of a slow reduction, that the iron may retain some silicon and some tenths of a per cent. of the earthy metals. All things considered, then, the preceding analyses, even admitting that they do not represent the exact composition of the irons examined, nevertheless support the general conclusion stated above, that crude wrought iron retains a part of all the elements present together in the blast furnace. But now, what becomes of these substances in working for steel? If the steel be obtained byfining, it is evident a priori, and analysis proves it, that only those elements can be entirely eliminated which are at the same time readily oxydized and have a slight affinity for iron. Of this number are manganese, calcium, magnesium, etc. On the other hand, aluminium, which is not readily oxydized, according to Deville, and which is intimately combined with the iron, must be more difficult to remove completely, and, in fact, certain cast steels contain traces of this substance. The greater part of ordinary steels retain, besides, some sulphur, phosphorus, and silicon. As to the copper in pig irons, it is found in greater part in the steels and wrought irons. It is sufficient to recall the 20 THE AilNUArlCTUJPE OF STEEL. researches of Lan in the Rivois process,* and those of Bromeis on the iron of Magdesprung in the Hartz.t The good natural steel of Siegen always contains, in like manner, 0.1-0.16 per cent. of copper, according to Stengel, and 0.2-0.4 per cent. of silicon.4 In a KIrupp cast steel for cannons, obtained probably by simple fusion of a mixture of pig and wrought iron from Siegen in a crucible, Mr. Abel, Manager of the Chemical Office of the London War Department, found the following elements:~ Carbon................. 1.18 Silicon................. 0.33 Phosphorus................. 0.02 Sulphur................. 0.00 Cobalt and Nickel................... 0.12 Copper............................. 0.30 Manganese........................ traces Metallic Iron.................. 98.05 Total........................ 100.00 In Indian Wootz Henry found some silicon, sulphur, and arsenic. Philipps, the engineer of mines, has proved the existence of silicon, cobalt, and nickel in all the cast steels mnade by Trinquet at St. Etienne. The greater part were obtained by melting the iron from the catalan forges of the Pyrenees after cementation; some others by combining pure pig and wrought iron in crucibles.ll It follows * Annales des Mines, Vol. XV., p. 103, 5th series. t Rammelsberg's Mltallurgie Chimique, p. 161. $ Archives de Karsten, Vol. IX. and X. ~ The Artisan, December, 1856, and Percy's Metallurgy. I Annales des Mines, 1848, Vol. XIV., p. 326. THE MANUFACTURE OF STEEL. 21 from this that the wrought iron must contain nickel and cobalt before the cementation. Finally, according to Parry, the skillful chemist of that vast establishment, the puddled steel of Ebbw Vale contains: * Carbon...........................0.501 Silicon...........0.106 Sulphur.............................0.002 Phosphorus.........................0.096 Manganese.........................0.144 Metallic Iron...................... 99.151 Total................ 100.000 We have just shown that wrought iron, like steel, quite frequently contains copper, cobalt, and nickel, and on the other hand, according to Karsten, it has long been kniown that there is carbon, silicon, sulphur, and phosphorus in all wrought iron. In reality steel and wrought iron are compounds almost as complex as pigr irons. The proportions only are less. We find, with very few exceptions, traces at least of all the elements of whichl pig irons are composed, and we believe that each of these elements must modify to a greater or less extent the especial properties of pig iron, wrought iron, and steel. Now, among these elements there is one which should especially occupy our attention-carbon. It is, in fact, easy to show, by examining the Bessemer steels made in Sweden and Austria, that hardness and the susceptibility of being tempered depend essentially upon the proportion of carbon held in solution. * Percy's MIetallurgy, p. 797. 22 THE MANUFA CTURE OF STEEL. In Sweden nine grades of Bessemer* steel are distinguished according to their hardness, estimated by the tempering power. They are desi(gnated by the numbers 1, 1, 2, 2,...... as high as 5, in passing from the hardest to the softest, and, at the works of Siljanfors, analysis has shown very nearly the following proportions of carbon: No. I....................... 2.00 % Carbon. 1.....1.75 " 21.......................1.5 X" " 2.....................1.50 %; " 3................1.00%.0 32...0.............. 75 " "1 4. nc>r0.50 41 0.25 " " 5............ 0.05 No. 1. Links white pig metal with the hardest steel; it may, with difficulty, be forged, and does not weld. No. 1I. Forges tolerably well, but does not weld. No. 2. Forges well, but does not weld. No. 2A. Forges well and commences to weld, although with difficulty. No. 3. Forfges very well, and may be welded in the hands of a skillful workman. It is hard steel. No. 3A. Forges very well and welds readily. It is ordinary steel. No. 4. Both forges and welds readily. It is soft steel. No. 41. Forges and welds perfectly, but has slight tempering power. It is hard, or granular iron. * Boman's German memoir, Sur le procede Bessemer en Suede. THE MANUFACTURE OF STEEL. 23 No. 5. Forges and welds perfectly, but has no tempering power. It is cast wrought iron, or homogeneous metal. It must now be remarked, that the proportions of carbon which I have quoted, have no absolute value as regards the tempering-power and the facility with which the irons may be welded or drawn out. The purer a steel is, the greater may be its proportion of carbon without destroying its welding and forging power. The greater part of foreign substances, excepting special metals, such as tungsten, titanium, nickel, etc., render steel short and destroy its welding power when the carbon exceeds a certain limit. Let us also observe that, all things being equal, the limit in question is lower according as the foreign substances are more abundant. This being the case, Bessemer steels made in France, and especially in England, must in general contain less carbon than Swedish steels. They are, in fact, made from pig iron of less purity; and we know that Bessemer steel, or rather Bessemer metal, in England rarely contains carbon enough to admit of being tempered. When the proportion of carbon is increased the product becomes more or less short. In Austria, as in Sweden, where very pure pi(g irons are treated in the Bessemer process, superior products are likewise obtained. Tunner, the eminent metallurgist of Leoben, has adopted a system of classification slightly differing from the Swedish. I[e has, however, omitted the two first Swedish numbers, which belong rather to 24 TTE il4N XfCFA TCTURE OF STEEL. white pig iron, and has replaced tlie half numbers by entire numbers increasing from 1 to 7. According to him the theoretical classification is as follows:No. 1. Containing 1.5 per cent. of carbon, is malleable but yet not weldable steel. It corresponds to No. 2 of the Swedish scale. No. 2. Containing 1.25 per cent. of carbon, is malleable steel, but difficult to weld. No. 3. Containing 1.00 per cent. of carbon, is very malleable steel, which may be welded by a skillfiul workman. It is hard steel. No. 4. Containing 0.75 per cent. of carbon, is very malleable steel, easily welded. It is ordinary steel. No. 5. Containing 0.50 per cent. of carbon, is, at the same time, very malleable and very easily welded. It is mild steel. No. 6. Containing 0.25 per cent..of carbon. It is granular iron, which is tempered with difficulty. No. 7. Containing 0.05 per cent. of carbon, is homogeneous iron, which cannot be tempered. When the fining is carried too far, a softer metal than No. 7 is obtained. It is short, without tenacity. It is the burnt iron of the blacksmiths; according to Fremy, an oxydized and not a nitrogenized iron. The theoretical classification has been verified at the works of Heft, in Carinthia. * Oestreildischc Zeitschriftfar Berg und tiiuttenwesen, year 1865, p. 153. THE.MANUFACTURE OF STEEL. 25 Below are given the proportions of carbon obtained by analysis: ACCORDING TO TUNNER'S ACCORDING TO NOTES TAKEN MEMOIR ABOVE CITED. AT THE EXPOSITION.