SOME ASPECTS OF POWDER METALLURGY 425 



lum, molybdenum, osmium, and iridium which could have been treated 

 in much the same manner as in WoUaston's process for platinum. 



There were, however, instances where real effort was made to develop 

 useful' products by means of powder metallurgy. As early as 1870, the 

 fundamental idea of a self-lubricating bearing was disclosed in a patent 

 by Gwynn^ and was the prototype for a large number of later developments 

 in the field. To 99 parts of tin prepared by rasping or filing, one part of 

 petroleum still residue was added, and the mass heated and intimately 

 mixed. The mixture was then pressed to give the shape and solidity desired. 

 It was specifically stated by Gwynn that journal boxes made by this method 

 or lined with the material would allow shafts to run at high speed without 

 other lubrication'". 



There were a number of metal powder producers in the nineteenth 

 century, most of them producing flake powders, but a virtual monopoly 

 in the field was held by Sir Henry Bessemer from about 1840 to 1885, when 

 he retired from the business". The process was a secret one and remained 

 so for almost his entire business career, and the profits were so large that 

 they financed the development of the Bessemer process for making steel. 

 Essentially, the method was one of machining very fine filaments from 

 solid metal bars and passing the filaments through rolls to flatten and break 

 them into flat tabular particles. Precautions were taken to prevent sticking 

 and give a high polish to the powder by adding a very small amount of olive 

 oil. The powder was graded by means of an air blast in a tunnel about 

 40 feet long and 2| feet wide, the finest powder fraction being collected 

 in silk bags attached to the end of the tunnel. Bessemer's powder metals 

 included copper, and most of the common alloys of copper. 



Even with the relatively large scale production of flake metal powders 

 by Bessemer up to 1885, and the subsequent preparation of powder metals 

 by stamp mills which pulverized the metal by severe working, there was 

 very little actual commercial manufacture of solid compacts from powder 

 metals. 



The electric lamp industry provided the stimulus for further study in 

 the search for a metaUic filament to replace the carbon filament first used. 

 This culminated in the production of the tungsten filament'*^''- and indicated 

 the technique to be applied in the development of the other refractory 

 metals as well as the production of cemented carbides, electrical contacts, 

 and electrode materials. 



Even with the promise shown by this development and the production 

 of other ductile heavy metals, there was little other commercial activity 

 in powder metallurgy as late as 1915-1920. 



Various types of porous bearings had received sporadic attention, and, 

 in 1921, a new porous bronze bearing was described'^. The material was 



