THE EUTECTOID TRANSITION POINT OF CARBON STEELS. 



197 



16. A Working Hypothesis as to the Mode of Crystallisation of the Eutectoid. 



We may next consider the eutectoid steel itself and describe some experiments 

 made upon it. 



The surface energy between the solid solution and iron is probably less than that 

 between it and the carbide. Accordingly, in such a steel, the lag in the precipitation 

 of the eutectoid is probably terminated by the crystallisation of some of the iron 

 which it contains. But the sudden precipitation of iron would increase temporarily 

 to a very high value the concentration of the carbide in solution in its immediate 

 neighbourhood. Hence, even if the retardations tended to be unequal, the crystalli- 

 sation of the one constituent would induce the appearance of the other. 



It is necessary to attempt to form a picture of the way in which crystallisation 

 proceeds. For simplicity in description we may suppose that, as frequently happens, 

 the eutectoid crystallises in grains, each consisting of, approximately parallel, 

 alternating layers of iron and of carbide.* 



It is unlikely that the production of these alternating layers of very different 

 composition and finite thickness is instantaneous. 



We may suppose that the first stage in the crystallisation of each grain results in 

 the formation of parallel and excessively thin threads at approximately uniform 

 distances apart, consisting of iron and the carbide in the eutectoid proportions, 

 separated by layers of still untransformed solid solution in the way represented 

 diagrammatically in fig. 10 (i.), and that the second stage is the growth of the 

 threads to their final thicknesses, represented in fig. 10 (ii.), by the diffusion in 



Fs, C 



Fig. 10. 



opposite directions, and deposition, of the constituents of the intervening layers of 

 the solid solution. 



Assuming the first stage, it is easy to see that the kind of diffusion postulated 

 in the second will occur. 



Owing to the lag the temperature is below the eutectoid point. Consequently 

 the chemical potential of the carbide in solution is higher than in the separated 



* Cf. e.g. BENEDICKS, 'Recherches sur 1'acier au carbone,' Photogramme 1. 