* No. 2......... 1.3.5 per cent. No. 3......... 1.15" ".. 1.00 to 1.10 per cent. No. 4......... 0.85 " " 0.75 to 0.85 " " No. 5......... 0.72 " " 0.42 " " No. 6......... 0.53 " " 0.25 " " No. 7........ " " * A similar series of Bessemer steel from Fagersta, in Sweden, containing fiom 0.10 to 1.30 per cent. of carbon, was also exhibited at the Exposition. At the imperial works at Neuberg, in Styria, the proportions of carbon, according to a statement published at the time of the Exposition, are the following: NA MBDNERS. PROPORTIONS OF CARBON. OBSERVATIONS. No. 1... 1.58 to 1.3S per cent. Cannot be welded, and is rarely used. No. 2... 1.38 " 1.12 " No. 3... 1.12 " 0.88 " "Welds easily, used for bitts, chisels, etc. No. 4... 0.88 " 0.62 " U" sed for cutting-tools, files, etc. No. 5... 0.62 " 0.8.... Mild steel for tyres, etc. No. 6... 0.38 " 0.15 " " Tempers slightly; steel for boiler-plate and axles. No. 7... 0.15 " 0.05 " " Does not temper. Steel for pieces of macelinery. These results show that 0.25 per cent. of carbon, more or less, suffice to make steel pass from one grade to another. They fully confirm the earlier theory, accordingr to which, all things being equal, the grade of a steel is otherwise nearly relative to the proportion of carbon. I say, all things being otherwise equal, for I must repeat that other elements may increase or retard the facility with which the metal welds, increase or diminish its hardness after tempering, and render it more or less short or tenacious, so that the proportions above set forth have really only a relative value. Notwithstanding 2G4 THE MANUFA CTURE OF STEEL. this, we may certainly say that the qualities of steel depend chiefly upon the proportion of carbon. In Sweden and Austria, a very great importance is justly attached to the classification of Bessemer steels. If the new product is to be commercially accepted, the hard steels for cutting-instruments (Nos. 2 and 3 of the Austrian scale), should be carefully distinguished from medium steels for springs, pieces of machinery, and tyres (Nos. 4 and 5), and these again should not be confounded with the extra soft steel (homogeneous irons), used for sheet-iron, axles, gun-barrels, etc. (Nos. 6 and 7). In both these countries, each bar is marked with the number of the class to which it belongs, before delivxering it to the consumer. It would certainly be of utility, if a like custom prevailed in English and French works. In the Austrian establishments, Nos. I and 7 are rarely made, the first, because its extreme hardness has a tendency to make it short; the latter, because of its diminished tenacity. The tenacity varies little* between lNos. 2 and 6, but the elongation which the bar undergoes before breakingr is greater as the proportion of carbon is less. In the Swedish section of the Exposition -were a number of cylindrical bars tested in this way. On the side of each bar the breaking weightwas marked, the relation between * According to the already quoted note concerning Neuberg, the absolute tenacity would seem to decrease regularly with the proportion of carbon. This is not perfectly confirmed, however, by the observations made in Sweden and elsewhere. Still, steels containing little carbon are, generally, the least tenacious. THE MANUFACTURE OF STEEL. 27 the sections of rupture and the original section, and the proportion of carbon in the specimen tested. The proportions appear to have been determined by the Eggertz method, which I shall explain at the end of this treatise. As this method appears to me to possess all desirable accuracy, I think it proper to quote the figures. The steels are from the works of Fagersta, near Ncuberg. A steel containing 1.2 per cent. of carbon broke without elongation. The softer steels gave the following results: PROPORTION BREAKING WEIGHT, RELATION BETWEEN OF TIlE ELONGATED AND CARBON. Per square millimetre. Per square inch. ORIGINAL SECTIONS. 1.()0 per cent. 9)5-109 kilo(g, 135131-155042 lbs. 0.91-0,95 0.70 " 8;-102 1" 122330-145105.5 " 0.80-0.90 0.45' 90-103' 128021-146505.6" 0.68 0.35 " 138.8 " 190334 " 0.36 A still softer steel, a kind of homogeneous iron, containing( from 0.1 to 0.3 per cent. of carbon, is employed at Fagersta for gun-barrels. With a bursting charge the metal expands and tears without flyingr to pieces. The same establishment exhibited an extra hard steel containing 1.50 per cent. of carbon, with which steel containing 1.00 per cent. of carbon may be bored, even when tempered. At Neuberg, steel No. 3 elongates 5.00 per cent. before rupture. No. 4 elongates 5.00 to 10.00 per cent., No. 5, 10.00 to 20.00 per cent., No. 6, 20.00 to 25.00 per cent., and No. 7, 20.00 to 30.00 per cent. Let us also state that, according to Vickers of Shef 28 THE IMANUFACTU'RE OF STEEL. field, the tenacity of steels diminishes as soon as the proportion of carbon exceeds 1.25 per cent. (Journal de Le'oben, by Tunner, Vol. XV., p. 300.) In resume, then, it results from all which precedes that the irons, steels, and pig metals of commerce are analogous compounds of iron and carbon always combined with other foreign substances; that the peculiar properties of these different irons are especially dependent upon their puiity and the proportions of carbon in combination; that, in fact, a continuous chain links soft irons containing a minimum of carbon with steels, and the latter again with pig irons. It has long been known that all the carbon in white irons and tempered steels is really combined, or held in solution, while in gray irons and steels not tempered, a portion of the carbon remains separated in the form of graphite. Caron has recently demonstrated that hammering steels hot produces the same effect as tempering, and therefrom concludes that hammering, like tempering, causes a combination of the carbon and iron.* I should rather judge that hammnering, like tempering, prevents the separation of the two substances already combined. It is. known from the phenomena which take place in blast furnaces and in cementation furnaces, that iron holds all the more carbon in solution, when the temperature is more elevated and longer sustained, while by slow cooling the excess of carbon is again separated. But in * Conzptes rendus, Vol. LVI., p. 46 and 212. METHODS OF MANUFACTURE. 29 soft bodies a certain time is necessary for the movement of the molecules. If, then, the cooling is sudden, the carbon cannot separate; a sort of supersaturation is produced. Hammering produces the same effect, but rather, in my opinion, because it prevents the independent crystallization of the iron and carbon, by bringing together, and continuously intermingling, the molecules of the two bodies. This proves that tempering and hammering cannot act in the same manner; that hammering, as we know, increases the density, while tempering diminishes it. Caron has, in fact, discovered this very curious circumstance, that a bar of steel becomes shorter under successive temperings, while it increases in size, laterally, in such a manner, that there actually is an increase of volume. METHODS OF MANUFACTURE. Steel may be obtained by two methods: on the one hand, by direct fining, in the same manner as wrought iron; on the other, by recarburation of this wrought iron. With a given pig iron, the latter method will, of necessitv, give a purer product than the former. The foreign bodies will be more completely removed by continuing the fining till wrought iron is reached. The process, however, will be more expensive, because the cost of recarburation is added to that of fining. It is nevertheless applied in two very different cases, which correspond exactly to the two extreme instances in which the Extra 30 THE MANUFACTURE OF STEEL. Process is employed in the treatment of copper in Enoland. In the first place, we resort to it when we wish to produce very superior steel from pig metals of the first quality. Thus the Dannemora pig iron is converted into wrought iron, which is then cemented in England for superior steel. In the second place, we employ it when we wish to obtain common steel from ordinary pig metals, which, owingr to the excess of foreign substances, can be fined only for wrought iron. Such are the English pig irons, smelted from the ores of the coal formation, which are converted into cast steel for rails, by the application to puddled iron of the method of recarburation, invented by Parry, of Ebbw Vale, or, more simply, by fusion in crucibles, with the addition of charcoal or pure pig iron. But before occupying ourselves with the indirect method, let us first see in what direct fining consists. In fining pig metal for iron or steel, we may pursue three different courses. First: We may operate upon solid pig iron at a more or less elevated temperature, but without fusion. When the decarburation is complete, the product is ordinary nalleable cast iron, and malleable steely cast iron (Glfthstahl of the Germans) when the fining is partial. Second: We may operate upon softened or fluid pig iron so as to obtain a solid product which, according to the degree of decarburation, will be a wrought iron or natural steel; that is to say, the wrought iron, or steel of the I METHODS OF MAXUFACTURE. 