December 14, 1888.] 



SCIENCE. 



285 



notwithstanding they require several days' preparation in the lathe, 

 •owing to their hardness, yet, under a compression load of lOO tons 

 per square inch, shorten .25 inch, and the harder kind (manganese, 

 15 to 20 per cent) . I inch to .13 inch. Chilled iron or hardened 

 steel would stand this test without any alteration. A cast speci- 

 men, No. 24 B (manganese, 14.75 pef cent), not forged, made into 

 a standard Whitworth test-piece, took nearly a fortnight to tool and 

 finish. 



The test-bar of the specimen (12.75 per cent) tested by Mr. 

 Wellman, of the Otis Iron and Steel Company, Cleveland, O., was 

 a day and a half in the lathe, as against half an hour for ordinary 

 mild steel. 



Owing to this peculiar hardness, its general application to cast- 

 ings has been limited by the difficulty of inachining them, no 

 method having yet been perfected by which the heads or runners 

 can be cut off, or the castings otherwise tooled to shape. Tool 

 steels from the best makers have been tried, including self-harden- 

 ing kinds and tools made of manganese steel, but without success. 

 The tests and applications, therefore, have necessarily been confined 

 to castings, where the runners could be broken off cold or pared 

 off hot. 



Water-Quenching, and the Effect of Heat upon Manganese 

 Steel. 



Naturally one of the questions asked, when examining a new 

 material said to be steel, or possessing the properties of steel, is, 

 'What effect has water or other cooling medium on it when plunged 

 therein in a heated condition ; in other words, will it harden ? ' 

 Again : the behavior was found to be quite different when com- 

 pared with ordinary carbon steel, no hardening action taking place. 

 Water certainly causes the material to become stiffer, but in an en- 

 tirely different degree to hardened carbon steel ; for a piece of man- 

 ganese steel, after such treatment, is slightly more easily touched 

 by a file : therefore, for the following reasons, the process now de- 

 scribed is termed ' water-toughening.' The increase in stiffness is 

 most marked, the tensile strength rising from 40 to 60, and in 

 some cases over 70, tons per square inch ; but this is not a mere 

 stiffening or hardening effect in the ordinary sense of the term, for 

 in all carbon steel such rise is invariably accompanied, when the 

 cooling medium is water, by a considerable decrease in the ductility 

 or elongation, whereas in this material just the opposite effect is 

 produced. In specimens Nos. 22 B and D^ the tensile strength of 

 the bar as received from the forge was only 36 tons per square inch, 

 with 1.56 per cent elongation. This latter is exceptionally low, 

 usually being 6 to 8 per cent. After water-toughening, it rose to 

 the extraordinary amount of 67 tons, with 44.44 per cent elongation; 

 and even then the specimen was not fractured, as at this point it 

 was considered worthy of being retained unbroken. The same re- 

 sult occurs if the piece under treatment be dipped when at a weld- 

 ing heat, though the carbon be as high as i per cent or more. With 

 regard to those samples containing below about 7 per cent man- 

 ganese, this treatment seems to exercise little or no influence, and 

 the material is comparatively valueless where toughness is requi- 

 site. While touching upon this point, the results obtained by the 

 Terre Noire Company of France with high manganese steel (1.75 

 to 2.25 per cent) should be referred to. It is stated that it was not 

 possible to obtain test-bars when dipped in water or oil, as they 

 either cracked or broke into pieces. Strange to say, not a single 

 bar in these experiments has behaved so. Take, for example, No. 

 4 B, with 6.95 per cent of manganese, which may be termed com- 

 paratively low, and more approaching to the Terre Noire material : 

 the test-bar, when heated to a white heat, could be safely plunged 

 into either water or oil without being water-cracked. 



After a large number of tests with regard to the action of heat 

 and sudden cooling upon this material, generally speaking, it has 

 been found that the higher the heat of the piece treated, and the 

 more sudden and rapid the cooling, the higher will be the breaking 

 load, and the greater the toughness or elongation. Six of the bars 

 were heated as uniformly as possible to a yellow heat, and plunged 

 into water of 72^ F. These gave breaking loads varying from 57 

 to 63 tons per square inch, and elongations of 39.S per cent to 50 

 per cent. As a comparative test, another test-bar of the same 

 material, heated in precisely the same way and to the same de- 



gree, but plunged into water at a temperature of 202° F., gave 

 only 53 tons and 32.8 per cent. The more rapid cooling of the 

 other test-bars was evidently the cause of their superiority, the 

 chemical composition of all being the same. 



