ON THE ANALYSIS OF IRON AND STEEL. 



189 



combustion of the carbonaceous residue in a stream of oxygen. Cupric 

 chloride was first prepared by roasting copper scale in a muffle, dissolving 

 in pure hydrochloric acid, and passing chlorine into the solution to 

 convert the whole into the cupric condition. To a concentrated solution 

 of the cupric chloride so prepared, 10 per cent, of sodium chloride 

 was added, and the solution so obtained, which is distinctly acid, is 

 employed for dissolving the steel. The separated carbon was collected 

 on an asbestos filter, and burned in a glass tube with pure recently 

 ignited oxide of copper. 



Sulphur and FhospJiorus. — The steel was oxidised by pure nitric acid 

 (sp. gr. 1'40), redistilled in Mr. Riley's laboratory, evaporated to 

 di'yness, heated, dissolved in pure hydrochloric acid (also redistilled in 

 the laboratory), and the silica separated; the excess of hydrochloric acid 

 evaporated off until only just sufficient acid was left to keep the iron in 

 solution. The sulphur was precipitated by barium chloride, allowed to 

 stand twenty-four hours, and weighed as BaS04. The filtrate from the 

 determination of sulphur was reduced by addition of sulphide of sodium, 

 the precipitated barium sulphate was filtered off, and the phosphorus 

 precipitated as basic perphosphate of iron by boiling with ammonium 

 acetate. If there is not sufficient ferric oxide present, a little bromine is 

 added. The precipitate is dissolved in hydrochloric acid, citric acid is 

 added, then excess of ammonia and a small quantity of ' magnesia mix- 

 ture.' After standing twenty- four hours the precipitate is filtered off and 

 weighed as Mg,2P207. 



Manganese. — The iron was separated as basic ferric acetate, the 

 manganese precipitated by bromine and weighed as Mn304. 



Silicon. — The silica separated in the determination of sulphur and 

 phosphorus was washed, ignited, and weighed ; it was afterwards treated 

 with hydrofluoric acid and any residue weighed and deducted. 



Mr. Stead reported as follows : — - 



Mr. Stead has supplied the following brief outline of the methods he 

 adopted, and which are almost identical with those employed by Mr. 

 Pattinson : — 



Carbon. — The sample was dissolved in double chloride of copper and 

 potassium, the carbon separated by filtration through asbestos, and after 

 careful washing burned in a porcelain tube with copper oxide in air or 

 oxygen. 



Silicon. — By solution in hydrochloric acid, evaporation to dryness, 

 and strongly heating ; redissolving in hydrochloric acid and water, 

 filtering and weighing as SiOa- 



Sulphur. — This was determined both by oxidation with aqua regia 

 and also by the evolution method ; no further details were supplied. 



Phosphorus was precipitated and weighed as phospho-molybdate of 

 ammonia. 



Manganese was estimated by the acetate and bromine gravimetric 

 process, and also by Pattinson's volumetric ii-.3thod. 