31 low hearth and the puddled iron or steel of the reverberatory. Instead of finingv the pig iron proper, the sponges obtained directly from the ore by the Catalan and Chenot methods may also be treated in the same manner. Third: Finally, we may operate upon molten pig iron at a temperature so high that the product remains fluid. This result is obtained by the so-called processes of reaction and Bessemer. According to the degree of the decarburation, ordinary cast steel or cast wrought iron will be pr6duced; the latter is called of late wrought iron or homnogeneous metal. This name appears to have been adopted, according, to Percy, in the first instance, by Joseph Bennett Howell, of Sheffield, in his patent of 1856. Finally, we may decarburize still less, and obtain a product intermediate between steel and cast iron. It is the.Feineisen, or Reineisen, and the Hartguss of the Germans. I will adopt the term re fitedpig iron for these mixed products. Let us glance in review at these different processes, pausing at those which present any new peculiarities. DIRECT FINING WITHOUT FUSION. Direct fining without fusion consists in slow oxydation of the carbon of the pig iron by roasting, or by some solid oxydizing agents. The latter especially are used, and this decarburation in closed vessels by oxydizing cementation we owe to Reaurnur, who recommends softening 32 THE MAANUFLACTURE OF STEEL. pig iron by heating it enveloped in burnt iron, then called safran de mars.* This is evidently the least perfect of the three methods, for, if the foreign elements could all be oxydized, only those whose oxyds are volatile could really be eliminated-carbon, sulphur, and arsenic. It can only be applied then to very pure pigt irons, and as graphite is difficult to oxydize, they must be white, or rendered so by recasting. Charcoal irons, made from the red hematites in the north of England, are chiefly used. White pig irons smelted from spathic ores may also be employed, but it is necessary that they be freed in great measure from manganese, which renders iron and steel brittle, as Breant t long since proved, and Caron: recently confirmed, by rernelting (a sort of Styrian refining). The following analysis by Dr. Miller, quoted by Percy,~ proves that the combined carbon is oxydized almost exclusively. COMPOSITION OF THE PIG IRON BEFORE FINING. COMAPOSITION OFTHE MALLEABLE CASTIRON. Combined Carbon.........................2.217 % 0.434 % Grapbite...................................0.583 % 0.446 % Silicon......................................0:951 % 0.409 % Snlphur....................................0.015 % 0.000 % Aluminium and Phosphorus.................. taces traces We see that the sulphur is completely eliminated, and it would appear that a portion of the silicon is likewise * Reaumur. Art defabriquer l'acier, 1722, p. 472. t Annales des mines, 1st series, 1824, Vol. IX., p. 325,. t Comptes rendus, Vol. LVI,. p. 828. ~ Percy's Metallurgy, Vol. II., p. 111. METHODS OF MANUF-ACTURE. 33 removed, but the silicon which has disappeared ought certainly to exist in the softened product as silica or silicate of iron. Tunner also now affirms that the silicon is partly eliminated, but I avow that I have some difficulty in believing( the fact possible if there is an actual elimination, and not simply a partial oxydation of the silicon.* I will again quote the following analysis of a malleable cast iron made at St. Etienne. We are indebted for it to Philipps,-r the engineer of mines. Carbon......................0.54 per cent. Silicon......................0.44 " " Sulphur, etc., were not sought for. Malleable cast iron may be forged, but is a product possessing little tenacity and density, and in the presence of 0.40 to 0.50 per cent. of silicon, as shown in the two analyses just quoted, this is not surprising..Steely malleable cast iron differs from malleable cast iron proper only in the higher proportion of carbon whicih remains in it. It requires a pig iron of still greater purity, a properly fined white iron. The metal to which we now turn our attention, the Gtlthstahl of the Germans, has been manufactured chiefly by De Mayr of Leoben (Styria)'. It can be forged, and files have been made of it, but the product is nevertheless short. Owing to the low price of Bessemer steel, Mr. De Mayr has lately discontinued the * Stabeisin und Stahlbereitung, Vol. II., p. 49. t Annales des mines, 1848, Vol. XIV., p. 327. 3 34 THIE MANUFACTURE OF STEEL. manufacture, and it may be said that the method of fining by oxydizingD cementation has no longer any reason to exist. II. FINING FLUID PIG IRON WITH SOLID FINED PRODUCTS. This is the usual method of fining for wrought iron and steel. Iron and forge steel are obtained in the low hearth with charcoal. In the reverberatory, iron and puddled steel are made. These are familiar modes of treatment. I (lo not propose to describe them again; I will only remark, that owing to the increasing cost of charcoal, the low price of wvrought iron, and the perfection to which puddling has been brought, working in the low hearth is becoming more rare, except in countries like the Alps, Ural mountains, and Sweden, where the ores are very pure and wood still abundant. Everywhere else puddling is gradually replacinr it, but the reverberatory, as well as the low hearth, gives impure products. In both cases blooms are produced, from which all the scoria cannot be expelled. They are sponges steeped with silicates. A solution of continuity exists between the particles of iron, and, consequently, a lack of cohesion and tenacity more apparent as the blooms are larger.* * The exceptionally good quality of wrought iron removed by attachment in certain German forges is owin to the smallness of the samples and the high temperature, which favors the fusion of the scoriaceous particles. MIETHODS OF MANUFACTURE. 35 The small weight of the blooms is a grave defect in the method in question. Large pieces of wrought iron are only to be had by uniting together a number of bars by welding. Now at the centre of the piles this welding is always imperfect. Another difficulty is consequent upon this. When a large piece of wrought iron is kept hot without being drawn out, it tends to assume a crystalline state. We are aware of the semi-fluidity of soft solid bodies from Tresca's experiments. The molecules of iron are mobile at this temperature, and may be grouped in regular crystals. This is what takes place at the centre of every large mass which cools slowly. We cannot control this tendency to crystallization except by sudden chilling, but the temper which the exterior of the mass receives must be destroyed by gentle reheating. Bombproof plates are prepared in this manner, and all large forgings should be thus treated. In this respect the Exposition was remarkable; specimens weighing 15,000-30,000 kilogrammes (33,000 — 66,000 pounds), from Petin, Gaudet & Co., Marel Brothers, Lacombe & Russery, Creusot, etc., were exhibited. As regards the quality of wrought irons and the weight of the masses handled, the working of puddled iron has made immense progress. But a fundamental defect still exists, and is inherent in the mnethod itself. We cannot have tenacious and pure wrought irons and steels, a really homogeneous mass, when the product fined is not liquid and cannot be cast in the form of an ingot. Now, this is 36 THE