Page:Encyclopædia Britannica, Ninth Edition, v. 13.djvu/352

 336 IRON placed together the ore to be reduced and the fuel ; the blast being applied, partly by the direct action of the cirbon, partly by the carbon oxide generated, the iron ore is gradually reduced to a spongy mass of metal which by stirring is gradually agglutinated into a ball which is re moved and worked into bars, Arc. The Pyrenean forge essentially consists of a silicious stone bottom (covered over with a &quot; brasque &quot; of charcoal powder rammed down), with a tuyere inclining doAvnwards, as indicated in fig. 54 ; the front part of the forge is filled up with the ore to be reduced, and the hinder part with charcoal, and the whole coated over with a layer of moist mixed fine ore and charcoal dust (greil- lade) to moderate the combustion. A. gentle blast is at first applied, and the formation of flame channels through out the mass avoided by putting on more greittade wherever any flame of magnitude ap pears on the surface : iu this way the ore gets largely reduced to spongy metal by the carbon oxide ; a portion is only reduced to ferrous oxide, and this unites with the silicious gangue, forming a fusible cinder which, bathing the mass of spongy metal, prevents its becoming highly carbonized ; finally, the blast being increased and the whole contents of the hearth being gradually stirred together, the reduced metal becomes agglomerated into spongy masses. According to the way in which the materials are manipulated, the resulting iron is more or less steely in character : a slow and prolonged reduction facilitates the carbonization of the metal, yield ing a product containing 5 per cent, of carbon and upwards ; a more rapid blast, and one impinging more directly on the bath of melted cinder and spongy metal at the base of the hearth, produces a softer and far less carbon ized iron. Even after well working under the hammer until fibrous in texture, the bars produced are apt to be non- homogeneous, steely portions being irregularly interspersed. Owing to the non-addition of flux, a great waste of metal is produced by the formation of cinder containing usually some 30 per cent, and more of iron as silicate ; the exact amount of loss of course varies with the ore employed, being less the purer the ore and the more open its texture, so that reduction takes place more readily ; for this reason brown haematites of not too compact texture are the ores preferably worked by the Catalan forge, the expulsion of water on first heating leaving the residual oxide in a con dition in which it is more readily attacked by the reducing agents than is the case with nearly anhydrous compact red haematites, &c. In various places where the Romans smelted or reduced iron from different kinds of ore by processes substantially the same as this, large accumulations of highly ferruginous cinders exist, e.g., in the Forest of Dean, Elba, Spain, &c. ; many of these contain suffi cient iron to be capable of being profitably smelted in the blast furnace either alone or admixed with other ores, being frequently quite as rich as average Cleveland clay ironstone after roasting ; they usually approximate in composition to the orthosilicate type, 2%0,Si0 2. The Catalan forges of the south of Europe are usually of such dimensions as hold from 3 to 10 cwts. of ore ; those formerly in use iu Austria, and to some extent still in America (e.g., the Champlain forge), have the blast heated to a greater or lesser extent by the waste flames from the forge, which are made to heat a serpentine through which the blast is blown, thereby causing a considerable saving in the amount of charcoal required, sometimes amounting to about 20 per cent, of the quantity requisite with cold blast when the blast temperature averages about 300 (Sterry Hunt). The titaniferous iron sand smelted at Moisie in similar forges is much less easily reducible than most other ores, and consequently con sumes a much larger amount of fuel in proportion ; owing to the almost complete absence of sulphur and phosphorus, a very fine quality of iron can be produced therefrom. The American bloom- T3ry processes, whilst resembling in general principles the old European Catalan forge methods, differ considerably therefrom in working details, mode of construction of hearth, dimensions, &c. An account of them and the ores worked by them, abridged from Sterry Hunt s Reports to the Canadian Geological Survey, is to be found in the Journal Iron and Steel Inst., 1871, ii. 103 and 126. 30. Spongy Metal Processes. The essential chemical reaction taking place in the Catalan forge being the reduction of oxide of iron to the metallic state by carbon and carbon oxide, it has been attempted by many inven tors to effect this reduction on a larger scale and in separate apparatus, the spongy metal thus obtained being employed either to form malleable iron by heating to a welding heat and hammering, &c., or to produce steel by fusion in crucibles with carbonaceous matter or intermixture with fused cast iron (free from any considerable quantity of sulphur and phosphorus) so as to reduce the per centage of carbon to any required limits in proportion as more or less spongy non-carbonized metal is employed. This latter application comes rather into the category of processes described later on ( 36 et seq.) than into the class at present under discussion ; its employment has met with some considerable degree of success in one form or another, which can hardly be said of the former methods of working up the spongy metal, at least from a commercial point of view. Some of the earliest experiments in this direction were made about 1837 and 1840 by Clay, 1 whilst since then and even quite recently several attempts have been made to prepare either iron or steel by operations substantially of the same description by Gurllt, Larkin (who reduces very pure fine magnetic ore by heating vith charcoal powder, separates the spongy metal by a magnet, and fuses it with spiegeleisen in crucibles), Renton (of Cincinnati and Newark), Henderson (of Glasgow), Yates, and Snelus, who utilizes the &quot; Gherstonofer&quot; furnace used for burning pyrites smalls in vitriol making, substituting powdered iron ore for the pyrites, and a reducing atmosphere for the hot air employed to oxidi/c the pyrites and burn off the sulphur. Moderately large scale experi mental trials in this direction have been made by Chenot, and sub sequently by Siemens and by Blair of Pittsburg. Chenot s reduring furnace was essentially a series of vertical conical retorts heated externally, the ore (Bilbao, Sommorostro, or other ores containing but little sulphur and phosphorus) being either mixed with car bonaceous matter and the mixture heated in the retorts, or else being placed therein alone whilst a mixture of carbon oxide and nitrogen (prepared by blowing air through incandescent charcoal) was passed through them ; from time to time the reduced spongy metal was drawn off at the base of the retorts into covered iron boxes, so as to prevent reoxidation as much as possible, and then heated in a charcoal hearth and made into a ball when pasty. It is mainly in this last stage that the practical difficulties of the pro cess are encountered : if the sponge be not powerfully compressed into comparatively solid blocks, an enormous waste by oxidation during the balling is produced ; in any case, as it is practically impossible to continue the reduction of the ore in the first stage sufficiently long to remove all oxygen from it (the time requisite and the cost of fuel being then excessive), a great waste of iron ensues. These difficulties have hitherto proved fatal to the com mercial success of the process ; but it has been shown conclusively that a very good iron may be produced by its means, provided sufficiently pure ore be used. The same causes of failure for the most part apply to the earlier methods of Clny, and the subsequent ones of Gurllt, Renton, Yates, Snelus, and Blair, which in principle are all much the same, the nature of the apparatus employed in producing the spongy iron being the main difference in the various processes respectively. Chenot s attempts to produce steel from the reduced spongy iron answered no better, commercially speaking, than the malleable iron manu facture from that source ; the sponge was simply compressed into small blocks after being mixed with charcoal powder, or after being moistened with melted resin, tar, or fatty matters, and heated to char the organic matters, and then melted up in crucibles. The 1 For details of these experiments, see Percy s Metallurgy.