56o 



NATURE 



[April 13, 1893 



ON THE CARBURISATION OF IRON. 

 II. 



IN a previous communication (Nature.voI.xIvI. p. 283) 

 the problem of the distribution and absorption of 

 carbon by iron has been discussed, and it has been 

 shown that the process is akin to that of the solution of 

 a salt soluble in water or an acid liquid, that at low tem- 

 perature solution proceeds slowly, the solubility in- 

 creasing with the temperature, until at the final high 

 heat of Bessemer blown metal, or fluid nearly pure iron, 

 the reaction is almost instantaneous ; the carbon, and 

 also manganese, contained in the spiegel-eisen used 

 for this purpose diffusing throughout the fluid metal in a 

 very short space of time. The same occurs when 

 carbon only, in the form of charcoal or coke, is added in 

 lieu of Spiegel, as in the Darby process of carburising. 

 By this latter process, however, about 30 per cent, 

 excess of carbon must be added over and above the 

 theoretical quantity required to insure a given percentage 

 of carbon, for instance, | per cent. For lower percent- 

 ages the excess must still be maintained, but with a 

 corresponding diminution of the total weight of carbon 

 used. In some instances more than 30 per cent, is used, 

 according to the methods of procedure. In practice 

 this holds good and the quantity of carbon required can 

 thus be regulated. 



A priori this would seem impossible. An excess 30 per 

 cent, above the quantity necessary being used, it seems 

 strange that, at the high temperature in the presence of 

 a considerable excess of fluid metal, that nearly the whole 

 of the carbon is not taken up, more especially when 

 iron, as is well known, may absorb as much as 5 per 

 cent, of carbon in the blast furnace ; usually, however, 

 cast iron contains not more than 4 and spiegel eisen 

 5 per cent, carbon, the latter alloy of manganese and 

 iron apparently conferring greater solubility. It even 

 suffices to pour the fluid metal on the pulverised 

 carbon previously placed in the ladle, and a very even 

 product is thus obtained, sufficing for all practical pur- 

 poses, the variation in the percentage of carbon ab- 

 sorbed or dissolved falling within'the limitsof experimental 

 error. It is possible that after absorption of carbon 

 equalling say \ per cent., if the iron were left in contact 

 with carbon for a longer period, more might be taken 

 up ; and that with iron already charged with carbon, 

 solution ma) be retarded ; .the rate at which the 

 latter is taken up probably bearing a certain ratio to 

 the amount previously absorbed. If carbon simply exists 

 in solution this is very probable, and yet the theory would 

 hardly afford at first sight a feasible explanation of the 

 even absorption of carbon which thus takes place, were 

 it not well known that most chemical reactions, so to 

 speak, fall into the same category. 



Chemical affinities are not entirely governed by actual 

 values ; or the affinity of one element for another ; the 

 mass or relative weight of the bodies present influences 

 the final result ; and it is conceivable that, assuming we 

 have two bodies in solution, the addition of a reagent 

 having a greater affinity for one of these may not, in the 

 presence of an excess of the other, exert its full power, 

 the greater mass or weight of the latter apparently 

 weakening, or rather partly neutralising, the chemical 

 force of the reagent added. 



Further cases can be quoted where relative masses in 

 solution are so evenly balanced that a slight excess of 

 the reagent added determines the precipitation of one or 

 the other at the will of the operator. 



Barium sulphate is somewhat soluble in acids, and by 

 prolonged digestion a portion is dissolved. Either barium 

 or sulphuric acid may be precipitated by merely, as 

 regards barium, adding a slight excess of sulphuric acid. 

 On the contrary the addition of a little barium chloride 

 determines the precipitation of sulphuric acid. Apparently, 

 NO. 1224, VOL. 47] 



then, excess or mass of one element overcomes the greater 

 affinity of the other for the reagent added, or, as often 

 happens, a portion is left uncombined and in solution, 

 requiring an excess of the reagent for the complete pre- 

 cipitation or combination. 



Such cases as those above quoted are not uncommon in 

 metallurgical processes conducted at high temperatures. 

 Thus in the case of the manufacture of Bessemer steel, 

 analysis indicates the presence of diverse elements 

 existing together. 



One has— silicon, carbon, hydrogen, oxygen, manganese 

 — also sulphur and phosphorus together with, it is said, 

 carbon monoxide in solution — also probably dissolved oxy- 

 gen in addition to iron oxide. Further, steel with more than 

 \ per cent, of carbon, and also silicon and manganese in 

 sensible quantities, always contains O and H ; and thus we 

 have the elements of water side by side in the presence 

 of a tolerable excess of no less than three bodies, Si, C, 

 and Mn, having affinities for oxygen. 



It is quite true that the abnormally high temperature 

 of the process may weaken ordinary chemical reactions 

 by a species of dissociation ; this has been acknowledged. 

 Yet mass or relative proportions of the elements present 

 must, one would think, influence final results, and thus 

 prevent the complete elimination of the elements named 

 for the reasons already stated. 



The treatment of fluid iron v/ith reagents such as C, Si, 

 Mn, or alkalies, as now practised, is as strictly a chemical 

 process as that pursued by the chemist in his laboratory. 

 In both, reagents are employed which are known to be suit- 

 able for the elimination or precipitation of substances 

 known to be present ; and, so far as can be ascertained from 

 actual practice, the steel-caster deals with molten metal 

 containing certain elements in solution, and endeavours 

 to get rid of some of these, or adds others assumed to be 

 beneficial, just as the chemist works with solutions known 

 to contain bodies possibly existing or combined with the 

 fluid solvent in much the same manner as the worker with 

 fluid iron. There seems but little difference, take it as 

 one may ; the same laws of combination, solution, &c., 

 seem equally applicable ; and differences of opinion as to 

 what is really meant by the terms solution, chemical com- 

 bination, or simply mixtures, are common to both. 

 Further, it must not be forgotten that pure fluid iron, 

 although exerting a direct solvent action on certain bodies, 

 may take up or dissolve a chemical combination or 

 double salt, just as pure water does. This, however, 

 remains an open question, but it would be interesting to 

 know if certain combinations of iron with other elements 

 are thus held in solution. 



As regards carbon there can be little doubt of the 

 existence of definite carbides of iron ; and it may be that 

 combinations of iron with bodies other than carbon may 

 play a part. Some recent work on certain alloys of iron 

 points to the probability of the formation of these. 

 Assuming the presence of a definite carbide of iron 

 which may not be in solution, but diffused evenly through- 

 out the fluid iron, although we cannot be absolutely sure 

 of this, the behaviour of steel under certain conditions of 

 heating and manipulation may be explained on the as- 

 sumption that iron carbide, being certainly more fusible 

 than pure iron, must become soft and plastic at a tem- 

 perature at which the mass of pure metal is scarcely at 

 all affected. This plastic compound would bind the non- 

 coherent particles of the greater mass of iron together, 

 and this mixed or heterogeneous body could be welded or 

 beaten out under the hammer. It is the general opinion 

 that the weldable metals are mixed bodies, are not homo- 

 geneous, inasmuch as bodies purely homogeneous cannot 

 as a rule be welded together. 



Wrought iron welds easily, far more easily than steel, 

 and it is certain that the former is not homogeneous, what- 

 ever may be said of the latter. Wrought or puddled iron 

 is well known as an irregular mixture, composed of grains 



