1^22.] Method of analyzing the Ores of Nickel. 207 



remains some sulphur mixed with a httle sihca. The residue is 

 weighed, the sulphur is burnt, and the remaining silica is again 

 weig;hed. 



The nitric solution evaporated to dryness is again treated with 

 concentrated nitric acid to acidify the arsenic, and render the 

 oxide of iron insoluble,* which is afterwards separated. This 

 latter process is nevertheless absolutely incorrect ; for the arse- 

 niate of iron readily dissolves when the solution contains acid in 

 excess. 



The acid solution, nearly neutralized by an alkah, is precipi- 

 tated by nitrate of lead, which separates arseniate of lead ; but 

 as this compound is soluble in nitric acid, it must be evaporated 

 to dryness, and the dry mass treated with water. It is very true 

 that if the iron could be separated by the method above 

 described, we should succeed in separating the nickel from the 

 arsenic by means of nitrate of lead. Usually, however, the pre- 

 cipitate thus obtained contains arseniate of iron and arseniate of 

 lead, in the mixture of which it is impossible to calculate the 

 quantity of arsenic. Added to this, when an arseniuret is 

 dissolved in nitric acid without the addition of muriatic acid, a 

 great quantity of arsenic is converted into arsenious acid, and the 

 precipitate becomes a mixture of arseniate and arsenite of lead. 



From the solution which contains the nickel and the lead 

 added in excess, the latter is precipitated by sulphate of soda ; 

 afterwards sufficient ammonia is added to redissolve the oxide of 

 nickel, and the alumina and every other substance mixed with 

 the ore of nickel, which is insoluble in ammonia, is obtained. 

 The ammoniacal solution is to be evaporated, and the nickel pre- 

 cipitated by subcarbonate of soda or potash, taking care to 

 evaporate every trace of ammonia liberated by the carbonate. , 



The oxide of nickel, thus obtained, usually contains cobalt. 

 M.Thenard and Fourcroy attempted to separate those oxides by 

 peroxidizing them by means of oxymuriate of lime, and treating 

 the peroxides with ammonia, which decomposes and dissolve^ 

 the peroxide of nickel, but not the peroxide of cobalt. This 

 method, however, is not quite correct ; for the portion dissolved 

 contains a little cobalt, and the residuum contains nickel. 



Mr. Philhps discovered another method, and one which is 

 more proper to be employed in analyses. It consists in diluting 

 the ammoniacal solution of the two oxides with a considerable 

 quantity of water, after which a solution of potash is to be added 

 as long as precipitation takes place. The nickel is precipitated, 

 and the cobalt remains in the liquor from which the ammonia is 

 obtained by evaporation. This method is not rigorously exact, 

 but the traces of cobalt which are precipitated with the nickel 

 may be entirely neglected with respect to analytical result. 



M. Laugier afterwards discovered another method of preparing 



* Aikin's Dictionary of Chemistry, ii. 136, 