MAXNUFACTURE OF STEEL. the especial characteristic of the third method, and which makes it so valuable. It is the future of iron metallurgy, and the incentive which leads us to turn our particular attention to it. But first, let us mention some exp-eriments, whose results have not been, and could not be, successful. Several metallurgists have thought that instead of smelting ores in a blast furnace, it would be better to simply reduce them to the condition of soft or carburized sponge. They hoped to obtain purer products and consume less fuel by operating at a lower temperature. They were completely deceived. When the sponges are made, instead of cast iron we have blooms of less purity, since they contain, besides the usual cinder, the earthy substances in the ore. And if the sponges are melted in crucibles, instead of forging them directly in the form of blooms, we shall have a homogeneous product, but it will be iron or crude steel of inferior quality unless the iron sponge undergoes fining like pig metal. This was done with the Chenot sponges at the works of Baracaldo in Spain, near Bilbao. But they thought, besides, to reduce the ore by a simple mixture of combustible gases, and did not think that if the oxyds of carbon and hydrogen would reduce the oxyd of iron, the carbonic acid and steam would oxydize the metallic iron, so that, to produce wrought iron by gases alone it is necessary to use a great excess of oxyd of carbon and hydrogen, or introduce solid carbon, as in blast furnaces, .METHODS OF MANUFA CTURE. 37 in order to effect the constant reduction of the gases oxydized by the action of the oxyd of iron. It is well to remember, on this subject, that, according to Debray, peroxyd of iron cannot be reduce-l to the condition of metallic iron with less than four equivalents of hydrogen for one of steam, while, if the proportion of lhydrogen is less, it only returns to the state of black protox3-d. It is the'same as a mixture of CO + CO; to obtain wrought iron there must be more than one equivalent of carbonic oxyvd per equivalent of carbonic acid.* In the direct methods whose object is the abolition of blast furnaces, the addition of coal mixed with the ore cannot be avoided; and it is this which destroys all profit in the processes invented by Chenot in France, Renton in America, Gurlt in Germany, etc. III. FINING FLUID PIG IRON WITH FILUID FINED PRODUCTS. Pig iron is fined at a temperature sufficiently high to obtain cast steel or wrou(lght iron, called homoyeneous metal, as its purified products. The scorias, being( fluid, separate completely from the metallic product, as in the blast furnace. Homogeneous ingots result, whllich, for that reason, are much more tenacious than the blooms obtained by the preceding method. This is the advan* Comtaes rendus, Vol. XLV., p. 1018. 38 THE MAANUFACTURE OF STEEL. tage of the processes remaining to be treated of, and the only secret of their superiority. This incontestable superiority also contains the germ of a complete transformation in the working of iron. This third method of fining comprises several processes. The most remarkable, and, at present at least, the most widely known, is the Bessemer process. As differing from it we may quote the Berard process. The fining is essentially effected, in both cases, by the oxygen of the air. But we may also fine by means of solid agents, such as iron and oxyd of iron; this is the method by reaction, so called, already alluded to by Reaumur, and even by Vanaccio for obtaining forge steel, and afterward recommended by Clouet, Mushet, Hassenfratz, Uchatius, and others, for making cast steel. The fusion was then effected in crucibles, as in establishments where blister steel is melted, whereas at present a reverberatory is employed. Hassenfratz mentions the latter furnace as early as 1812. Heath, John Davie Stirling, and Bessemer, have tried it in England, the former in 1845, the two latter in 1854 and 1855. Later, in 1858, Sudre and Petin and Gaudet employed it, for a very short time, in France; the process, however, did not become practical until after the prolonged experiments of Alexandre in the imperial foundries at Villeneuve and Ruelle, in 1861 and 1862, and especially throurgh the exertions of Martin since 1865, at his works at Sireuil. Let us now treat somewhat of each of these processes. BESSELVER, PROCESS. 39 BESSEMER PROCESS. Bessemer Process in France. The Bessemer process is at present widely disseminated throughout all countries, as the Universal Exposition has shown. In France it is regularly conducted at Imphy, Assailly, Terre Noire, and Mutterhausen: at several other establishments as an experiment. The experience gained confirms generally the results attained in my two memoirs of 1861 and 1862. It has been everywhere proved that pure pig iron is necessary to obtain good products. The finingr is successful on this condition only. The expectations of Fremy to the contrary have thus far not been realized. The most suitable irons for Bessemer steel are obtained in France from the ores of the Pyrenees (massive ores of Canigou and Vicdessos), and from the magnetic ores of Molita-el-Hadid (BOne), and from the island of Sardinia. The spathic ores of the Alps would be equally suitable, and probably also the Perigord hematites containing( mang(anese and certain granular ores from central France. The ores of the Pyrenees are brown hematites and spathic ores more or less decomposed. They are worked in the blast furnaces of Ria, La Nouvelle, Berdoulet, Pamiers, etc. Those from BOne and the island of Sardinia resemble Swedish ores. They yield 60.-65. per cent. of pi(r iron containings a small proportion of manganese. At some establishments the proportion of manganese is augmented by adding ore from Garrucha (South 40 THE MANUFACTURE OF STEEL. of Spain), or even oxyd of manganese, to the charge in the furnace. The magnetic oxyd of the island of Sardinia is smelted at Givors, in the coke furnaces of the Petin-Gaudet Company. The ore of Mokta is worked at Terre Noire, Saint Louis, Vienna, Givors (Rochette blast furnaces), Chasse, Creusot, etc., and it is owing to this ore that Creusot now produces irons of superior quality with raw coal. The English method is used universally in the Bessemer works of France; that is to say, the movable converter with the addition of pure spiegel pig iron, for recarburizing the burnt iron. Success is only attained by using gray irons. When white pig iron containing little silicon is treated, the immediate reaction of the oxyd of iron upon the carbon and the separation of carbonic oxyd which ensues, prevents the mass from being heated. It remains viscous, and this want of fluidity induces explosion. On the other hand, when the pig iron contains silicon and a small proportion of manganese, the mass is hotter, because at the outset solid oxydized products only are formed. That the operation be successful it is necessary that the first period, that of scorification or refining, be relatively of long duration. Jordan observes, in his interestinc memoir on the iron works of Siecen, that an excess of manganese also causes explosions, and that gray iron should not contain above two per cent.* *~Etat de la me'tallurgie dut fer dans le pays de Siegen, 1864, p. 32.-Upon the subject of these explosions Jordan says: " They are possibly owing to the facility with which manganese would absorb oxygen at a certain temperature, BESSEM'ER PRO CESS. 41: The same observation was made in France, except that the limit of two per cent. was not absolute, it should vary with the relative proportions of silicon and carbon. We know that, all things being equal, a pig iron contains less silicon and often more carbon, according as the charge in the blast furnace is richer in manganese. In all cases the injurious influence of an excess of manganese upon the working of the Bessemer process has been universally proved. The Bessemer converters in France, as in England, have, up to the present time, been charged with iron remelted in a reverberatory. The works of Terre Noire afford the first instance of deviation from this custom, imitating in this the practice previously adopted in Sweden and Austria in the treatment of charcoal irons. The works of Terre Noire possess two converters of 3,000 — 4,000 kilogrammes capacity (6,600-8,800 pounds) each. They blow several times a day. They fine directly almost the entire product of one blast furnace.