520 SUMMARY OF CUERENT RESEARCHES. 



segregation of cementite (supersaturated zoue) at the extreme edges of 

 some of the disks, but not in others. The view of Giolitti is taken that 

 these segregations are caused by slight oscillations of the temperature 

 during cementation, as in a repeat experiment in which the temperature 

 was made to oscillate between 800° and 1)50° during heating a much 

 more marked supersaturated zone was obtained. The thickness of this 

 cementite layer seldom exceeds 0-002 inch. (Four disks were clamped 

 together for polishing transverse sections, and thus rounding of the 

 edges and destruction of the cementite zone was avoided.) The carbon 

 content of the disks, determined after grinding a thin layer off the 

 surfaces to remove any cementite layer, are given together with photo- 

 micrographs. The saturation content of carbon by cementation at the 

 same temperature varies according to the composition of the steel. 

 None of the carbon-concentration temperature curves coincide with the 

 curve 8E of the carbon-iron equilibrium diagram, but lie well to the 

 left of it. This fact confirms the view that the carbon is transmitted 

 to the iron by gases rather than by diffusion of cementite. The 

 phenomenon of exfoliation of case-hardened and quenched objects — 

 i.e. the splitting away of the superficially cemented layer when subjected 

 to shock — is due to the existence of internal stresses arising from the 

 greater increase in volume on hardening of the high carbon exterior as 

 compared with that of the low carbon core. Nickel-steel is exceptional, 

 as the specific gravity after cementing and hardening is greater than that 

 of the same material hardened but not cemented. As the difference is 

 very small however, failures with nickel-steel are less frequent. 



Determination of the Line SE in the Iron-carbon Diag-ram.* — 

 The method adopted by N. Tschischewsky and N. Shulgin in a re- 

 determination of the line SE of the iron-carbon equilibrium diagram, 

 representing the solubility of cementite in y-iron, was to heat polished 

 specimens of steel to temperatures above and below the line in vacuo, 

 etch at the same temperature with gaseous chlorine, and then i-apidly 

 cool in vacuo. This method avoids decarburization at high tempera- 

 tures of the surface of the specimen. Only two steels, containing 1*25 

 and 1 • 50 p.c. carbon respectively, were investigated. Specimens which 

 at the temperature of etching were below the line SE show traces of a 

 pearlitic structure, while those which were above the line show a 

 completely polyhedric structure, more marked the further beyond the 

 line. An average between the two nearest temperatures where one 

 specimen showed a polyhedric structure and the other showed traces of 

 cementite was taken as the required temperature for that composition of 

 alloy. It is concluded that the line SE can be represented as a straight 

 line,_assuming the eutectoid point as 0'!) p.c. carbon and 700° C, and the 

 maximum limit of saturation of iron by cementite as 17 p.c. carbon and 

 1130° C. The secondary structures, martensite, troostite, etc., were not 

 observed in these experiments. Several microphotographs are given to 

 show the effects of chlorine etching. It was noted that etched specimens 

 were oxidized after a few minutes' exposure in air, and the etched 

 surfaces were, therefore, protected by varnishing directly after removal 

 from the furnace. 



* Journ. Iron and Steel. Inst., xcv. (1917, 1) pp. 189-98 (14 figs.) 



