Page:Encyclopædia Britannica, Ninth Edition, v. 13.djvu/351

IRON 335 ter, pig rich or poor in silicon and carbon as the case may be being then added to give the requisite composition. According to Stead the rate at which silicon is oxidized in the converter relatively to carbon is somewhat different according as the temperature attained is extremely high or comparatively low. In the latter case the silicon is wholly eliminated before the carbon disappears, even when the original amount was large, say 3 per cent, silicon and 3 5 of carbon ; in the former case, however, the relative rate of oxidation of the silicon is less rapid, so that the carbon becomes oxidized before the last traces of silicon have dis appeared. Thus the curves illustrated in fig. 53 indicate the 40 2-0 50 I the blast furnace, chiefly on account of the much higher tempera ture of the converter gases and the intermittent character of their generation during a series of blows ; but stoves for heating the blast for the cupolas used in melting the spiegeleisen, &c., have been successfully heated in this way, the flame being applied in much the same way as that obtained from blast furnace gases. For the descrip tion of a method used for this purpose in Sheffield, see a paper by C. B. Holland, Journal I. aiid 8, Inst., 1878, p. 104. 28. Heaton s Process. The chemistry of this process is closely allied to that of Bessemer s, the oxidizing gases used to decarbonize the pig iron being blown through the fused mass so as to produce either a semi-steel, a harder steel, or something approaching malleable iron, according to the amount of carbon oxi dation effected, the gases employed being, not ordin ary air, but the mixture of nitrogen, nitrogen oxides, and. oxygen evolved by the action of heat upon sodium nitrate. This salt is packed at the bottom of the con verting vessel (usually a vertical cylinder of iron 25 ?0 At lower temperature. Fig. 53. At higher temperature. percentages of carbon and silicon contained in the metal in different stages of the blowing, the dotted lines representing the former and the continuous lines the latter, the abscissas representing times and the ordinates percentages. Just at first the carbon percentage slightly increases owing to oxidation of metal and silicon and their removal as slag ; but by and by the rate of carbon removal becomes rapid, whilst the rate of silicon removal is a maximum at first and gradually diminishes, more rapidly at the higher temperature. Attempts have been frequently made to use the spectro scope as a means of determining the moment when the last portions of silicon and carbon have become oxidized ; but it is very doubtful whether the indications of the character of the flame as seen by the unaided eye are not as a general rule practically at least equally valuable with those of the spectroscope, and much more readily attainable. At any rate, the spectroscope is but little used in actual manufac turing practice in England ; on the Continent, however, its use is somewhat more frequent. 1 The more smoky the flame the less distinct are the spectroscopic indications, so that with highly manganized pig (e.g., pig from Austrian spathose ores) the instrument is all but useless. With certain kinds of pig the progress of the decarbonization can be readily judged by the colour of the slag ; the peculiar roar of the blast alters slightly in character when the decarbonization is complete, so that an experienced hand can judge by the sound alone when the operation is about finished. For details of apparatus, &c., see 36. The slag formed during the blowing usually approximates in composition to the metasilicate formula K 2 O, Si0 2 , where R. 2 is either iron or manganese, in this respect differing from the tap cinder of the ordinary puddling forge, which is much more basic. This arises from the highly silicious nature of the lining of the converter (see 36). It has been proposed by Snelus to utilize the gases discharged from the converter, especially during the latter half or so of the blow, in the same way that blast furnace gases (which they closely resemble in general composition, 18) are utilized ; the practical difficulties are, however, considerably greater than in the case of 1 A report of a lecture on the use of the spectroscope in Besse- merizing, given by Roscoe to the Iron and Steel Institute, is to be found in the Journal I. and S. Inst., 1871, ii. p. 38. Alleyne has attempted to utilize the spectroscope for the determination of phos phorus in steel, ibid., 1875, i. p. 62. lined with firebrick) to the extent of about 10 per cent, of the weight of the pig to be treated, and covered over with a perforated cast iron plate ; the pig, previously melted in a separate furnace, is run into the converter ; the heat melts the sodium nitrate and causes an evolution of gas, at first comparatively slow, but gradually increasing in violence as the perforated iron plate melts, until a rapid ebullition of the whole mass takes place ; after a few minutes the reaction is over ; the partly decarbonized fluid mass is then run into ingot moulds (if the mass operated upon is sufficiently large to render the product fluid enough) or otherwise removed from the con verter, conveniently by detaching the bottom portion, which is made removable purposely; the masses of " crude steel" are then reheated and rolled, or melted in crucibles, so as to produce either bar or crucible steel as required. Unlike the Bessemer process, Heaton s method brings about a sensible diminution in the quantity of phosphorus present. In all probability this is due to the alkalinity of the cinder owing to the soda from the nitrate, this acting like the lime lining to the converter in Thomas and Gilchrist s modification of the Bessemer process ( 37); it is evident, however, that the character of the resulting product depends on the uniformity of the pig iron used, and the amount of nitrate of soda employed. The practical difficulties in the way of carrying out the process on the large scale, and of securing uniformity of product, and the non-entire removal of phosphorus, have prevented this method from seriously competing with the other leading steel-making processes; but a number of experimental trials made on a moderately large scale have demonstrated the possibility of obtaining a good class of useful steel by its means. According to Griiner the elimination of phosphorus becomes almost imperceptible if the cinder formed contains upwards of 30 per cent, of silica produced by the oxidation of the silicon in the pig. VI. METHODS INVOLVING THE PKODTJCTION OF STEEL OR MALLEABLE IRON EIFECT FROM THE ORE WITHOUT PASSING THROUGH THE STAGE OF FUSED PlG IRON. 29. Catalan Forge. This variety of bloomery may be taken as being a typical development of the earliest crude apparatus for extracting iron from its ores, represented in almost its simplest form by the rough clay furnace used for the first stage of wootz making ( 35), and by the analogous small furnaces in use in Burmah, Madagascar, Borneo, &c. ; with various modifications it is still in use in different localities, e.g., the Pyrenees, Corsica, and especially in some parts of America and Canada. In principle all these forges may be considered as a more or less enlarged black smith s or ordinary rivetting forge, in the bed of which are