July 21, 1892J 



NATURE 



283 



I do not for one moment suppose that a natural history 

 museum of this kind is suitable for all places. The genius loci 

 must be, in all places, the genius of the museum. The principles 

 however of success are the same in all, and success can only he 

 achieved in a limited degree if there be no signs of the worship 

 >f some of the Muses in the arrangements. 



The Work of the Museums Associatiou. 

 In ending this address, all too long, I fear, for my audience, all 

 too short for my subject, I must add a few words as to our work 

 as a Museum Association. It is twofold. First, we must arouse 

 ourselves to the present situation and note the directions which 

 the intellectual movement of the day is taking. Next, it is our 

 duty to arouse the public to the importance of museum develop- 

 ment, and to take care that the claims of museums as instru- 

 ments of education shall not be ignored in the grants made by 

 public bodies for the good of the commonweal. 



ON THE CARBURIZATION OF IRON. 

 I. 

 'T*HE conditions under which carbon combines with iron have 

 ■*■ been closely studied, and the observed phenomena fully 

 discussed. Even now, however, it is doubtful whether true 

 chemical combinations of carbon and iron are formed. It 

 has been alternatively assumed that carbon is with difficulty 

 soluble in iron, and that at low temperatures solution may pro- 

 ceed very slowly. In other words, carbon is not easily dissolved 

 except at high temperatures ; and it follows that if highly heated 

 iron fully charged with carbon be cooled, a portion of the carbon 

 must he precipitated in this state, existing simply as foreign 

 matter in the metal, but that, on reheating, it may again enter 

 into solution. Low carbon steels may be regarded as dilute 

 solutions of carbon in iron ; pig or cast iron as saturated ; and 

 intermediate grades may be termed moderately concentrated 

 solutions. 



Against this, however, there is a mass of evidence which de- 

 serves attention and cannot be ignored. Hercy states that for 

 the full carburization of iron a high temperature is necessary, 

 and further, considering the absolute infusibility of carbon, it 

 seems reasonable to assume that these elements must enter into 

 chemical combination. It is, however, admitted that this com- 

 pound may have the power of dissolving additional carbon ; this 

 explains the copious deposition of carbon in the graphitic form 

 when iron is cooled. Dr. Percy finally concludes that there must 

 be at least one definite compound of carbon and iron, but adds 

 that there seems to be no reason why solution should not occur, 

 as in the case of mercury, which liquefies gold, silver, or copper. 



Prof. Roberts Austen also ("On Certain Properties common 

 to Fluid? and Metals," Royal Institution, March 26, 1886) 

 speaks of the power which certain solid metals have of even 

 rapidly taking up fluids — clearly cases of solution. Abel 

 claims to have proved the existence of a definite compound of 

 carbon and iron. Prof. Roberts Austen also finds that heated 

 iron combines with pure carbon in the form of diamond dust. 

 The author also has succeeded in directly combining iron 

 fused in vacuo with pure sugar charcoal presumably freed from 

 gases by repeated heatings in vacuo. Yet it is obvious these 

 instances may all be explained on the theory of solution at 

 elevated temperatures, with the exception of Prof. Abel's, who 

 claims to have isolated a definite carbide of iron from the metal, 



Matthieson, as the result of an elaborate research, states that 

 " with few exceptions " most of the known two-metal alloys are 

 solidified solutions of one metal in another. Carbon-iron 

 alloys may be looked upon as solidified solutions of carbon in 

 iron, and the analogy of cast iron with other alloys indicates the 

 non-existence of chemical combination between carbon andiron. 



Again, viewing alloys as definite chemical combinations, the 

 facility with which heated iron absorbs certain gases does not 

 admit of easy explanation. 



Deville, however, imagines a kind of porosity in the metals, 

 terming it an intermolecular porosity, sufficient to admit of the 

 passage of gas at a low temperature ; and supposes it developed 

 by the expansive agency of heat. Graham assumes that the 

 affinity of the gases for iron and platinum is as the attraction 

 admitted to exist between a soluble body and its solvent. 



Other metallurgists are of opinion that carbon does not 

 directly combine with iron, attributing their union to the in- 



direct action of carbon monoxide gas always present in iron ; 

 by the agency of this gas carbon is indirectly transferred to iron ; 

 but it would appear that this cannot be maintained, for it has 

 been proved that carbon combines directly with iron one way 

 or the other, i.e. by solution or chemical combination. 



