W. A. Davis and J. A. Prescott 137 



Experiments with Sodium hydrogen 'phosphate, Na2HP04, I2H2O. 



The specimen of sodium hydrogen phosphate used had been carefully 

 recrystallised by Prescott^ in his work on the estimation of phosphoric 

 acid by the molybdate method. It had, however, slightly effloresced 

 so that when 1 grm. was dissolved in 200 c.c. of water 100 c.c. of the 

 solution contained 0-1096 grm. P2O5 instead of the 0-0990 grm. 

 corresponding to the formula Na2HP04, I2H2O. 



Several experiments were made with the salt, in each case 1-000 grm. 

 being fused with 8 grms. of pure ammonium fluoride ; the residue then 

 remaining was then gently ignited twice with sulphuric acid to expel 

 all the hydrogen fluoride, and dissolved in dilute nitric acid, and made 

 up to 200 c.c. The results obtained were variable, showing varying 

 losses of phosphoric acid. Thus, in the successive experiments, 100 c.c. 

 of the final solution contained 0-0989, 0-0886 and 0-0942 grm. P2O5, 

 i.e. 90-2, 80-9 and 86-0 % of the phosphoric acid originally present 

 in the salt. 



Experiments with Potassium dihydrogen phosphate, KH2PO4. 



With this salt, containing a smaller proportion of the alkali metal, 

 the loss of phosphoric acid during the treatment with ammonium 

 fluoride was considerably greater than with the disodium salt. 



0-500 grm. of the commercially pure (Kahlbaum's) salt dissolved 

 in 100 c.c. of water gave a solution containing 0-2539 grm. P2O5, as 

 estimated by the molybdate method. 



When 0-5000 grm. of the salt was fused with 4 grms. of pure 

 ammonium fluoride and treated as in the case of the sodium phosphate, 

 being finally made up to 100 c.c, quantities of the P2O5 varying between 

 0-1623 and 0-2078 grm. were found in the 100 c.c. Thus only 63-9 to 

 81-9 % of the phosphoric acid remained after the treatment w^ith fluoride. 



Experiments with Calcium phosphate. 



Kahlbaum's precipitated calcium phosphate was used. 0-5000 grm. 

 dissolved in 200 c.c. of dilute nitric acid gave a solution containing 

 0-2030 grm. P2O5, as estimated by the molybdate method. 



0-500 grm. of the same material was fused with 4 grms. of 

 ammonium fluoride, and after volatilising the excess of the latter, 

 ignited twice with sulphuric acid and the residue dissolved in dilute 



1 This Journal. 1914, 6, 111. 



