﻿114 Pen field and Harper — Precipitation and 



4. On the separation of aluminum and beryllium. 



"Various experiments convinced us that the following methods 

 are not satisfactory: pPrecipitation of the alumina as basic 

 acetate, some beryllium being also precipitated ; precipitation 

 of the alumina with barium carbonate ; solution of the beryl- 

 lium in a boiling solution of ammonium chloride; by long boil- 

 ing of a solution of ammonium chloride the solution becomes 

 slightly acid and according to our own experience some alumina 

 is dissolved, while Genth* found that the beryllium was not 

 completely dissolved. Solution of the beryllium in and pre- 

 cipitation of the alumina with ammonium carbonate gives un- 

 reliable results, and the method, when carried out as described 

 by one of us in making a series of beryl analyses,! gave too 

 low results for beryllium. 



We have made a large number of separations by dissolving 

 the mixed chlorides in the least possible excess of caustic soda, 

 diluting the solution largely and precipitating the beryllium 

 by boiling. This method has recently been proposed by 

 (xenth;j; as the most reliable of the known methods, and Baker§ 

 used it successfully in his analysis of beryl. Before using the 

 method in our own work we made a series of experiments to 

 determine its accuracy and to find out the best conditions for 

 making the separation. A standard solution of beryllium 

 chloride was made from purified K 2 BeF 4 containing -1107 

 grams BeO in every 50 c.c. Our standard solution of alumi- 

 num chloride contained -1002 grams A1 2 3 in every 50 c.c. 

 Neither of the solutions contained any alkali and not more 

 than traces of sulphuric acid. 



The method was as follows: 50 c.c. of each solution were 

 measured into a platinum dish and evaporated to dryness, the 

 chlorides were dissolved in the least possible quantity of water, 

 and a rather strong solution of pure soda, made from metallic 

 sodium, was cautiously added till the precipitate which at first 

 formed was completely dissolved. The contents of the dish 

 were then rinsed with cold water into a beaker containing about 

 800 c.c. of boiling water and the contents of the beaker boiled 

 for one hour, replacing from time to time the water which 

 evaporated. The beryllium separates out as a granular pre- 

 cipitate which is easy to filter and wash. After acidifying and 

 concentrating the filtrate the alumina was precipitated with 

 ammonia. Our experience has taught us that the best results 

 are obtained by dissolving the dried chlorides in the least possi- 

 ble quantity of water and using as little soda as possible for 

 dissolving the aluminum and beryllium. 



Below we have given in tabulated form the results of our 

 experiments. 



*Proc. Amer. Phil. Society, xxi, 694, 1884. f Loc. cit. 



\ Loc. cit. § American Chem. Journal, vii, 115. 



