b THE DETERMINATION OF LITHIUM. 



QUANTITATIVE SEPARATION OF LITHIUM. 

 REVIEW OF METHODS DESCRIBED IN THE LITERATURE. 



Before the appearance of Gooch's article in 1887, the method most 

 used by chemists was that of BerzeHus ^ modified by Mayer.^ Ber- 

 zeHus added phosphoric acid and sodium carbonate to the mixed 

 alkali chlorids and evaporated to dryness. On taking up wdth cold 

 water, a white powder was left behind whic'i was insoluble in tlie 

 sodium phosphate mother liquor and could be washed with cold 

 water. This precipitate he considered a double phosphate of sodium 

 and lithium wdth equivalent amounts of each metal. One analysis 

 of a precipitate showed very nearty this relation. 



Rammelsberg ^ found that the filtrates and wash waters from the 

 phosphate precipitate of Berzelius contained lithium on account of 

 the solubility of the lithium phosphate. He also found the sodium 

 to vary from 7.845 to 28.38 per cent and the lithium from 34.36 to 

 21.89 per cent accorcUng to the variations in the amounts of sodium 

 and litliium in the solution from which the precipitate was formed. 

 He concluded therefore that the phosphate separation was unreliable 

 and suggested the use of ether and alcohol to make a quantitative 

 separation. He extracted the dried mixed chlorids with a mixture 

 of equal parts of ether and absolute alcohol. This solvent cUssolved 

 the lithium chlorid and left the sodium and potassium chlorids. 

 After evaporation of the solvent, the lithium could be converted into 

 sulphate for weighing. In a further study * of the phosphates of 

 lithium he found varying proportions of lithium oxid and phosphoric 

 acid with different methods of formation of the precipitate. 



J. Lawrence Smith, ^ writing on the determination of the alkalis, 

 gave the method of Rammelsberg as affording a satisfactory separa- 

 tion and reported some results which do not show as clean separation 

 as might be desired. Wliile the results are not very clearly expressed, 

 the calculation b}^ Gooch ^ seems to be the only one that will show 

 just what quantities of lithium were used and recovered. AccorcUng 

 to these figures, Smith, by treating 0.5 gram of potassium chlorid 

 with 10 grams of ether- alcohol mixture chssolved 0.0003 gram of the 

 salt and from 0.5 gram of sodium chlorid dissolved 0.0005 gram with 

 the same amount of ether-alcohol, giving a total weight of 0.0008 

 gram dissolved from the two salts. From a mixture of 0.2 gram each 

 of potassium clilorid and sodium chlorid and 0.0080 gram of lithium 

 chlorid, he recovered 0.0101 gram, presumably lithium chlorid, an 

 excess of 0.0021 gram. From a "similar mixture," supposedly 0.2 

 gram each of potassium cldorid and sodium clilorid, with 18.1 per 



1 Pogg. Ann., 1825 4-- 245" * Pogg. Ann., 1849, 76: 261-270. 



2 Ann. Chem. Pharm., 1856, 98: 193. ' Amer. J. Sci., 1853 (2) 16: 56. 



3 Pogg. Ann., 1845, 66: 79. « Loc. cit. 