* WVhen the iron is gray the operations usually last fiom 20 —25 minutes; when the pig iron is whitened at the edgles it lasts 15 minutes only. Explosions are then frequent, and part with it at a lower temperature." This explanation. seems to me scarcely admissible. The explosions result rather fron the high proportion of carbon which manganese irons always contain, and probal)ly also from the refractory condition of the metallic manganese, which renders pig irons less fluid. Finally, it must not be forgotten that pig irons rich in manganese contain little silicon, a circumstance which facilitates the reaction of the oxyds upon the carbon. *The company is about to enlarge its Bessemner shop at Terre Noire, and establish another at Bess6ges. 42 THE MA4NUFACTURE OF STEEL. because the metal remains slightly cold. The precise results of this treatment are not known to me, but in all cases at least 80 per cent. merchantable products are obtained; and what proves the economy of the process is that these works made a tender to the Paris, Lyons, and Marseilles Railway for an order of 22,000 tons Vignoles rails of Bessemer metal, at 315 francs ($58.56)* per ton, at the works. At this establishment the rail steel may be slightly tempered; the plate steel does not take a temper, and rather belongs to the categorv of homogeneous irons. The former corresponds to Nos. 5-6 of Tunner's scale; the latter, to Nos. 6-7. According to experiments of tensile strength made at Terre Noire, the ordinary Bessemer steel (No. 5) breaks with a strain of 70 kilogrammes per square millimetre (99,554 pounds per square inch), and the soft plate steel with a strain of 55-60 kilogranmmes (78,263-85,360 pounds per square inch), while charcoal plate yields with 35 kilogorammes per square millimetre (49,809 pounds per square inch). The first elongates 5-8 per cent. before rupture, the second, 15-20 per cent. In 1866, and toward the end of 1865, iron rails were made at Terre Noire with a Bessemer steel cover; but this bastard system was abandoned on account of the danger of separation of the weld, to which they were liable after several months of service. * In this and similar subsequent instances the value of a franc is taken at 18.6 cents.-Translator. BESSEMER PRO CESS. 43 About this time the Orleans Company, before deciding in favor of rails made entirely of Bessemer steel, required from Terre Noire two series of tests to prove the uniformitv of manufacture. The following are the specifications given for these tests by Mr. Nordling, engineerin-chief of the Central Railway System of Orleans.* 1st. Two inoots from the same cast to be taken at random, submitted to the tests here given, and the results shown to be practically identical. This same test must be repeated with three different casts. 2d. A standard cast to be made, from which an ingot shall be chosen to which the test just stated is to be applied, six casts to be made in order to arrive at the same standard; an ingot to be taken from each cast, and the same results to be attained by the tests as those attained with the ingot from the standard cast. These preliminaries settled the tests to be conducted in the following manner: Ist. The ingots, chosen according to the directions just given, are rolled into rails in the usual manner; the pattern of rail adopted is that of the Paris Mediterranean Railway, with inclined foot for switches. 2d. Each rail is subjected to a flexion test, under the following conditions: The rail placed upon two points of support I metre (3.28 feet) apart, is subjected to pressure. The flexion under the weight and the permanent set after its removal are to be noted. * Extracted from a manuscript note furnished by the Terre Noire Company, under date of December 15, 1865. 44 THE MAN17FACTURE OF STEEL. The hydraulic press used in these tests is a very perfect machine, made in the shops of Graffenstaden; the pressure is communicated by three eccentric pumps, so as to insure the utmost regularity. 3d. Pieces of the rails, 2 metres long (6.56 feet), are then tested by shock under the following conditions: The rail is placed upon two points of support 1.100 metres (3.6 feet) apart from centre to centre; these supports rest directly upon a block of cast iron weighing 10,000 kilogrammes (22,000 pounds), and the ram weighs 300 kilogrammes (660 pounds). The annexed tables give the results of the tests made as explained above. By a careful examination of the two tables showing the results obtained, we may draw the following conclusions: Ist. The first series of tests clearly shows the practical ideztity of the two ingots taken at random from the same cast. The regularity of the results up to twenty-five tots weight, which represents nearly the linit of elasticity, is especially remarkable. It may even be said that the differences are much less, relatively to the weights sustained, when the elastic limit is exceeded. 2d. This series of tests was not made to show that the casts were identical, notwithstanding which a remarkable regularity is nevertheless shown. Cast No. 581 is perhaps a little harder than the others, but this is explained, until the weight reaches twenty-five FIRST SERIES OF TESTS. _ ~ —z_~~ 4~.-e~i~ —-----— ~ COMPARISONBETWEEN TWO INGOTS FROM THE SAIVlE CAST. CAST No. 577. CAST No. 580. CAST No. 58!. PRESSURE TESTS. INGOT A. INGOT B. INGOT A. INGOT B. INGOT A. Ir~co'r B. ~ Deflection under Permanent De- Deflection under Permanent De- Deflection under Permanent De- Deflection under Permanent De- Defle(~lt~O~geU. nder Permanent De- Deflection under Permanent I)eCharge. flection. Charge. flection. Charge. flection. Charge. flection. flection. Charge. flection. 1 Pressure. millira inches. millim. inches. millim. inches. mill m. inches. millire. inches. nfilP. m. inches. millire. inches. millire. inches. millire. inches. millim. inches. miliim. inches. millire. inches. ~ooo ~o~9...... ~7 o o6~ o o~ o o~ ~7 o o~7 o o~ o oo~ ~ o o~ i o~ 0.004 lit o oI o~ ~ o o~9 o o oooo ~9 o o~ o~ ooo ~ ~ t o o~ o oI I o o~ ~~1 2oooo o oI oooo o o~ o c ooo 44,092. 2.2 0.087 o.I 0.004 2.2 o.o87 o.l 0.004 2.2 {5.o87 o.I 0.004 2.2 o.ooc o.o79 2.I ~ooo I ~o7....... ~o o~ o~ oo~s ~ o~o o o o~ ~ o~o~ o~ o oo~ o~I o~~ ~1 o~c ~t o~t 0.004 ~ o~o~ o~ ooo ~~~~1 30,000 66,I39...... 5.2 0.205 I.O 0.075 4.7 O.185 I.C 0.075 3.5 O..138 0.5 0.020 3.6/ O,14: O.6I O.O2q 3.2t O.126] 0.008 3.4 O.134 0.2 O.OI 0.2[ 35,OOO 77,162...... 10.8 0.425 7.5 0.295 8.9 O.350 5.2 O.2OC 5.~c 0.228 2.3 O.O90 2.3t O.O9C 6.O[ O.23( 1.6t O.063 4.8 O. I8C I., 0.04 4o,000 5.3[ 0.209[ ~~..... ~ o6~ ~9 o4~a ~6~ o~6~ ~ o49~ ~o= o4~ 64 o~;/ I~,~: 9/ I,~,o;,~ ~m xI 3 o 44~ 7.2 0.282.8 0.386 5.5 ] o.216.4 0.3% 5.: 0.205........ kil%,., / lbs., [ millim., inclies, / ~he~ Limit........... kilo~., lbs., millim., inches, kilog., lbs."7 —— Ruptur~:Rupture kilog'., lbs., lgupture. Ruptllre kil%., I lbs., Ituptt~'re./ 55,0o0 I21,254 6O 2.36 54, IOO II9,267 56,500 I24,56I 56,5O0 [ I24,56~ 61,5OO I~;5,384 6O 2.36 62,OOC I36,686 58 2.28 SHOCK TESTS. Deflection after Shock. Deflection after Shock. Deflection after Shock. Deflection after Shock. Deflection after Shock. Deflection after Shock. Height of Fall. millim. inches. millim. inches. millim. inche.~. millire. inches. railtim. inches. millire. inches. [5~etres. Feet. 1.5oo 4.92..'..... 5 o.2o 5 o.2o 4 o. I6 o.2o 0. I6 o. I6 t 1.75o 5.74........ II 0.43 IO 0.39 xO 0.39 IO 0.39 9 0.35 9 0.35 t 2.ooo 6.56........ T9 0.75 x9 0.75 ~6 0.63 I7 0.67 I6 0.63 x4 0.55 2,250 i 7.38........ 27 I.O6 27 1.O6 25 0.98 24 0'94 23 O.90 23 0.9O t SECON SERIES OF T ST COMPARISON BETWEEN SEVEN INGO-I'S FROM DIFFEKEN STANDARD CAST, NO. 564. CAST 582. CAST 585. CAST 8 6. CAST 589 PRESSURE TESTS. DEFLECTION. DEFLECTION. DEFLECTION. DEFLECTION. D E.',,L E C T I 0 N. Under Charge. Permanent. Under Charge. Permanent. Under Charge. Permanent. Under Charge. Permanent. Under Charge. Perr m. inches. millim. inches. millin kilog., el lbs., millim. inches. millim inches. millim. inches. millim. inches, millim. inches. millim. inches. millim. inches. milli I5,000 33,079.............. 1.6 o.o63 O-I 0.004 I-7 o.o67 O-I 0.004 I.8 0-07I O-I 0.004 I.S' 0-07I 0. I 0.004 I.9 0.075 0.1 0.1:20,000 44,092.............. 