

Broxoning and Spencer — Separation of Ccesium, etc. 279 



Art. XXXI. — On the Separation of Ccesium and Rubidium 

 by the Fractional Crystallization of the Aluminium and 

 Iron Alums and its Application to the Extraction of these 

 Elements from their Mineral Sources ; by Philip E. 

 Browning and S. R. Spencer. 



[Contribution from the Kent Chemical Laboratory of Yale Univ.— cclxxxi.J 



Robinson and Hutchins* have recommended the crystalliza- 

 tion of the aluminium alums for the separation of caesium and 

 rubidium from potassium and lithium in lepidolite after the 

 decomposition of that mineral by fluorspar and sulphuric acid. 

 They have also called attention to the difference in solubility 

 between the caesium and the rubidium alumsf and have sug- 

 gested fractional crystallization for the separation of these ele- 

 ments. The marked difference in solubility between the potas- 

 sium alum and the alums of caesium and rubidium makes the 

 method quite satisfactory for the separation of potassium from 

 these rare alkalies, but the difference in solubility between the 

 alums of caesium and rubidium is not sufficiently great to bring 

 about a rapid separation of these elements. 



The work to be described was undertaken to obtain some 

 definite information as to the value of the process of fractional 

 crystallization when applied to the problem of separating the 

 alkalies. 



The process may be briefly described as follows: 



A solution obtained from the decomposition of lepidolite by 

 heating with fluorspar and sulphuric acid after the removal of 

 the calcium sulphate was evaporated until on standing the 

 mixed alums crystallized out. The mother liquor was poured 

 off into a second flask and this liquid was evaporated until 

 another crop of crystals was obtained and the new mother 

 liquor poured into a third flask, and so on. The crystals in 

 the first flask were dissolved in a small amount of water, by 

 warming, and again allowed to crystallize, the supernatant 

 liquid being poured into the second flask upon the crystals 

 which had formed there; these crystals, in turn, were dissolved 

 in this liquid and allowed to recrystallize ; and the process was 

 continued through all the series of flasks. The crystals separat- 

 ing in flask number one were repeatedly dissolved in fresh 

 water and allowed to recrystallize in this way, and the mother 

 liquor was kept moving along the series of flasks in succession. 



By this method the more insoluble alum was concentrated 

 at the upper end of the series while the more soluble alum 

 moved toward the lower end. 



* Amer. Cheni. Jour., vi, 74. 



fNote: 100 parts of water at 15-1 7" C. will dissolve 0'62 parts of caasium 

 alum, 2 - 3 parts of rubidium and 13 - 5 parts of potassium alum. 



