QUANTITATIVE SEPARATION OF LITHIUM. 9 



cent, or 0.0885 gram, of lithium cliloricl, ho recovered 17.65 per cent, 

 or 0.0862 gram, a deficiency of 0.0023 gram of Hthium clilorid. Again 

 from a similar mixture containing 67.2 per cent or 0.8200 gram of 

 lithium clilorid, he recovered 68.4 per cent, or 0.8340 gram, equivalent 

 to an excess of 0.0140 gram of lithium chlorid. While in judging 

 most methods, it is fairer to rate the results by the percentage of the 

 substance sought wliicli is recovered, in the case of a determination 

 with constant sources of error, such as the increased solubility of the 

 sodium and potassium clilorids due to moisture in the ether-alcohol 

 ami the failure to obtain complete extraction of small amounts of 

 litliium from large amounts of the other cldorids, the value of the 

 method for determining small amounts depends upon the absolute 

 amount of error. From this point of xiew, Smith's errors of 0.0140 

 gram, 0.0023 gram and 0.0021 gram of lithium cldorid show the 

 method to be very poorly suited to the determination of lithium in 

 natural waters which very rarely contain more than 0.01 gram per 

 liter. 



Mayer ^ noted that Liebig had found a carbonate left in the i)re- 

 cipitate formed by the use of phosphoric acid and sodium carbonate. 

 In view of the work of Rammelsberg and the fact that the phosphate 

 method was in such general use, Mayer made a study of the precipi- 

 tation under different conditions. He obtained only pure lithium 

 phosphate (LigPOJ by evaporating to dryness with sodium phos- 

 phate and enough sodium hydroxid to render the liquid alkaline. 

 If sodium carbonate were used in neutralizing the liquid, some 

 carbonate was found in the precipitate. He found it difficult to 

 wash out the last traces of sodium from large precipitates, but by 

 thorough washhig was always able to obtain a residue free from 

 sodium and consisting of tri-basic lithium phosphate (Li3P04). He 

 claimed that the three lithium phosphates of Rammelsberg were the 

 same with varying amounts of water. The thorough washing re- 

 quired by Mayer dissolved so much lithium that he found it necessary 

 to evaporate the filtrate and wash water and take up with water and 

 wash the precipitate. With careful work he obtained only about 

 1 .5 per cent of the total lithium precipitate in the second separation. 

 He washed with equal vohmies of water and ammonia, and noted 

 that in the presence of relatively large amounts of sodium and potas- 

 sium salts, the greater part of these must be removed before separat- 

 ing the lithium as phosphate, as the amount of water necessary to 

 wash out the alkali salts would dissolve too much lithium. He dis- 

 solved 0.110 gram of sodium chlorid in 100 grams of a mixtm-e of 96 

 per cent alcohol and 98 per cent ether, and noted that in tests made 

 by Rammelsberg 1.33 per cent of the sodium chlorid used went into 

 solution with the lithium chlorid. He also observed the tendency 



1 Ann. Chem. Pharm., 1856, 98: 193. 

 40061°— Bull. 153—12 2 