2.2 0-087 0.2 0.008. 2.2 0.087 0.2 0.008 2.2 0.087 0.2 0.008 2.3 0.090 0.1 0.004 2.4 0.094:25,000 53,IO7.............. 3.5 0-I38 1.1 0.043 2.9 0.1 I4 0.3 MIS 3.3 0-I30 0.7 0.027 2.9 0. I 14 0.3 0.118 3.2 0.126 o., 3oooo 66, 1 I.2 I.0 0.2 -.9 0.I5- 5. I 0.20I 1.8 0.07I 5.61 0.220 2.1 I39......... 0-441 8.5 O-335 4.7 0-I85 0.039 7.2 3 3 i62.............. 23.6 0.930 2 I.0 0.827 10.0 0.394.4 0-2 5 2 i6.2 o.6-8 I2.1 I 1.4 0.449 7.6 0.299 I 5.2 35,000 77, 6 -1 0.476 1 0.597 II., 0.949 20.8 0.8ig I 6.5 o. 6'0' 26.8 22.1 40,000 88)IS5.............. 40.6 1,598 -6.i I-42I i8.6 0.732 I4.0 0-55I 29.0, 24. I.1) 1.055'k lbs., R'pture R'pture kil., lbs., milli., kil., lbs.9 mill., inches, kil., lbs., mill., inches, kil., lbs., RIpture R'pture -il., Limit.................... 50,000 I10,23I so 3-I5 59,000 I30,073 63 2.48 50,200 iio,672 44,500 98,106 54,000 119,049 7( SH011-1K TESTS. Deflection after Shock. Deflection after Shock. Deflection after Shock. Deflection after Shock. Deflection after SI 3letres. Feet. millim. inches. millim. inches. M1111M. inches. Milan. inches. millira. i r, I-500 4.92................ 7 0.28 6 0.24 4 oJ6 7' 0.28 6 I-750 5.74................ I5 0.59 I3 0.5i I4 0.55 15 0.59.4 2.000 6.56................ 26 1.02 23 0.91 0.9s 26 I.02 25 0 1.22 I..,6 2.2'0 7-38................ 37.46 3 I -3.)O 37 35 BESSEMIER PR OCESS. 45 tons, by some tenths of a millimetre difference in permanent set, and has certainly no practical influence. 3d. The second series appears to us likewise to show the practical identity of the six casts made with the object of reproducing type No. 564. This type cast is a little softer than the three casts of the first series; this was done purposely, and the results correspond within the limits of possibility; the differences in permanent set approach some tenths of a zmillimnetre. As in the preceding series, it is only when the limit of elasticity is exceeded that the difference becomes a little greater. 4th. The shock tests show remarkable regularity. The results harmonize perfectly with those of the pressure tests. The deflections under shock shown demonstrate the regularity of the ingots compared with each other, and likewise show small differences of nature. For instance, we observe that cast 581 gives a somewhat stiffer metal than 577 and 580. It is equally easy to prove that the second series of casts is generally much softer than the first. We think we may draw the conclusion that these experiments have shown the results desired, and, as nearly as possible, solve the questions proposed. We will add, that the tests were conducted in a perfectly reliable manner. Mr. Delom, engineer of the Central Railway System of Orleans, has made tests to verify those of the engineers of the Terre Noire Company. The exposition at Imphy contained specimens of Bessemer steel in great variety; the usual manufacture 46 TIlE M,iNUFACTURE OF STEEzL. consists of rails and railway-crossings. I will only add, that lately the pig iron treated is melted in a cupola. It is a step forward, in my opinion, for the nature of the metal is less changed than in a reverberatory furnace. At the Petin-Gaudet establishment at Assailly, near Rive de Gier, two converters of seven tons each are employed. The product of the establishment in Bessemer steel exceeds five hundred tons per month; and, as a third nine-ton vessel is about to be erected, it will soon be greater. The conduct of the process has attained great regularity. Seven special workmen and fifteen to twenty laborers or common workmen are employed. Gray coke irons smelted at Givors from the ore of St. Leon in Sardinia are chiefly treated. This ore contains: Protoxyd of Iron............. 24.00 per cent. Peroxyd of Iron..62.00 " " Oxvd of Manganese.. 0.80 " Lime and Magcnesia... traces " " Quartz.... 13.00 " Sulphur...............0.20 " Phosphorus.................. 0.00 " Total.......... 100.00 It is perfectly suited for the manufacture of.Bessemer metal. Each operation lasts usually from 20-25 minutes; the loss by oxydation does not appear to exceed 12-14 per cent., and the merchantable product averages fully 80 per cent. After the addition of the spiegel pig iron, and before pouring the fined product into the BESSEMER, PROCESS. 47 casting-ladle, they again blow from two to three seconds, according to the nature of the metal desired. The spiegel pig comes fr'om the Petin-Gaudet blast furnaces. Until now both pig metals have been melted in the reverberatory. The Bessemer is likely to replace ordinary cast steel. It is already used for rails, springs, tires, sheet irons, etc. But for some time it has chiefly been employed for many castings, such as housing frames, toothed wheels, pinions, etc., and particularly for cannons of every calibre. According to the special catalogue of the Petin-Gaudet exhibition, the works at Assailly have already furnished above thirty cannons, several of them weighing ten to sixteen tons. I was present at the casting of a similar cannon of fourteen to fifteen tons weight, in October, 1866. Both converters were employed at the same time, or rather by turns at first, on account of the want of power of the blowing-engine; then, previous to casting, the product of both vessels was united in the same ladle. It was thus proved that, when the metal is very hot, it may be kept in repose in the vessel, when turned down, without fear of chillingr. A momentary rest in the casting-ladle is indispensable if perfectly sound castings are desired. Bessemer recommends, in fact, in one of his patents, that of February 12th, 1856, that the metal be allowed to stand before it is poured into the moulds, to allow the gas to be disengaged from the molten mass. He advises, also, in the same patent, the casting of large pieces by ascension, or front below. His advice is followed at Assailly. Can 48 THE IfAY'UFACTURE OF STE EL. nons of steel are cast from a tangential siphon, terminating at the base of the mould, as is done at Ruelle in casting iron cannons. When very hot, cast steel may be kept a longer time in the casting-ladle in a fluid condition than is generally supposed. I have witnessed some casts which lasted ten minutes from the time the vessel was emptied into the ladle until the filling of the last mould. Thus, allowing the molten mass to stand at rest before casting, is to be recommended if we wish to obtain ingots free from blowholes. It is also preferable to cast a single ingot rather than several small ones. Forging pieces of small weight are in that case obtained by dividing the ingot with a chisel, when drawing it under the hammer. Another means which may be employed to obtain castings without blowholes, consists, according to Bessemer himself, in mixingr one per cent. pure pig in the castingladle. The manganese and silicon of the pig metal then absorb completely all the oxygen dissolved, and prevent it from reacting upon the carbon. This means has been successfully employed in Sweden and Austria. The difficulty, however, in this case is to obtain perfectly homogeneous products. A later and more efficient device for obviating blowholes consists in causing a powerful pressure to act upon the molten metal during solidification. Cazalat proposed it in his communication to the Academ y, January 8th, 1866 (Comptes rendus, Vol. LXII., p. 87); and Revollier and Bibtrix have employed it at St. Etienne for some months. By employing r)ressure per BESSEMER PR 0 CESS. 49 fectly sound Bessemer steel car wheels are cast in moulds of iron. The pressure attains 500-600 atmospheres. BESSEMER PROCESS IN ENGLAND. During the last two years the number of Bessemer works in England has greatly increased. The total steel product of every description, in 1861, was barely 1,000 tons per week, whereas last year the total Bessemer steel produced alone was 3,000 tons per week.