54 RESEARCHES UPON ATOMIC WEIGHTS. 



THE ANALYSIS OF MANGANOUS BROMIDE. 

 PURIFICATION OF MATERIALS. 



All of the water used in either the purification or the analyses was twice dis- 

 tilled, once from a dilute alkaline solution of potassium permanganate and then 

 from a very dilute sulphuric-acid solution. Block-tin condensers were used in 

 both distillations and the apparatus contained no rubber or cork connections. 

 The water was collected as a rule in Jena glass flasks, although for special pur- 

 poses either platinum or quartz receivers were substituted. 



Acids and ammonia also were distilled shortly before use, either platinum or 

 quartz condensers and receivers being employed when necessary. Solid reagents 

 were recrystallized, usually with centrifugal drainage. 



Special pains were taken in all the work to prevent the introduction of alka- 

 lies or silica into the purest materials, by avoiding as far as possible the use of 

 glass vessels. 



MANGANOUS BROMIDE. 



Four different specimens of manganous bromide were employed, which were 

 obtained from different sources and were purified in different ways. In the 

 case of Samples A and B, purification of the manganese from other heavy metals 

 was accomplished by recrystallization of Merck's "chemically pure" potassium 

 permanganate. Sample A was crystallized three times only, while Sample B 

 was thus treated ten times, the last two crops of crystals being thoroughly 

 freed from the mother-liquors by centrifugal drainage. 



In order to free the manganese from potassium and convert it into the bro- 

 mide, the following processes were employed with Sample A: First the perman- 

 ganate was dissolved in water and was reduced by passing sulphur dioxide into 

 the solution. This sulphur dioxide was made by heating copper turnings with 

 concentrated sulphuric acid and was purified from copper compounds mechani- 

 cally carried along by passing through three gas washing-bottles, each con- 

 taining a solution of sulphurous acid, and one column of beads moistened with 

 a similar solution. From the solution of potassium and manganous sulphates 

 the manganese was precipitated by the addition of an alkaline solution of am- 

 monium carbonate. The manganous carbonate was washed with water until 

 the washings were free from sulphates, then it was dissolved in nitric acid, 

 which had been redistilled until free from chlorine, and the manganous nitrate 

 was recrystallized six times from a solution strongly acid with nitric acid, four 

 times in a glass vessel, twice in platinum. 



Usually it was necessary to start crystallization by inoculation, and cooling 

 with ice was found advisable for the sake of economy in material. From a dilute 

 solution of the purified nitrate in a platinum vessel, the manganese was again 

 precipitated as carbonate, by means of ammonium carbonate which had been 

 freshly made by passing pure carbon dioxide into distilled ammonia in a plati- 



