METALLURGY. 



485 



treatment with a mixture of sulphuric and nitric 

 acids in the proportion of 8 parts of sulphuric 

 to 1 of nitric acid. Sulphuric acid attacks silver 

 very slowly in the cold, and has no action on 

 copper; while nitric acid attacks silver instantly, 

 forming nitrate of silver, and acts violently on 

 copper, producing nitrate of copper and large 

 quantities of dioxide of nitrogen, which is con- 

 verted on contact with the air to the peroxide. 

 The mixture designated gives the desired effect 

 of dissolving the silver, while the copper is not 

 attacked. The coating of silver is converted into 

 the nitrate and sulphate; these salts are easily 

 precipitated, after 8 or 10 volumes of water have 

 been added, by a few drops of hydrochloric 

 acid. The chloride of silver is then separated 

 by agitation. The acid liquid is drawn away 

 by successive siphonages, and the pure salt is ob- 

 tained by nitration and washing. In case colored 

 spangles are the objects treated, the coloring 

 varnish is removed by immersion in a bath of 

 caustic soda or Roman potash at a temperature 

 of 45-50 C. The copper, not having been acted 

 upon, is recovered by simply withdrawing it from 

 the bath and washing. The silver is in the state 

 of a chloride, and is reduced to the metallic state 

 by treating with carbonate of soda or by passing 

 a current of hydrogen over it. 



Aluminum. Some very extensive deposits of 

 bauxite a hydrated oxide of aluminum used 

 largely for the extraction of the metal have been 

 discovered by officers of the Department of Mines 

 in New South Wales. Hitherto, its value not 

 having been recognized, the ore has been used 

 largely for road making. Three samples, how- 

 ever, recently analyzed were found to contain 

 respectively 58.31, 35.28, and 39.82 per cent, of 

 alumina, the first sample being richer than the 

 bauxite ores of France, Austria, or the United 

 States, which are at present the main sources 

 of supply. 



The results of experiments by M. A. Ditte on 

 the corrosion of aluminum by saline solutions 

 show that the metal is at first vigorously at- 

 tacked, but that a coherent protective layer of 

 alumina is soon formed. In presence of air, how- 

 ever, the corrosion goes on, and if an aluminum 

 plate has been immersed in a salt solution and 

 then only imperfectly washed the attack slowly 

 continues, the surface becoming more and more 

 easily affected by other reagents. 



Five thousand tons of aluminum electric con- 

 ductors were used on railways in the United 

 States in 1898, representing an outlay of $2,000,- 

 000. While the conductivity of aluminum is 

 inferior to that of copper, its weight is very much 

 less, and while it does not solder so well its 

 tensile strength is much greater, so that con- 

 siderably longer spans can be constructed be- 

 tween poles. 



Several explosions having been recorded within 

 .a few years past in factories where aluminum- 

 bronze powder is ground, M. Stockmeier has made 

 experiments with a view to ascertaining under 

 what circumstances explosions are possible. He 

 has found that the powder itself is stable both 

 to shock and while being ground. When mixed 

 with chlorate of potash it detonates when struck 

 or by simple rubbing. It can also be detonated 

 by means of an electric spark when shaken up 

 in a vessel containing air. Aluminum bronze is 

 capable of decomposing water; the quantity of 

 hydrogen produced varies under certain condi- 

 tions, such as the nature and quantity of the 

 grease always present. The bronze powder is 

 hygroscopic, and, according to the author's ex- 

 periments, it is capable when dry of absorbing as 



high as 1.40 per cent, of atmospheric moisture. 

 Thus in grinding up 5 or kilogrammes of bronze 

 powder at 1.4 per cent, of water we can produce 

 from 43.4 to 52.08 litres of hydrogen. Thin hydro- 

 gen forms with air a detonating mixture, which 

 may be exploded by a spark produced by the 

 passage of a stone or other foreign body through 

 the rollers. The most effective method of pre- 

 venting these explosions would be to replace the 

 air which comes in contact with the powder by 

 an inert gas, but this is impracticable. All risk 

 of explosions may, however, be avoided by ob- 

 serving certain precautions as to the dryness of 

 the powder, evenness of temperature, ventilation, 

 cleanliness, etc. 



Alloys. In a review of the results of a study 

 of some alloys with iron carbides, mainly man- 

 ganese and tungsten, M. S. de Benneville says 

 that liquid iron does not differ as a solvent from 

 other liquids for example, water except in the 

 comparatively greater complexity of its molecular 

 structure. On the cooling of such a mass con- 

 taining in combination other elements, accord- 

 ing to the conditions of such cooling and the 

 volume and chemical affinity (for the solvent) 

 of such elements, the solidified mass would 

 present a structure homogeneous throughout, 

 through intermediate forms, to a highly segre- 

 gated and crystalline condition, in which a num- 

 ber of definite compounds have been produced. 

 Such definite compounds in iron are present in 

 two forms: (1) Those common to alloys of wide 

 range of composition, Fe3C, and the prismatic 

 forms found in manganese and chromium car- 

 bides, phosphides, and sulphides; and (2) com- 

 pounds varying in composition from one alloy to 

 another, and dependent within narrow limits on 

 the conditions of cooling, and often on the im- 

 mediately surrounding magnesia from which the 

 constituents are drawn. Such are the granular 

 residues, differing only in degree from the ground 

 mass of the alloy. With the ground mass of the 

 alloy, directly comparable with vitreous struc- 

 ture of basalts, there is found a regular grada- 

 tion in iron alloys, and they are to be classed 

 according to the amount of differentiation under- 

 gone. Iron carbides tend to form homogeneous 

 masses, in which definite compounds are of minor 

 importance, the great bulk of the alloy entering 

 into the ground mass. Rapid cooling preserves 

 this structure. Slow cooling promotes crystalliza- 

 tion. Owing to the tendency to form ground 

 mass, iron carbides are more subject to control 

 even in presence of elements with which iron 

 forms inert compounds; if present in small quan- 

 tity, limits of homogeneity are comparatively 

 wide. To a considerable extent this is also the 

 case with ferro-manganese, but definite compounds 

 (of form 1) take on a more regular form, and 

 the granular masses constitute a considerable per- 

 centage of the total mass. In carbide with the 

 sixth family tungsten, molybdenum, and chro- 

 mium definite crystallized compounds are an im- 

 portant factor, and the dominant influence of the 

 added element is shown in the chemical inertia 

 of the granular residues sharply differentiating 

 them from the soluble portion to an extent not 

 found in iron carbide or ferro-manganese, in which 

 the separation chemically from the ground mass 

 is not marked. Such elements, when present in 

 large quantity, are less capable of control. 



In his fifth report to the Alloys Research Com- 

 mittee of the Institution of Mechanical Engineers 

 Sir William Roberts-Austen describes a long 

 series of investigations by means of cooling curves 

 of the carburized iron series of alloys, usually 

 known as steel and cast iron. He shows from 



