Page:Encyclopædia Britannica, Ninth Edition, v. 13.djvu/355

IRON 330 conjointly. When coal is completely burnt by cold air to carbon dioxide and water vapour (not liquid water), the products of com bustion escaping at a temperature of say 300, the actual heat de- velopment is variable with the character of the coal, but may be taken as approximately near to 7600, the unit of weight being the weight of coal burnt ; for the heat of combustion of ash-free average coal may be taken as near to 9000, or about 8550, allowing 5 per cent, of ash (10), when the resulting carbon dioxide and liquid water are at about 20 s C. ; a less amount of heat, however, is generated if the products of combustion escape at a higher temperature, say 300 C., the dif ference being 4 x 593 + (0 4 x O 48 + 3 O x 0-216) (300- 20) = 462 when the coal is considered to yield 3 times its weight of carbon dioxide aud 4 times its weight of water on complete combustion, 593 being the latent heat of water vapour at 20, and 4 8 and 216 the specific heats of water vapour and carbon dioxide respectively. The nitrogen of the air used for combustion, however, is also heated to 300, starting originally say at 20; making allowance for the oxygen supplied by the ore, suppose that in the Siemens process the nitrogen escaping is 6 times the weight of the coal used, its specific heat being 24 ; then the heat carried away by the nitro gen is about 6 -0 x 24 x (300 - 20) = 403. On the whole, therefore, the effective calorific value of the coal will be 8550 - (462 + 403) = 7685, or 7600 in round numbers. In order to reduce ferric oxide to metal, the heat consumption per unit weight of iron may be taken as about 1700 (contrast 20); the heat carried out from a Siemens furnace hearth by a ball o^ iron will be somewhat greater than that by an equal weight o* fused pig iron from a blast furnace on account of the higher temperature, say 350 instead of 330 ; the same will apply to the cinder, but this increase will be more than counterbalanced by the smaller quantity thereof, so that, assuming 600 heat units to be carried out by one part of slag by weight, and the cinder to amount to 50 per cent of the iron, the heat thus carried out per unit weight of iron will be 0-5x600 = 300. Altogether, therefore, 1700 + 350 + 300 = 2350 units of heat would be requisite per unit weight of iron were it possible to reduce the ores in the Siemens rotator without loss by radiation, &c., and imperfect combustion, the gases leaving the regenerators at 300 ; this would correspond to about f$ = - 309 parts of coal, or somewhat less than 6J cwts. per ton of iron. If then 25 per cent, of the total heat generated by the fuel be utilized, 75 per cent being wasted through incomplete combus tion, gases leaving at a higher temperature than 300, and radiation, &c. , still reduction would be accomplished by an expenditure of only 25 cwts. of coal per ton of iron. By a somewhat different mode of calculation Siemens arrives at much the same result (Chcm. Soc. Journal, 1873, p. 677), viz., that about 6 4 cwts. of carbon aceous matter should theoretically suffice to reduce a ton of iron iu the precipitation furnace ; and hence that about 25 per cent of the heat actually capable of being generated is actually utilized. This high "duty" (as compared with other operations of the iron industry, especially with the blast and puddling furnaces con jointly) arises from the circumstance that whilst the reaction is proceeding carbon oxide is copiously evolved from the materials, and this is burnt in the furnace itself by admitting air and very little other gas so as to keep up the temperature almost without extraneous fuel; the carbon dioxide produced by the combustion, being above and not in contact with the reacting substances, does not in any way interfere with their action, in which respect the process of reduction in the precipitation furnace markedly differs from that in the blast furnace. The following table, prepared by L. Gordon for Siemens (Journ. I. and S. Inst., 1873, p. 57), is of interest as representing the rela tive consumption of fuel during the production of one part by weight of iron by various of the processes largely used at different epochs up to the present date, Charcoal : Ancient Direct Processes. Average. Equivalent in Wood. I. East Indian forges ("Catalan 5-0 to 8-16 2-75 2 98 G-58 2-87 21-9 9-G Siegen 4-40 14-7 II. { Styrian and Carinthian ... Stiickofen 2-85 3-07 2-89 4-00 9-G 13-3 1 Chenot s process 2-66 2-90 2-78 9 3 III. Siemens rotator process . 