A REVISION OF THE ATOMIC WEIGHT OF BROMINE. 57 



SYNTHESIS OF SILVER BROMIDE. 



The ratio of silver to silver bromide was determined as follows: Weighed 

 quantities of silver were dissolved in the purest redistilled nitric acid diluted 

 with an equal volume of water, in the dissolving flask described on page 12. 

 Next, the acid solution of the silver was diluted with an equal volume of water, 

 and was heated until free from nitrous acid and oxides of nitrogen. After still 

 further dilution, the solution was added slowly with constant agitation to a 

 dilute solution of an excess of ammonium bromide in a glass-stoppered precipi- 

 tating flask, and the whole was violently shaken for some time to promote coag- 

 ulation. By adding the silver solution to the bromide, occlusion of silver nitrate 

 was almost wholly precluded. In some experiments the solutions were as dilute 

 as twentieth normal, in others as concentrated as fourth normal. The final re- 

 sults seem to be independent of the concentration of the solutions. At the 

 end of about 24 hours the flask with its contents was again shaken, and then it 

 was allowed to stand until the supernatant liquid was perfectly clear. The pre- 

 cipitate of silver bromide was collected upon a weighed Gooch crucible, after 

 thorough washing by decantation with water, and was dried in an electric oven, 

 first for several hours at 130°, finally for about 14 hours at 180°. Then it w^as 

 cooled and weighed. 



The operations of precipitation and filtration were performed in a large cup- 

 board lighted with red light, and if the flask was taken out of this cupboard it 

 was enveloped in several thicknesses of black cloth. 



Even after the prolonged drying, traces of moisture were retained by the 

 salt, and could be expelled only by fusion. This was done by transferring the 

 bulk of the silver bromide, freed as completely as possible from asbestos, to a 

 small porcelain crucible which was weighed with its cover. The silver bromide 

 was then fused by heating the small crucible, contained in a large crucible to 

 prevent direct contact with the flame of the burner. A temperature much above 

 the fusing point of silver bromide was avoided so that volatilization of the salt 

 could not take place. This treatment must have eUminated occluded ammon- 

 ium salts as well as water. Finally, in order to convert any occluded silver ni- 

 trate, metallic silver, or silver sub-bromide into silver bromide, the salt was again 

 fused in a current of dry air containing bromine vapor. This treatment seldom 

 produced any measurable effect either upon the weight or the appearance of the 

 salt, which was perfectly transparent and of a light yellow color, even after the 

 first fusion in air. 



A few shreds of asbestos displaced from the crucible, together with an occa- 

 sional trace of silver bromide which escaped the crucible, were collected upon a 

 tiny filter paper, which was then ignited in a porcelain crucible. Before weighing, 

 the ash was either treated with a drop of nitric and hydrobromic acids and again 

 heated, or else was heated for some minutes in a current of air and bromine. 



The filtrate and washings were evaporated to small bulk. The precipitating 

 flask and all other glass vessels used in the analysis were rinsed with ammonia 