It was also thought that sulphuric acid, being a rapid conductor 

 of heat, might give good results as a cooling medium. The ex- 

 periment was therefore made with a bath consisting of equal vol- 

 umes of water and of sulphuric acid, and on 8 inches the extraor- 

 dinary elongation of 50.7 per cent was reached with a breaking 

 load of 65 tons, the bar being thus drawn cold 4^,y inches before 

 fracture. Another specimen on a 4-inch length gave 56.75 percent. 

 The operation of merely heating the forged test-bar to a yellow 

 heat and cooling it in air has a very beneficial effect, the elon- 

 gation in most instances being increased to 15 and 20 per cent, the 

 tensile strength also rising 8 or 10 tons per square inch. 



As before pointed out, the temperature to which the bar is sub- 

 jected has a marked influence. Although good tests result when 

 the specimens are treated at lower temperatures, the best are ob- 

 tained with as high a temperature as possible, the bars being thor- 

 oughly soaked, and plunged into cold water. Care, of course, 

 must be taken that they are not burnt, or heated beyond a welding 

 heat. In those specimens where the alloy is not so pure a mixture 

 of iron and manganese, and the material cannot be heated so hot 

 without crumbling, lower temperatures also give good results, viz., 

 40 to 46 per cent elongation. The best tests have been obtained 

 with material containing 12 to 14 per cent of manganese, though 

 those with 10.83 P^'' cent are also good, considering their high 

 breaking loads as compared with mdd steel. However, special at- 

 tention is drawn to the peculiar fact that an increase of 4 per cent 

 in the manganese causes such a considerable rise, both in tenacity 

 and elongation. The cause of this is very obscure, the only ex- 

 planation offered being that the peculiar crystallization in the cast 

 ingots seems to disappear gradually after passing about 1 1 per 

 cent, and the fibre noticed is not so much a cause of weakness. 

 This is only surmise, as to the eye the fibre in even the lower per- 

 centages entirely disappears in the hammered bar. 



It is not easy to understand the action of the water-quenching 

 process. As so ably explained by Chernoff, the effect of oil-tem- 

 pering on ordinary steel is to produce a metal of fine grain, which 

 possesses much greater strength than open, coarse-grained steel. 

 If, however, forged manganese steel possesses any real difference 

 of structure, after being heated and water-toughened, it is rather in 

 the direction of a more open than a closer grain. But the most 

 puzzling case in the author's experience is that of the cast-tough- 

 ened 9-per-cent specimens, at which percentage, as before pointed 

 out, the crystallization is very peculiar. An ingot 2i inches square 

 and 2 feet long was cast in an iron mould. When cold, a piece 

 was broken off, requiring four blows under a steam-hammer. The 

 fracture showed the usual peculiar form of the 9-per-cent material, 

 — a form which, to outward appearance, is unchanged by any heats 

 short of the actual melting-point. The other piece was reheated 

 to a yellow heat, and water-quenched. In this the toughness was 

 increased in a remarkable manner, ten blows of the steam-hammer 

 being required to break the bar. The appearance of fracture was 

 unchanged. What caused the increase of toughness ? In this 

 case, certainly, it was not owing to structural changes, the pro- 

 nounced form of ingot not being to the eye in any way altered. It 

 will therefore be understood how difficult it is to offer any satisfac- 

 tor\' explanation of these peculiarities. 



Considering the effects of water-toughening, special attention is 

 drawn to a specimen containing, carbon, 1.S5 per cent : manganese, 

 9.42 per cent. Ordinary steel with this amount of carbon would be 

 excessively hard if water-quenched even at a dull-red heat : in fact, 

 it is questionable whether it could be hardened at all without being 

 water-cracked. Yet the above specimen was heated to a high heat, 

 plunged into cold water, and the bar was not water-cracked, and, 

 if changed at all, slightly softer. Carbon seems, therefore, entirely 

 deprived of its usual hardening properties, and it is probable that 

 manganese must be partly considered as the cause of the high ten- 

 sile strength of this material, that is, unless iron itself possesses the 

 property of taking some other form not hitherto suspected. Fur- 

 ther, iron so combined with manganese is rendered capable of 

 elongating 50 per cent on 8 inches, against about 30 per cent in 



