.352 



METALLURGY. (!RON AND STEEL.) 



dition in the carbon, a large proportion of which 

 passes from the combined into what is for con- 

 venience called the graphitic state, although the 

 condition is different from that of graphite either 

 as found free in nature or as modified out from 

 gray iron during cooling. The substance is an 

 allotropic form of graphite corresponding with 

 what J^edebur has called "tempering graphitic 

 carbon," and differs from graphite and from amor- 

 phous carbon in specific gravity, specific heat, and 

 calorific power. It can, however, be separated 

 from the iron and chemically determined in the 

 same manner as graphite. The author found, 

 contrary to a statement of Ledebur that iron 

 containing this allotropic graphite hardened and 

 tempered exceedingly well, and when not fully 

 developed by overannealing the whole of the car- 

 bon will pass into the hardening state and leave 

 the iron without a trace of free carbon. By 

 prolonged heating this allotropic form becomes 

 changed into graphite identical with that found 

 in gray iron, when it occupies a similar position 

 in the "iron structure and is very injurious to the 

 physical qualities of the material. The effect of 

 other components of iron on the carbon change 

 has been investigated by the author as to silicon, 

 manganese, and phosphorus. Silicon and man- 

 ganese exert great influence upon the carbon dur- 

 ing the annealing process. The presence of silicon 

 is a necessary condition of the change. Indeed, 

 the relative amount of carbon that can be changed 

 from the combined to the graphitic state during 

 annealing seems to be directly proportionate to 

 the amount of silicon present. The influence of 

 manganese is not so readily marked as that of 

 silicon, and its action is further obscured by the 

 high silicon which always accompanies high man- 

 ganese. Comparatively high manganese, however, 

 assists the change and shortens the time necessary 

 for its completion. The action of silicon is appar- 

 ently of an essentially different character from 

 that of manganese. That of silicon is direct upon 

 the carbon, while the influence of manganese is 

 only indirect through the silicon by protecting 

 it from oxidation during the melting of the iron, 

 and so causing the percentage of silicon in the 

 casting to be higher than it would be if the man- 

 ganese had been more completely oxidized from 

 the bath. In every case irons low in silicon and 

 manganese are practically worthless after anneal- 

 ing. Such irons may be softened by the process 

 so as to machine easily, but are always deficient 

 in tenacity. They generally have a white fracture, 

 in appearance like that of a 30-per-cent. carbon 

 Bessemer steel bar. Low carbon at least com- 

 paratively low carbon does not prevent the 

 change in the carbon condition from taking place. 

 This change can also be produced by the author's 

 method of annealing in other combinations of iron 

 and carbon than white iron. The carbon change 

 described, though gradual, was found to be coex- 

 tensive throughout any given section of the 

 casting. 



The results of the researches of Dr. A. Stans- 

 field into the solution theory of carburized iron 

 are that carbon is less soluble in iron when pre- 

 sented in the form of graphite than when pre- 

 sented in the form of cementite, and that the 

 apparent reversal of this in steel is due partly 

 to the absence of nuclei of graphite, on which 

 further deposits might take place; partly to the 

 length of time required for the separation of .the 

 graphite, involving as it does the gradual passage 

 of carbon through the iron to reach the nuclei; 

 and partly to the mechanical pressure which must 

 oppose the formation of graphite in solid steel. 



In his experiments on the effect of aluminum 



on carbon in cast iron, H. W. Waldron has ob- 

 served that, as compared with silicon, traces of 

 aluminum have a greater effect in producing 

 graphite in the material, although the whole of 

 the carbon is never thrown down as such by it, 

 as occurs with silicon when present in active 

 quantities. With less than 0.5 per cent, of silicon, 

 practically no separation of graphite takes place. 

 A few comparative tests showed that 0.35 per 

 cent, of aluminum reduced the resistance by nearly 

 60 per cent, when slowly cooled, and that further 

 quantities still rendered the metal soft, but not 

 quite so much so. Rapid cooling increased the 

 resistance by compression of those alloys, but not 

 in anything approaching the original strength of 

 the cast iron. 



One of the features of the Paris meeting of the 

 Iron and Steel Institute was a demonstration by 

 Mr. J. E. Stead, of Middlesborough, of the ways 

 in which iron is affected by phosphorus, more 

 especially in conjunction with carbon. According 

 to the late Dr. Percy, there are at least seven 

 definite chemical compounds of phosphorus and 

 iron. The way in which carbon drives out phos- 

 phorus was illustrated in a manner which may 

 possibly lead to further important results. Mr. 

 Stead pointed out that the white constituent in 

 gray iron was formerly looked on as carbide, but 

 it had been proved to be phosphide. He exhib- 

 ited a highly phosphoric specimen of iron having 

 crystals from 1& to 2 inches across. He ex- 

 plained how, following the researches of Sir W. C. 

 Roberts-Austen and of Dr. Stansfield. he accounted 

 for brittle and imperfect steel by the formation 

 of a eutectic alloy. The substitution of carbon 

 for phosphorus led him to suppose that pieces of 

 steel might be welded together without pressure 

 by placing plates 'of steel together with layers 

 of phosphide between, and he had succeeded in 

 obtaining this result. He explained, however, that 

 the junction was more in the nature of a brazed 

 joint than a weld. 



From experiments with wrought iron and vari- 

 ous steels exposed to sea water, river water, and 

 the weather for two periods of about a year each, 

 H. M. Howe has drawn the conclusion that the 

 difference in the rate of corrosion between wrought 

 iron and soft steel is rarely enough to be of much 

 moment, except in the case, perhaps, of marine 

 steam boilers; and that the ratio of corrosion of 

 given soft steel to that of given wrought iron 

 may vary greatly with the conditions of exposure. 

 In view of the widespread belief that wrought iron 

 oxidizes less readily than soft steel, the author 

 puts forward a theory, based on the different pro 

 tective powers of the " cinder " or " scale " flakes 

 in wrought iron and the cementite flakes in soft 

 steel, to explain the discrepancy between the belief 

 and the results obtained in his experiments. 



The question of the most desirable form of steel 

 ingot for gun tubes and propeller shafts was dealt 

 with at the spring meeting of the British Iron 

 and Steel Institute in a paper by Mr. F. J. R. 

 Carulla. Although the circular form of ingot pos- 

 sesses many advantages, it has the serious draw- 

 back of being liable to develop skin cracks during 

 cooling. The author said that the octagonal ingot,, 

 said to have been used in the gun factories at 

 Woolwich, was a sounder form, but a hexagon wa8 

 better, and a square section the best. The form 

 of ingot which had come into use, but not so 

 generally as its merits deserved, was a hexagonal 

 section with concave sides, each of which offered 

 the resistance of the arch to pressures arising from 

 within, and which are set up by the expansion 

 of the metal in cooling. 



The Taylor- White tool-steel process in use at 



