Gooch and Phelps — Separation of Arsenic from Copper. 489 



any traces of the magnesia mixture left on the precipitate. 

 The residue in the crucible was dried at low temperature over 

 a Bunsen flame, and, when the ammonia had been driven off, 

 was ignited to the condition of the pyroarseniate, cooled in a 

 desiccator, and weighed. 



The magnesia mixture used in all this work was prepared 

 according to Blair's directions.* That is — a solution one hun- 

 dred and ten grams of the purest magnesium chloride of com- 

 merce containing water of crystallization was made in water and 

 filtered. Into this solution was poured a solution of fifty-six 

 grams of the commercially pure ammonium chloride in water 

 further purified by treating it at boiling temperature with bro- 

 mine water and then with ammonia in faint excess, and filter- 

 ing. This solution, diluted to about two liters, was made 

 ammoniacal by the addition of ten cubic centimeters of strong 

 ammonium hydroxide. It was filtered after standing, and 

 was filtered again each time as required for the experiments 

 given below in the tables. 



The solutions containing copper and arsenic from which the 

 arsenic was separated, were made by treating the boiling solu- 

 tion of pure copper sulphate required for each experiment 

 with ammonium hydroxide in such amount that all the copper 

 was kept in solution, then filtering on asbestos in a perforated 

 platinum crucible under pressure, pouring into a platinum dish 

 and adding the arseniate. 



In experiments (1) and (2) of Table I the solution of potas- 

 sium dihydrogen arseniate was added from a burette to the 

 ammoniacal solution of purified copper sulphate with the 

 magnesia mixture held in a platinum dish ; and, after stand- 

 ing about two hours, the precipitate was filtered off on asbestos 

 in a perforated platinum crucible under pressure, rinsed with 

 the least amount of distilled water made faintly ammoniacal, 

 20-50 cm3 , dried carefully over a Bunsen flame, ignited after 

 the ammonia was removed, cooled, and weighed. The results 

 shown by (3), (4), (5), (6), (7), and (8) were obtained in the 

 same way as in (1) and (2) excepting that the nitrations were 

 made as soon as the precipitates subsided. The first filtrate 

 was used to wash all traces of the precipitate into the cruci- 

 ble, and the precipitate was washed carefully, but not exces- 

 sively, with distilled water made faintly ammoniacal, and then 

 dried, ignited, and weighed. The results in (1), (2), and (3) 

 show that it is possible to gather with accuracy any traces of 

 arsenic present in solution. The presence of a compound of 

 copper in precipitates considerable in amount was shown in 

 some cases by dissolving the ignited pyroarseniate and getting 

 the characteristic blue upon addition of ammonia to the solu- 



* The Chemical Analysis of Iron, Blair, p. 59. 



