368 IRON cnliar to steel ; but, properly considered, they belong to all compounds of iron and carbon. Wrought iron contains too little carbon to show much hardening when rapidly cooled from a high temperature, but it is generally rendered more rigid by such treatment. Cast iron becomes very hard and brittle on sudden cooling, but, since it is much more complex in composition than steel, the circumstances con- trolling the hardening are not so well under- stood. Oast iron which has been hardened may by a process of tempering be rendered soft again. The precise nature of the harden- ing process is not understood. It has been supposed to result from a chemical union of the carbon with the iron, formed at high tem- perature, and maintained under rapid, but re- solved by slow cooling. It has also been ascribed to a state of tension or polarity of the particles, which is relieved by tempering. In the case of cast iron a change in the condition of the carbon may be often observed ; some dark graphitic irons become perfectly white (chill) on sudden cooling. As to the character of the union of iron and carbon in cast iron, a dif- ference of opinion exists. Gurlt, Mayrhofer, Hahn, and others, have endeavored to estab- lish the existence of definite combinations of iron and carbon, such as FeO, Fe a C, Fe0, Fe 8 0, and suppose the different varieties of cast iron to be compounds or mixtures of these definite carburets with iron. The formula of spiegeleisen, in which the carbon is all com- bined, was supposed by Karsten to be expressed by FeC, which would require 5-8 per cent, carbon, but this amount is never found in re- ality. Gurlt proposed a lower carbide, Fe 8 C, which he supposed to stand in the same rela- tion to gray iron as Karsten's tetracarbide did to white iron. These formulas, although in- teresting and attractive in a theoretical point of view, must be regarded as purely imaginary. Isolated analyses may seem to indicate their existence, but extended investigations show that the variations of composition in cast iron are too great to admit of any definite formu- las. In the molten condition all the carbon is most probably combined with the iron. The separation of carbon as graphite takes place on cooling, and the amount separated is, other things being equal, determined by the rate of cooling. When we consider the number of factors that enter into the case, it is not sur- prising that we fail to detect any regularity in the composition of cast iron. Durre proposes a classification of cast iron based on physical characters. He considers all pig irons to be mixtures of two different substances, namely, graphite and a white or light gray matrix or ground mass. He recognizes three types of iron, represented by spiegeleisen, in which the ground mass forms bbld, brilliant, reed-like bundles of crystals; Swedish cannon iron, in which it appears as thin thread-like bundles ; and Scotch iron, in which it presents short in- terlaced figures, almost obscured by the graph- ite. The manifold properties possessed by iron in its various forms constitute its great value in the arts. No other metal or metallic combination possesses such a wide range of properties. The hardness and rigidity of pig iron, and the facility with which it can be cast into any desired form, adapt it to use in con- struction for the resistance of a crushing weight, and also to an infinite variety of uten- sils. The purer kinds often possess moreover great toughness, and are available for ordnance. Wrought iron, having a high degree of tenaci- ty and elasticity combined with malleability and ductility, is applicable to numberless uses in every-day life, particularly those which re- quire not only strength, but the ability to re- sist shock. Steel is stronger than wrought or cast iron, but is intermediate between the two in rigidity. It replaces wrought iron ad- vantageously in construction where strength is required in small bulk; but it is excluded, except in the softest varieties, where shocks are to be encountered. Its property of hard- ening, combined with malleability and duc- tility, adapts it for the manufacture of cutting tools. Until the comparatively recent intro- duction of the Bessemer process and the Sie- mens regenerative heating furnace (see FUR- NACE), it was impossible to melt wrought iron on the large scale; and the distinction be- tween wrought iron and cast steel was there- fore well marked in their physical characters, steel showing a homogeneous crystalline, and wrought iron a more or less fibrous structure, due to the intermingled cinder. This distinc- tion in physical characters disappears when soft iron (that is, iron with 0'25 per cent, or less of carbon) is melted and cast in moulds ; and the tendency of metallurgists at the pres- ent time is to call this product steel, without regard to its contents in carbon or its suscep- tibility to hardening. Bessemer and open- hearth (Martin) steels include products varying from hard steel to soft iron ; they have, how- ever, the common property of homogeneity, whence the name sometimes applied to them of " homogeneous metal." In both steel and wrought iron, therefore, the distinction is to be observed between welded and cast pro- ducts. (See STEEL.) Manufactured iron has thus far been considered in the present article merely as a compound of iron and carbon. It is generally, however, much more complex in composition, and we will now consider each kind separately in greater detail. I. CAST IEON. This is the product of the blast furnace (see IRON MANUFACTURE), and contains a number of elementary substances derived from the ore, flux, and fuel used in its production. The sub- stances most commonly met with (besides car- bon, which must be regarded as essential) are silicon, sulphur, phosphorus, manganese, and more rarely, or in smaller quantities, chro- mium, copper, nickel, cobalt, titanium, arsenic, antimony, aluminum, calcium, and magnesium. The following analyses will serve as examples :