* The principal establishments employing the Bessenler apparatus are: Sheffield, Brown & Company's works (Atlas Iron Works), and Cammell & Company's works (Cyclops Iron Works); the former contains several ten-ton converters. We may mention the Mersey forcges at Liverpool as possessing two five-ton converters; the great establishment of the Northwestern Railway at Crewe, containing four five-ton vessels; the Dowlais and Ebbw Vale, in Wales, having twelve five-ton vessels (six in each forge); in the north of England, Tudho', and some others at Manchester, and in the Cumberland hematite region. When the product is designed for making tires, springs, sheet-metal, etc., gray hematite pig is used almost exclusively, or else charcoal iron from Sweden, Canada, India, etc. White spiegel pig, from Siegeil, is always added, to recarburize, toward the close of the operation. When, however, the pig metal is to be fined for rails, the * The information concerning England is, in part, extracted from two manuscript memoirs by the pupil engineers Ichon and Michel Levy, written aftersome travels in 1866. 4 50. THE MANUFACTURE OF STEEL. most ordinary ores are charged in the blast furnace. Thus, at Dowlais, the charge per ton of gray iron for rails is ~ TONNES. LBS. 1.455 (3,201) argillo-quartzose hematite, from Cumberland and Lancashire. 0.485 (1,067) roasted ore from coal-fields selected from the purest and most ni'anganesiferous ores. 1.940 4,268 total. Hot blast is employed, and the slags are very basic, yellowish-white, stony, and opaque. The pitg iron thus obtained contains: 4.00-4.50 per cent..............Carbon. 1.10-1.30 " "............... Silicon. 1.50-2.00 " "...............Manganese. Traces................. Sulphur. Less than 0.10 per cent........... Phosphorus. This pig is worked in the Bessemer process with an equal weight of hematite iron, containing a little manganese, very little sulphur, and at most 0.1 per cent. of phosphorus. Both irons are melted together in the reverberatory, with a loss of 6. per cent., and sometimes as much as 10. per cent. Bessemer rail-ends are chargced in the vessel before the molten iron is poured in. These rail-ends are charged at the same time with the coke wrhich heats the vessel. The operation itself lasts twenty minutes. The loss is from 12-15 per cent., besides some 6. per cent. due to remelting. To avoid a violent reaction the spiegel pig is only mixed with the fined metal in the casting-ladle. The fined metal is first poured in from the vessel, and then 7.00 per cent. pure pie b-F-l.. I,, i~' / o,' ~ 3 ~- & 6 7 9 /ofeee. f: —=:,:_:,,. P'~. V. RA.MSBOTTOM'.S DOUBLE HAMIMER Fi~.g. ~!..-........,~,z'f ~ ~ ~ ~ wleIt.......d,,ttpe"lf -1 ~ ~ ~ ~ ~ 1, 1~~~~~~~3 ~ ~' ".........,:' L, o 7~q i..Omd~;7'~ ec 1,4Z,~.r't~r. LU Lili __~~~~~~~~/IL _.. i't' I I 1~%.~ /.~~~~~~~~~~~~~~~~~~~~~~~,;I~~~~~~~~~~~~__:..i ~'.~I [JF~l, ~~~~~~~~I I''ia~~~~~~~~~~~~~~~~'"; —~~~i —- ~~~$ -t,~ I [ ~~'; i'!, i' [,' Nlrl ZIMI~IIJ.,... / " t 2 *d d 78' TEEL. FRA. I RAMSSBOTTOMS DlRECTACTINGDOU3LEIAMMR HEATH S STEEL FURNACE..:~~Fi.5.. Fig 5. 9). | EiZ; l~o~ (i~~ J ~Seldwinorv ott, U. Z..e. Ot ~ Y (Hammer coit t' )e~r I f( tls,.e7/ CO(-e'") tt G 1;,:- 7-,.:~.....''Yahi'e of F&. 6 a 7/to 250a...U I r= fI, - 1 "~'''.....S | s~~~~~~~~~~~~~~~~~~~~~~~~~~~~~US = i. STEEL i=FEURACE ~;'''t' S~t,/et' *i. P ito Io=/t 67. /%0 Y' 20 30 40 5,1 0 y1 a Fl,~": _,:2 c~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ID El~: l YkitoO0vt 7 L~~~~~~~~~~~~~ AP-9":': " o.~](,/.ib Seiw LIE:] lu IMF OF 101 - ---------- ----------------- swagfinlj ---------- --------- - ------------ mom log, IWO OY ITA'. -T.,T STANDARD ENGLISH PLANT. ( I K N / I - ( E X' 1'' p' ",/.-. 2%i4~~~ MN: \ ii ~ ~ ~'__, P171K STANDARD AMERICAN PLANT. 11 - - - - - - - - - - - -. I1i.........,J....AUI,"' i~, I~~~~~~~ I /h "...... =: -4 tP~~~~~~~~~~~t ~ill ili111 1 PHILIP S. MILLER. LENOX SMITH. MILLER & SMITH, No. 53 Exchange Place, New York, EXPERTS AND DEALERS IN Railway Steel & Iron. BESSEMER STEEL RAILS OF SUPERIOR QUALITY. EXTRA IRON RAILS MADE TO SPECIFICATION, AND Railway Equipment in General Supplied to Order. ALSO, BEST GERMAN AND SWEDISH SPIEGELEISEN. RAILWAY SECURITIES NEGOTIATED. A THEITF1SS ON RFGOLLu TURENI FOR THE VEANNTJFACTWUREF OF IIRON, 1ly P}ETER T UINNE1tR. Translated and adapted by JOHN B. PEARSE, of the Pennsylvania Steel Works. Illustrated with 50 Wood-c uts in the text, together with a folio Atlas of 10 Lithographed Plates of Rolls, with Measurements, &c. 1 Vol. 8vo of Text, and 1 Vol. Folio of Plates. Cloth. $10.00. "We commend this work as a clear elaborate, and practical treatise upon the department of iron manufacturing operations to which it is devoted. The writer states in his preface that for twenty-five years he has felt the necessity of such a work, and has evidently brought to its preparation the fruits of experience, a painstaking regard for accuracy of statement, and a desire to furnish information in a style readily understood. The book should be in the hands of every one interested either in the general practice of mechanical engineering or the special branch of manufacturing operations to which the work relates. "-American Artisan. D. VAN NOSTRAND, Publisher, 23 Murray St. & 27 Warren St. p** Copies sent free on receipt of price. TREATISE ON THE METALLURGY OF IRON, Containing Outlines of the History of Iron Manufacture, METHODS OF ASSAY, AND ANALYSIS OF IRON ORES PROCESSES OF MANUFACTURE OF IRON AND STEEL, &c., &c. fBy 11. 1B3,UIEHRMAN. FIRST AMERICAN EDITION. REVISED AND ENLARGED. With an Appendix on the Martin Process for laRing Steel, FROM THE REPORT OF A. S HEWITT. Illustrated with numerous Wood Engravings. 12mno, Cloth, $2.50 "This is an important addition to the stock of technical works published in this country. It embodies the latest facts, discoveries and processes connecting with the manufacture of iron and steel, and should be in the hands of every person interested in the subject, as well as in all technical and scientific libraries. "-Scientific American. D. VAN NOSTRAND, Publisher, 23 Murray St. and 27 Wa-ren St. **e Copies sent free by mail on receipt of price. WEISBACH'S MECHANICS NEW AND REVISED EDITION. A Manual of the Mechanics of Engineering, AND OF THE CONSTRUCTION OF MACHINES. By JULIUS WEISBACIH, PH. D. TRANSLATED FROM THE FOURTH AUG3IENTED AND IMPROVED GERMAN EDITION. BY ECKLEY B. COXE, A.M., MINING ENGINEER. Vol. I.-THEORETICAL MECHANICS 1 vol. 8vo, 1100 pages, and 902 Wood cut Illustrations. Cloth......... $.... $10.00 ABSTZRACT OF CONTENTS.-Introduction to the Calculus-The General Principles of Mechanics-Phoronomics, or the Purely Mathematical Theory of Motion -Mechanics, or the General Physical Theory of lotion-Statics of Rigid Bodies -The Application of Statics to Elasticity and Strength-Dynamics of Rigid Bodies-Statics of Fluids-.-Dynamics of Fluids-The Theory of Oscillation, &c. "The present edition is an entirely new work. Greatly extended and very much improved, it forms a text book which must find its way into the hands not only of every student, but of every engineer who desires occasionally to refresh his memory or acquire clear ideas on doubtful points. An earnest student might take this work, and even without the aid of a teacher, acquire a very thorough knowledge of the subject."-'2fanufacturer and Builder. "We hope the day is not far distant when a thorough course of study and education as such shall be demanded of the practising engineer, and with this view we are glad to welcome this translation to our tongue and shores of one of the most able of the educators of Europe." —The Technologist. D. VAN NOSTRAND, Publisher, 23 Murray St. and 27 Warren St. *** Copies sent free by mnail on receipt of Price, THE KANSAS CITY BRIDGE, WITH AN ACCOUNT OF THE REGIMEN OF THE MISSOURI RIVER, AND A DESCRIPTION OF METHODS USED FOR FOUNDING IN THAT RIVER. By O. CHANUTE, Chief Engineer, and GEORGE MORISON, Assistant Engineer. Illustrated with Five Engraved Views and Twelve Engraved Plates of Plans, &c, 4to, Tillted Paper. Cloth, $6.00. D. VAN NOSTRAND, Publisher,.?3 Mlurray St. ((nd 27 WaiTaren St. *** Copies sent free by mail on receipt of price. THE QUINCY BRIIDGE, AN ACCOUNT OF THE IRON RAILWVAY BRIDGE ACROSS THE MISSISSIPPI RIVER, AT QUINCY, ILL. By T. C. CLatRKE, Engineer-in-Chief ILLUSTRATED WITH TWENTY-ONE LITHOGRAPHIC PLATES OF PLANS. 