12 THE DETERMINATION OF LITHIUM. 



chlorid and sodium pyrophosphate. By varying the proportion of 

 sodium pyrophosphate to htliium chlorid from 20 to 1, to 2 to 1, the 

 percentage of Htliium oxid in the precipitate varied from 10 per cent 

 to 24 per cent. One part of lithium chlorid to 10 parts of crystal- 

 lized sodium pyrophosphate gave a precipitate of the composition 

 found by Berzelius. 



Merling ^ made aifalyses of different phosphates of lithium and 

 was able to obtam figures correspondmg to the fornuilas LijPO^, 

 LiPOg, and Li^PjO;. He found in usmg the j)hosphate method of 

 separation as used by Fresenius ^ that the third evaporation of the 

 filtrate gave liim only 0.6 mg of insoluble phosphate. His total 

 precipitate contained 6.8 mg of silica, which, subtracted from the 

 amount of phosphate, left exactly the theoretical amount to be 

 obtained from the lithium carbonate he started ^\•ith, that is 104.53 

 per cent instead of 104.50 per cent. 



GixTL 3 separated the Uthium from the greater part of the other 

 alkali salts by extraction with strong hydrochloric acid until the 

 residue was free from lithium. He then evaporated the solution 

 and repeated the extraction on the new residue until the amount of 

 other alkali chlorids was small enough to permit the separation of 

 the lithium either as phosphate or by ether-alcohol. 



GoocH,* reviewing the methods of Mayer and Rammelsberg and 

 finding neither completely satisfactory, made the separation of the 

 lithium chlorid from the chlorids of sodium and potassium by means 

 of amyl alcohol. In his test experiments he separated from 0.01 to 

 0.1 gram of lithium chlorid from 0.1 or 0.2 gram of sodium or potas- 

 sium chlorids. Haywood,^ before pubHsliuig the Gooch method as 

 a regular procedure for water analysis, verified the results obtained 

 by Gooch and determmed the value of the method in the se])aration 

 of the small amounts of lithium usually found mixed ^^^th large 

 amounts of other salts in mineral waters. 



Carnot ^ made a separation based on the differing solubihties of 

 the alkahne fluorids. One part of lithium fluorid he found soluble 

 in 800 parts of water, 1 part in 1,900 parts of a mixture of equal 

 volumes of ammonia and water, and 1 part in 3,500 parts of a 

 strongly ammoniacal solution of ammonium fluorid. It took, how- 

 ever, 70 parts of this reagent to dissolve 1 part of sodium fluorid. 

 In the application of this method a solution of ammonium fluorid 

 With an excess of ammonia was added to the concentrated solution 

 of the mixed chlorids. After standing, the precipitate was washed 



1 Zts. anal. Chem., 1879, 18: 563. 



2 Loc. cit. 



' Ber. osterreich. Ges. chem. Ind., ISSO, p. lO.. 



< Amer. Chem. J., 18S7, 9: 33-51. 



6 U. S. Dept. Agr., Bureau of Chemistry Bui. 91, p. 29. 



"Bull. soc. chim. Par., 1889, (3) 1: 280. 



