

METALLURGY. 



433 



which deposit on it and produce oxidation, while 

 it is difficult to insure the absence of such 

 metals. In Dieffenbach's process, which is in suc- 

 cessful operation at a German establishment, a solu- 

 tion of zinc chloride, obtained by leaching a zinc- 

 iferous iron pyrites after submitting it to chlorinat- 

 ing roasting, is electrolyzed. The Ashcroft process 

 obtains coherent zinc by employing a somewhat 

 basic solution of zinc sulphate or chloride in the 

 cathode compartments of the electric cells. The 

 Siemens and Ilalske process employs a somewhat 

 acid zinc-sulphate solution. An almost chemically 

 pure zinc was obtained at Tarnowitz from an alloy 

 of zinc and silver with a little lead and copper, pro- 

 cured by desilverizing lead with zinc containing 

 about 0.5 per cent, aluminum, refined electrolyt- 

 ically. 



In introducing his new apparatus for the electro- 

 lytic production of zinc, Dr. Carl Hoepfner, of 

 Giessen, Germany, affirms that zinc has not hitherto 

 been recovered electrolytically on a commercial 

 scale from zinc-chloride solution, owing to the 

 nature of the cathodes used. Such solutions, he 

 says, can be economically electrolyzed only by the 

 use of discoidal rotating cathodes when partly im- 

 mersed in the electrolyte, and the zinc is obtained 

 in a homogeneous pure solid condition. With 

 such cathodes currents of high density can be used 

 when it is desired to produce chlorine together with 

 zinc. 



Attempts to deal on a commercial scale with 

 complex refractory ores, such as those existing at 

 Broken Hill, which contain on an average about 25 

 per cent, of lead, 20 per cent, of zinc, and 20 ounces 

 of silver per ton, have been abandoned on account 

 of the great difficulties met in practice. It is 

 claimed that the Cowper-Coles electrolytic process 

 overcomes all these difficulties, and has other sub- 

 stantial advantages, as regards economical working 

 and the purity of the products obtained. The ore, 

 after being crushed dry and roasted, is screened, so 

 as to remove the very fine dust or slimes, which are 

 treated by a special process in separate vats, while 

 the screened ores are placed in vats provided with 

 false bottoms or filter beds. A weak solution of 

 sulphuric acid, which maybe made from the sul- 

 phur dioxide driven off from the ore during the 

 process of roasting, is employed for leaching off the 

 zinc. The leaching solution, after being drawn off 

 from the operation tanks to remove any copper 

 that may be present, is circulated through a series 

 of electrolyzing colls, where the zinc is deposited in 

 the form of metallic plates on an aluminum cathode, 

 or in the form of zinc sponge. By using a suitable 

 strength of solution only the zinc and copper are 

 removed from the ore, while the other metals pres- 

 ent are not dissolved. The residue, after the /inc 

 has been leached out within a few per cent., is washed 

 to remove the soluble zinc sulphate. The lead and 

 silver remaining in the residue can be extracted by 

 any of the well-known method?. 



Alloys. The mechanical defectiveness and con- 

 sequent electrical instability of alloys used for elec- 

 trical wires are discussed by Mr. Rollo Appleyard 

 with regard to the constitution and metallurgy of 

 the alloy, and the subsequent treatment and en- 

 vironment of the wire. Instances are given of the 

 failure of German silver and platinoid wire that 

 have occurred among several thousand resistance 

 coils distributed over widely different latitudes. 

 In periods of time, varying from six weeks to sev- 

 eral years after manufacture, some of the bobbins 

 became brittle and bulky, both on the outer layers 

 and within the coils. The towns where the faults 

 appeared are all within the tropics, and are in- 

 cluded nearly within the isotherm, of 25" C. Other 

 coils of nominally the same material, manufacture, 

 VOL. XXXVIIT. 28 A 



and environment have retained their original good 

 condition. It follows that metallurgical differences 

 exist between different samples of the same nominal 

 quality of alloy. Examples are given to prove that 

 failure sometimes occurs with platinoid through 

 which no electricity has passed. Provided that the 

 wire is good, the effect of environment is almost in- 

 significantthat is, the question is one of metal- 

 lurgy rather than of instrument-making. 



The experiments of Prof. Guilluumc show that, 

 the higher alloys of nickel with steel up to 50 per 

 cent, of nickel are subject to very slight expansion 

 by heat, and are therefore suitable for the construc- 

 tion of standards of length. An alloy containing 

 36 per cent, of nickel showed an extremely small 

 variation in length under wide changes of tempera- 

 ture. The hardness and strength of this alloy, and 

 the fact that it readily receives a high polish and 

 is little subject to rust, add to its suitability for 

 the purposes indicated. The same property 

 renders it eminently suited for the construction of 

 pendulum rods forelocks, as the variation in length 

 due to changes of atmospheric temperature would 

 be almost inappreciable. It is possible that bars of 

 this alloy may prove satisfactory for measuring 

 rods for geodetic base measurements, thus dispens- 

 ing with the use of the troublesome compensating 

 compound bars, or with the device of employing a 

 bar packed in ice. 



The use of coins made of an alloy of copper and 

 nickel in two of the French colonies has made ne- 

 cessary at the mint a rapid and exact method for the 

 analysis of these coins. They are not of the same 

 composition, one containing 25 and the other only 

 15 per cent, of nickel. They being binary, it is only 

 necessary to determine one of the metal's the cop"- 

 per, for instance. But as the nickel is the more ex- 

 pensive, it is preferable to estimate it directly 

 rather than to deduce its weight by difference. In 

 the process a gramme of the alloy is attacked with 

 the smallest quantity possible of nitric acid on a 

 sand bath, after which the directions are : Add a 

 little water containing five or six drops of sulphuric 

 acid and evaporate to dryness. Take up with 

 water still containing a little sulphuric acid, and 

 again evaporate to dryness to make sure that all 

 the nitric acid is driven off ; redissolve in water 

 with a little sulphuric acid, and pour into an elec- 

 trolytic crucible, filling it about one third full. 

 The copper alone is deposited. The remaining so- 

 lution is saturated with ammonia and electrolyzed 

 with three Paniell cells. This brings down' the 

 nickel, and the whole process gives very concordant 

 results. The difference between the total weights 

 found and 100 represents the impurities consist- 

 ing of sesquioxide of iron and oxides of manganese 

 and aluminum in the niekel, 



In experimenting with Babbit's antifriction 

 metal, 82 Sn. 9 Sb, 9 Cu, Prof. H. Behrens and Mr. 

 II. Bancke found that by slow cooling the alloy 

 was split up into compounds of different fusibility. 

 By pressure between the plates a metallic mother 

 liquid was squeezed out. The remaining cakes of 

 crystalline metal were treated with hydrochloric 

 acid and washed with water. An alloy, containing 

 90 Sn, 10 Sb, on being thus treated, yielded the 

 same cubic crystals as Babbit's metal, which were 

 found to answer to the formula SbSn. With 42 

 Sb prismatic crystals of the compound SbSn were 

 obtained. In Babbit's metal the copper forms brit- 

 tle needles of whitish bronze containing no anti- 

 mony. Such bronzes show less stability than the 

 compounds of tin and antimony. From an alloy of 

 90 Sn, 10 Cu, the compound CuSn was obtained. 

 Repeated healing and cooling brought the copper 

 percentage up from 35 to 58. Miscroscopical exam- 

 ination of bearings showed that cushions heated by 