4to, Cloth, $7 50. D. VAN NOSTRAND, Publisher, 23 Murray St., and 27 Warren St. *** Copies sent free by mail on receipt of price. THIRD EDITION, CAREFULLY REVISED AND CONSIDERABLY ENLARGED. N~ow RBeadp, with 1T5 Double Plates (20 of which now first appear in this Edition), numerous Wood Engravings, and l iportant additions to the Text, in 2 Vols., Imperial 4to., price $60, handsomely halU-bound in morocco. A COMPLETE TREATISE ON CAST AND WROUGHT-IRON BRIDGE CONSTRUCTION, INCLUDING IRON FOUNDATIONS. IN THREE PARTS: THEORETICAL, PRACTICAL AND DESCRIPTIVE. Illustrated by numerous Examples, drawn to a Large Scale,:By IILTTIAA IM -EIBE. Fti, By the same Author, STRAINS IN GIDERS, CAlCULATED DY rORXULI AND DRIARAI, 18mo, Cloth. $2.50. D. VAN NOSTRAND, Publisher, 23 Murray St. and 27 Warren St. Valuable Werks on Iron, Steel, &e. Francis on the Strength of Cast-Iron Pillars, with tables for Engineering, Architects, and Builders. 8vo. Cloth, $2. Whilden on the Strength of Materials used in Engineering Construction. 12mo. Cloth, $2. Campin on the Construction of Iron Roofs. Illustrated with wood-cuts and plates of Roofs lately executed. 8vo. Cloth, $3. -lenrici's Skeleton Structures, in their application to Steel and Iron Bridges, with folding plates. 8vo. Cloth, $3. Humber on Strains in Girders. 18mo. Cloth, $2.50. Crooke and Rohrig's Practical Treatise on the Metallurgy of Iron and Copper. 1 vol., 8vo. Cloth, $10. Percy's Traite Complet de Metallurgie. 5 vols., half Mlor., $45; paper $37.50. IKohn on Iron and Steel. 4to. Half Morocco, $15. Truran on Iron. 4to. Half Moroceo, $20. Fairbairn on Iron Ship Building. 8vo. Cloth, $9. Fairbairn's Researches on the Application of Iron to Building Purposes. Third edition. 8vo. Cloth, $8. Rankine's Machinery and Mlill Work. 8vo. Cloth, $5. Scott Russell on Iron Ship Building. 3 vols., folio. Half Russia, $85. Styffe on Elasticity, Extensibility, and Tensile Strength cf Iron and Steel, with 9 plates. 8vo. Cloth, $6. Roger's Elementary Treatise on Iron Metallurgy. 8vo. Cloth, $5. Kirkaldy's Experiments on Wrought Iron and Steel. Second edition, with numerous plates. 8vo. Cloth, $9. Tredgold and Hodginson on the Strength of Cast-Iron and other Metals. Second edition. 8vo. Cloth, $6. Osborn on the Metallurgy of Iron and Steel. Theoretical aTnd Practical in all its branches, with special reference to American Materials and Processes. Illustrated. 8vo. Cloth, $10. Scofiren's Useful Metals and their Alloys. 8vo. Cloth. $3. Makin's Manual of Metallurgy, more particularly of the Precious Metals. 12mo. Cloth. $3.50. Baker on the Strengths of Beams, Columns, and Arches. 12mo. Cloth. $4.50. Von Cotta's Treatise on Ore Deposits. Translated from the German. By Fred. Prime, Jr., with numerous Illustrations. 8vo. Cloth. $4. Quartz's Operator's Hland Book. By P. M. Randall. Revised and enlarged edition. 12mo. Cloth, $2. Roger's Geology of Pennsylvania, a Complete Scientific Treatise on the Coal Formations. 3 vols. 4to, boards, plates and maps. $30. i- Copies of our new Catalogue of American and Foreign Scientific Books, 80 pages, 8vo, sent to any address on receipt of 10 cents. D. VAN NOSTRAND, Publisher, 23 M3urray & 27 Warren Street, New York. *** Copies sent free by mail on receipt of price. Iron Truss Bridges for Railroads. Nethods of Calculating Strains, with a Comparison of the iMOST PROMINENT TRUSS-BRIDGES And New Formulas for Bridge Computations3 ALSO, THE ECONOMICAL ANGLES FOR STRUTS & TIES. By BVT. COL. W. E. MERRILL, Major Corpa of Engineers. Illustrated with Nine Lithographic Colored Diagrams. 4to, Cloth, $5.'The work before us is an attempt to give a basis for sound reform in this feature of railroad engineering by throwing additional light upon the method of calculating the maxima strains that can come upon any part of a bridge truss, and upon the manner of proportioning each part, so that it shall be as strong relatively to its own strains as any other part, and so that the entire bridge may be strong enough to sustain several times as great strains as the greatest that come upon it in actual use."Scientific American. " He has presented his views in a clear and intelligent manner; and the ingenuity displayed in coloring the figures so as to present certain facts to the eye, forms no inappreciable part of the merts of the work. The reduction of the'formulae for obtaining the strength, volume, and weight of a cast iron pillar under a strain of compression,' will be very acceptable to those who have occasion hereafter to make investigations involving these conditions. " —Railroad Gazette. D. VAN NOSTRAND, PUBLISHER, 23 Murray Street and 27 Warren St. *** Copies sent free by mail on receipt of price. PLATTNER'S MANUAL OF QUALITATIVE AND QUANTITATIVE Analysis with the Blow-Pipe. From the last German Edition, Revised and Enlarged. BY PROFESSOR TH. RICHTER, Of the Royal Saxon 1Mining Academy. Translated by Prof. HENRY B. CORNWALL, A. M., E. M, Assistant in the Columbia College, School of Mines, New-York; Assisted by JOHN H. CASWELL, A. M. ILLUSTRATED WITH Eighty-Seven Wood Cuts, and One Lithographic Plate. One Vol., 8vo. 560 pages. Cloth, $T.50. From Journal of Applied Chemistry. " Mr. Cornwall's edition is something more than a translation, as it contains many corrections, emmendations and additions not to be found in the original. It is a decided improvement on the work in its German dress." " There is nothing so complete to be found in the English language. Plattner's book is not a mere pocket edition; it is intended as a comprehensive guide to all that is at present known on the use of the Blow-pipe, and as such is really indispensable to teachers and advanced pupils. The typographical execution of the work, the numerous wood-cuts, and the lithographic plate of the spectra of metals, are all that could be desired. Ftrom Sillimans Journal. "Plattner's celebrated work has long been recognized as the only complete book on Blow-Pipe Analysis. The fourth German edition, edited by Prof. Richter, fully sustains the reputation which the earlier editions acquired during the lifetime of the author, and it is a source of great satisfaction to us to know that Prof. Richter has co-operated with the translator in issuing the American edition of the work, which is in fact a fifth edition of the original work, being far more complete than the last German edition." "The American editor, Mr. Cornwall, has done a very great service for all students of chemistry and mineralogy who use the English tongue, in thus adding to our scientific literature a work of such rare merit. He has showed excellent judgment in rendering the work into good English, in avoiding needless repetitions, in adding a large amount of valuable material, and in adopting a mineralogical nomenclature which is familiar to Amerioan scientific men." _From Professor Egleston, School of _Mines, Columbia College. "This work is a complete translation of the latest German edition of Plattner's unrivalled Manual of Blow-Pipe Analysis; nothing of the slightest importance having been omitted. Numerous qualitative tests, a quantitative assay for mercury, descriptions of new apparatus, and upwards of one hundred and fifty new minerals have been introduced, with other matters of minor importance, while a general index has been added." DESCHANEL'S NATURAL PHILOSOPHY. TRANSLATED AND EDITED, WITH EXTENSIVE ADDITIONS, By J. D. EVERETT, Professor of Natural Philosophy in the Queen's College, Belfast. IN FOUR PARTS. Part I.-MECHANICS, HYDROSTATICS, AND PNEUMATICS. Illustrated by 181 Engravings on Wood, and one Colored Plate. Svo, Flexible Cloth..................,............$1..80 Part II.-HEAT. Illustrated by 151 Engravings on Wood. Svo, Flexible Cloth...........................................................8.... Part III.-ELECTRICITY AND MAGNETISM. Fully Illustrated. Svo, Flexible Cloth.......................................................................... Part IV. Ready Soon. D. VAN NOSTRAND, Importer and Publisher of Scientific Books, 23 Murray Street and 27 Warren Street. *,* Copies sent free by mail on receipt of price. SPECTRUM ANALYSIS, IN ITS APPLICATION TO TERRESTRIAL SUBSTANCES AND THE PHYSICAL UONSTITUTION OF THE HEAVENLY BODIES, FAMILIARLY EXPLAINED. By Dr. H. SCHELLIEN. TRANSLATED FROM THE SECOND ENLARGED AND REVISED GERMAN EDITION, By JANE and CAROLINE LASSELL. EDITED, WITH NOTES, BY WILLIAM HUGGINS, LL. D. WVith numerous Wiood-Cnuts and Colored Plates; also Angstr;tnis and Kirlchoffs Maps. 8vo, Rich Cloth Binding.................. $12.00 D. VAN NOSTRAND, Importer and Publisher of Scientific Books, 23 lRlirrray Street and 27 Warren Stareet. *** Copies Sent free by mail on receipt of price.