2-0 Ciwircoal : Blast Furnace and Puddling Forge. Blast Furnace. Puddling Forge. Total. Equivalent in Wood, i f Styria and Carinthla jy 1 Rhine Average. 0-71 0-96 Average. 0-90 0-95 1-61 1-91 5-4 6-4 1-43 1-00 2-43 8-1 [ Sweden 1-21 1-00 2-21 7-4 100 paris of wood reckoned to yield 30 of charcoal. Coal : Blast Furnace and Puddling Fire. Blast Furnace. Puddling Forge. Total Coal. Average. Average. 1-00 3 75 O.QQ 90 3 28 2 39 0-90 3 29 VI. ! Scotland 272 1-00 372 Cleveland 1 99 1-00 2 99 3-02 1-25 4 27 S. Wales (Dowlais) 1-48 0-85 2-33 VII. Siemens rotator process ... 1-25 Details of the manufacture of iron by this method, of its conver sion into steel by further treatment with pig, &c. , in the rotator itself, analyses of the metal and cinder produced, &c., are to be found in the Journ. I. and S. Inst., 1877, p. 345; the total consump tion of fuel for the production of wrought iron of highest quality is there described as being about 3 parts per unit of iron (60 cwts. per ton), of which quantity about one-third is assigned to the rotator, the remainder being used in the reheating furnaces. Within the last twelvemonth Holley has communicated to the American Institute of Mining Engineers the results obtained with a large rotating furnace set up at Tyrone, Pennsylvania, to produce material for open hearth steel furnaces by Siemens s direct process. The charges were ore (containing about 50 per cent, of iron) 2000 lt>; reducing coal 600 to 700 Ib; limestone 250 lt>; scale and cinder 800 lt>. The yield of blooms was 1600 to 1700 Ib per charge, or 80 to 85 per cent, of the metal contained in the ore; nineteen operations per week producing 1 4 tons of blooms were made, the producer coal being 3800 K> per ton of blooms. The total coal consumption was thus on an average near to 4600 Ib per ton. Jones s Process. A. peculiar process, in principle some what analogous to the Siemens precipitation method, has been proposed by F. F. Jones (Journ. I. and S. Inst, 1873, 251), consisting of a cupola furnace into which ironstone slag and coke or other fuel is charged, an air blast being applied so as to melt the ores by the heat developed by the combustion of the fuel ; another blast is then turned on through a second set of tuyeres directed downwards obliquely towards the bottom of the hearth, the jet of gas thus introduced being the mixture of carbon oxide and nitrogen produced by blowing air through a second cupola full of coke only after the fashion of the Tessie 1 du Motay gas producer ( 10) ; by this means rapid reduction of the iron oxide is brought about, the process being a sort of inverted Bessemer blow, oxygen from fused oxide being burnt out by a stream of carbon oxide instead of carbon being burnt out from fused cast iron by a stream of air. Carbon is taken up by the metal thus produced to the extent of several tenths per cent., but it is remarkable that silicon is not thus reduced. Phosphorus is largely present in the resulting metal if contained in the materials used. VII. CONVERSION OF MALLEABLE IKON INTO STEEL BY DIRECT CARBONIZATION. 32. Cementation Process. It has been known for a long period, some two centuries at least, that when wrought iron is enveloped in powdered charcoal and heated to redness for a long time it gradually becomes carbonized and con verted into steel, the deposition of carbon commencing at the outside and gradually penetrating inwards in precisely the same way as that in which the decarbonization of iron proceeds in the manufacture of malleable cast iron ( 22), a longer time being consequently requisite for the car bonization of thicker than of thinner bars ; the name of the inventor of the process, however, has been forgotten. In the middle of the 16th century it was known that when a bar of wrought iron was kept immersed for a long time in molten cast iron it gradually became acierated by taking up carbon from the cast iron ; this process is clearly closely allied to cementation in solid carbon, and was pro bably the forerunner thereof ; very likely it was in the first instance an accidental observation ; it was described as being in actual use about that period by various writers, notably Biringuccio in 1540 and Agricola (De Re Metallica,