By utilizing the thermal analysis and the microscopic observation the equilibrium diagram of the Fe-Fe₃C-FeS system was thoroughly investigated.
The materials used were melted together in a Tammann furnace, under a eover of fused glass in which Na₂CO₃ added in 6:1, to protect the oxidation of the melt. 75 alloys thus prepared were subjected to the thermal analysis
In this system, there exist two liquid phases m1 and m₂ in a certain composition range, which is evidently proved by two layers in the solidified mass. The upper layer m₂ consists of the phase rich in FeS, while the under layer m1 of that rich in iron and Fe₃C. Between these phases a monotectic reaction, m₁₃ m₂+γ, and a monotectoeutectic reaction, m₁ m₂+γ+Fe₃C at 1, 103°C take place. A ternary eutectic reaction, m γ+FeS+Fe3C, is observable in a nearly whole area of the ternary ststem. The A₄ A₃ A₂ and A₁ transformations of iron are almost unaffected by addition of sulphur, showing that the sulphur is almost insoluble in the solid iron.
The binary equilibrium diagram of Fe-FeS system obtained by the present investigation agree almost with those of the former investigators; the A₄ A₃ and A₂ critical peints of the alloys were determined experimentally by the present writer. The existence of two polymorphic transformations at 135° and 308° in FeS was evidently shown by the thermal analysis, using a specially deviced apparatus. These transformations were also observed in alloys of Fe-FeS and Fe-Fe₃C-FeS system
From the experimental results the isothermal diagrams, sectional diagrams, and projectional diagram of the space model of this ternary system were constructed. A qualitative binary diagram of Fe₃C-FeS system was also deduced from the diagrams of the ternary system.

View full abstract

The ordinary process for the determination of nickel in iron and steel is precipitation by Dimethylglyoxime. This method is very accurate in the presence of all other elements except cobalt. When more than traces of cobalt are present, the method must be slightly modified. of modifications, the one apparently in most general use is the Methods of U.S. Steel Corporation that is based upon the oxidation of cobalt by sodium or potassium chlorate in ammoniacal solution. By this method the author determined nickel in iron and steel in the presence of varying amounts of cobalt. And he found the fact that this is an excellent process giving good results when the amount of cobalt in the solution of the sample does not exceed about 0·02 grams but when it is in excess of this limit the precipitate of the nickel glyoxime obtained may not have its true scarlet color owing to the contamination with cobalt and iron and too high results are obtained. So any remedy for this effect of cobalt is necessary if the solution of the sample contains more than 0·02 grams of cobalt. Some experiments were made to study the methods applicable in such a case. and of these the following method proved most successful:-
Proceed as for nickel as in the Methods of U.S. Steel Corporation obtaining the impure nickel precipitate. The precipitate is dissolved off the filter with nitric acid allowing the solution to run into the beaker in which the precipitate of nickel was made. Heat and destroy any remnant of dimethylglyoxime. Cool and make the solution ammoniacal. Boil and filter the iron hydroxide. The precipitated iron hydroxide willcarry out with it a small amount of nickel. This iron hydroxide is redissolved in hot I; I nitric acid and again precipitated with ammonia. From the combined fitrate the pure nickel precipitate can readily be obtained.

View full abstract