president's address — SECTION B. 99 



going into bottoms are brought into direct contact with the blast in 

 the agitated whirl within the vessel, while in the reverberatory furnace 

 they are shielded from the influence of the air by the cover of the matte 

 and slag and by their affinity for the metal. That most injurious of 

 all impurities, bismuth, is removed by volatilisation in both cases ; 

 but the reverberatory furnace gets rid of scarcely more than one-half 

 '(54 per cent.), while the converter drives out practically all (9+ per 

 cent, to 96 per cent.), only little remaining in the copper and less in 

 the slags. Lead is equally removed by both methods, all of it vanish- 

 ing (99 per cent.), but arsenic is largely volatilised in the vessel (73 per 

 cent, to 91 per cent.), while, in the reverberatory furnace, it goes into 

 the bottoms, imless eliminated in the calcinations (21 per cent, driven 

 off). Antimony goes into bottoms also (total eliininated, 50 per cent.), 

 and in tlie converter rather more is removed (62 per cent, to 73 per 

 ■cent.). Selenium and tellurium are very largely eliminated in the 

 rev^erberatory furnace ; in the converter much the same (57 per cent, 

 to 71 per cent.). To repeat, as far as the more common impurities are 

 ■concerned, the old and the new methods stand on the same level as 

 regards lead ; the old is inferior wiuh respect to antimony, very much 

 less capable as regards bismuth, and enormously behind as far as 

 arsenic is concerned. (r) 



As against the enthusiasm which the converter adept is apt to feel 

 over the superiority of his process in this important connection, the 

 savants of the Welsh process have entered certain quiet protests. As 

 expressed by Mi. Allan Gibb,(t<;) the degree of elimination of any im- 

 purity in the converter depends upon the proportion originally present 

 and upon the proportion of other impurities present, which behaviour is 

 analogous to that of the impurities in the case of furnace-bottoms. 

 As, in both instances, the commencement is matte and the end blister, 

 and as there is only one chemical element to do the oxidising, and, 

 moreover, since the concentrated energ}' of the converter only replaces 

 the attenuated energy of the other method, it follows that, whatever 

 the intermediate steps may be, the final results of both methods must, 

 on the whole, be much the same, and this fact is borne out by his ob- 

 servations. Making a comparison of the total roaster operation with 

 the action of the converter, Mr. Gibb shows that there is not much 

 difference in tlie derjree of elimination, but a very marlced difference in 

 the manner. Bismuth, not forming salts with cuprous oxide, is removed 

 almost wholly by volatilisation in both processes ; but arsenic and 

 antimony, for their part, wliile chiefly volatilised in the converter, are 

 both principally slagged in the reverberatory furnace. Mr. Gibb 

 contends that the difierence of action is less due to distinctions of a 

 physico-therraal character than to chemical ones, viz., to the com- 

 position of the respective slags. The Bessemer slags having only 1 per 

 cent, to 2 per cent, copper, and the roaster slags containing from 6 per 

 cent, up to 40 per cent, of copper, simply in consequence of the longer 



iv) E. Keller, Trans. Am. Inst. Min. Eng., 1898, vol. XXVIII., p. 127, and later 



papers, 

 (w) Trans. Am. Inst. Min. Eng., 1903, vol. XXXIII., p. 653, and other papers. 



