Page:Encyclopædia Britannica, Ninth Edition, v. 13.djvu/342

326 N sulphur, and phosphorus in the puddle bars obtained from a given kind of pig will vary. Many of the machines for puddling above mentioned, and of the more or less analogous ones invented by Menessier Jones, Gidlow and Abbott, Bouvart, and others, are reported by their inventors to answer far better in this respect than ordinary hand-puddling furnaces, in particular as regards the re moval of phosphorus ; so that with Cleveland pig, for instance, con taining 1 5 to2 - percent, of that element, whilst the hand -puddle bars usually contain about 5 per cent, of phosphorus, only 2 and 0"! per cent, and even less is contained in the machine-puddled metal. This result has been traced in some instances to the use of fettling materials containing but little silica : the production of a comparatively non-silicious cinder greatly facilitates the oxidation and removal of phosphorus (as in the " basic " Bessemer process, 37) ; whilst, conversely, well-puddled iron, if left in contact with a highly silieious cinder, is capable of removing phosphorus there from and taking it up again. Based on this principle, several pro cesses have been proposed for more or less completely dephos phorizing pig iron either during puddling or previously thereto ; among them may be mentioned the following. Bell s Processes. Lowtliian Bell has proposed to desili- conize the pig by blowing for a few minutes in a Bessemer converter, and then to transfer to a puddling furnace and finish the operation therein so as also to eliminate phos phorus ; in case the metal contains so much phosphorus as to come to nature before a sufficient amount of that im purity is removed, spiegeleisen or other carbonized iron free from phosphorus is added to the partially blown metal, so as to prolong the operation of puddling, and consequently to enable a larger amount of phosphorus to bo removed ; or the too rapid expulsion of the carbon may be prevented by blowing into the converter along with the blast carbonaceous matter, such as ground coke or charcoal, &c. Partially refined metal thus prepared charged into the furnace in the fluid state causes much less injury to the lining than ordinary pigs, especially when used solid, and consequently with a good deal of sand adherent to them; and the process has the additional advantage of diminishing the time dur ing which the furnace is required for each puddling heat, so that the yield per furnace is increased, whilst the quality of the metal is also improved ; thus when Cleveland pig was treated by blowing for five minutes in a Bessemer converter, and the fluid metal then puddled, the phosphorus in the final product was reduced to 3 per cent., whereas when puddled in the ordinary way it amounted to - 5 per cent, and upwards. The same result as regards partial purification of the pig iron can also be produced in a more simple way and to a greater extent by " washing" molten pig iron with iron oxide (ore, cinder, &c.) in a fused state, the two being agitated together at a temperature suffi ciently low to prevent the iron from being heated much above its fusing point ; under these circumstances the greater portion of the phosphorus present is oxidized and removed, whilst only a fraction of the carbon is oxidized, although at higher temperatures the carbon is rapidly oxidized ; to effect this either a rotating or an oscillating puddling furnace can be employed, or preferably a special arrange ment consisting of a trough of boiler plate closed at the ends and covered with a brick arch and lined with purple ore at the bottom and sides ; the trough rests on a horizontal axis, so that it can oscil late like the beam of an engine ; hence any material placed inside is continually rolled or made to flow from one end to the other and back by the oscillation. The trough is about 12 feet long, 3 wide, and 3 high ; the interior being red hot and the oxide introduced (either fluid or sufficiently hot to melt when in contact with fused pig iron), the pig is run in, and oscillation kept up for about ten minutes, when the partially purified metal is tapped out and puddled in the ordinary way ; during the oscillation the metal and oxide travel altogether some 700 to 900 feet backwards and forwards from one end to the other. In this way a pig containing much phosphorus is almost wholly dephosphorized, as, for example, in the following instance. Pig before Treatment. Refined Metal. Loss per 100 parts of original Non-metal. Silicon Per cent. 2-0 Per cent. 0-12 94 Carbon 3-5 3-1 11 Phosphorus 1-5 -23 84 Krupp s process (Narje s process) for dephosphorizing iron is in principle very much the same as Bell s method, the phosphorus being oxidized by fused iron oxide, only at a much higher temperature than that used by Bell. Wedding describes the process as carried out at Essen thus : the pig is melted in a 13 foot cupola with coke (taking an hour and a half), and is then tapped into a modified Pernot hearth ( 40) covered with a layer of fluxed ore almost a foot thick, melted on at a very high temperature ; before every heat from 1500 to 1700 lb of ore, heated until sintered, are added before the iron is tapped from the cupola. At first the furnace is made to revolve twice only in a minute, but later the speed is increased to five revolutions per minute. In from five to ten minutes the phosphorus is almost completely removed, the point when this is the case being indicated by the formation of jets of carbon oxide ; prior to the formation of these, the carbon percentage is barely altered, although the phosphorus is acted upon rapidly. Finkiner gives the following analyses of various specimens of metal thus treated : Sample Xo. 1. Sample No. 2. Sample Xo. 3, Before Treat ment. After Treat ment. Before Treat ment. After Treat ment. Before Treat ment. After Treat ment. Carbon (mean of duplicate deter minations) Phosphorus do. 3-94 0-631 8-76 0-132 3-80 0-448 3-57 0-107 3-17 1 -221 3-03 0-302 Manganiferous ores are employed in preference, the presence of manganese being said to increase the rate of removal of phosphorus whilst diminishing that of carbon. Both in Europe and America this method has been employed on the large manufacturing scale with consider able success, especially for the preparation and purification of metal for Siemens-Martin steel. Hamoir s process as carried out at Mauberge (France) essentially consisted in the removal of silicon, &c., by blowing air through the molten pig before puddling it in the ordinary way ; according to Deby an economy of 10 per cent, of fuel was thus occasioned, with an increase of 10 per cent, in daily output (10 heats being worked instead of 9), the quality of the metal being also improved. Smyth has proposed to refine pig iron by blowing in a Bessemer converter or analogous specially constructed vessel with a blast con taining petroleum disseminated through it, together with hydro chloric acid or caustic soda, these chemicals being introduced in a special piece of apparatus or "generator" somewhat analogous to an ordinary spray producer, and the partially refined metal being subsequently puddled in the usual way. A number of experiments made by Maynard at Gorton (near Manchester) indicated that the advantages of the process were hardly in proportion to the extra cost thereby entailed so far as the purifying action is concerned; but there is an advantage in the use of petroleum as a means of obtain ing the high temperature in the converter or "receiver" used by Smyth requisite to keep the blown metal in fusion and prevent "skulls " forming when it is run out into a casting ladle, in cases where the iron has not the requisite amount of silicon present to enable it to generate this heat with the use of air alone (see 27). Pctlitt s process for dephosphorizing pig iron, strictly speaking, is in no way a modification of the ordinary methods of mechanical puddling, being more akin to the "basic" dephosphorization method ( 37). Essentially it consists of the use of a blast carrying with it into the Bessemer converter iron oxide particles, the object being to bring iron oxide into intimate contact with the fused metal, and thus make the slag formed in the body of the iron "basic " ab initio, instead of its being mainly silicate, as in the ordinary pro cess. According to the inventor it is not necessary, in order to produce steel, to employ spiegeleisen, nor to alter the mode of lining with ganister in any way ; he gives the following analyses : Pig Iron before Treatment. Bar Iron Produced. Carbon 2-76 0-37 Silicon 2-01 0-07 Sulphur 0-29 nil. Phosphorus 1-44 0-07 Manganese trace trace