Whatever may be said of irons containing an excess of 

 carbon, i.e. cast iron and very hard steel — which, if one grants 

 that carbon is not very soluble in iron at a low temperature 

 may be termed supersaturated solutions— in the case of low 

 carbon steels there seems some ground for assuming that carbon 

 is merely dissolved in the metal. 



Sir L. Bell tells us that, on heating thin sheets of carburized 

 metal or steel piled closely together, the excess of carbon con- 

 tained in one or more of the sheets is transferred to the others. 

 Wrought iron is carburized in much the same manner by the 

 ceooentation process, and it is equally possible that heterogeneous 

 iron, i.e. iron containing intermixed carbon or graphite, and as 

 a rule not equally diffused, may by continued sufficient heat- 

 ing become practically homogeneous. 



It is a well-known fact that the carbon in low carbon steel 

 — for instance, Bessemer steel — exists in at least two different 

 forms ; Prof, Ledebur says four. Akerman (Iron and Steel 

 Institute) classifies these as (i) hardening carbon, or the carbon 

 which determines the quality of steel, (2) cement carbon, and 

 also graphite may be present. 



The united researches of many workers in this field of research 

 indicate generally that a portion of the total carbon is in inti- 

 mate union with the metal, and that the more intimately com- 

 bined or hardening carbon determines the quality of the steel. 

 The carbon incompletely combined (or intermixed carbon) is 

 termed cement carbon, because it occurs in the largest propor- 

 tion in blister or cement steel. 

 I Does not the above point to a case of solution of carbon, in 

 : which the quantity in solution is determined by temperature, 

 i just as with other solutions? 



j Metallurgists, however,-can hardly accept the theory of solution 

 j without qualification. 



Mr. Spencer states that " unhardened steel containing i-i8 

 I per cent, total carbon— of which the colour test indicated "89 per 

 cent, as combined carbon, and residual carbon or graphite '29 per 

 ' cent. — after bein;f hardened, gave only '58 by colour test, and 

 i only traces of graphitic carbon, showing a loss of •51 per cent. 

 j of carbon. A softer steel, containing "50 total carbon — equalling 

 I *45 per cent, by colour test, 04 per cent, graphitic carbon — 

 j after hardening, only '21 colour test carbon; graphitic carbon 

 ! "oo, showing a loss of "29 per cent. Other'nnalyses were made 

 confirming the above, and establishing the fact that after harden- 

 ing there is always a proportion of carbon which can neither be 

 determined as graphite or by the colour test, and this propor- 

 tion is found to increase according to the larger amount of 

 carbon in the metal, and the rapidity with which it was cooled" 

 (Mr. Spencer, Iron and Steel Institute). 



The facts above quoted are not apparently in accord with 

 the theory of solution ; but there are undoubted allotropic 

 modifications of carbon, and this peculiar form may be one 

 of these, " uncombined," and may be classified with the graphite, 

 orreally as merely intermixed foreign matter. 



There is the alternative assumption that the missing carbon 

 may exist in some form or combination with the iron, possibly 

 not capable of being registered by the colour test ; but as the 

 steel is treated with dilute nitric acid, in which it is completely 

 soluble, with the exception of the graphite, this assumption can 

 hardly be maintained. 



Referring to Akerman's assertion that only combined "harden- 

 ing " carbon determines the physical properties of steel — an 

 assertion with which Mr. Spencer agrees — "The apparent loss 

 of carbon shown by the latter, and which we have determined 

 as intermixed carbon or a form of graphite "—it may well be 

 that the missing carbon is so intimately mixed as to be in 

 a state closely bordering on solution, for it is well known 

 that it is difficult to draw the line between absolute solution 

 and matter finely suspended in a liquid. The latter practically 

 often presents the appearance of a solution scarcely to be 

 distinguished from it. Messrs. Harold Picton and L. E. Linder 

 (Chem. Soc, January 1891) are of opinion that there is a con- 

 tinuous series of grades of solution passing without break from 

 suspension to a crysiallizable solution. This seems very probable, 

 and in accordance with our chemical experience. 



Graphite, if the author has adequately grasped Prof. Aker- 

 man's views, has little or nothing to do with the quality of 



NO. II 86, VOL. 46] 



