NOTE ON THE LOSS OF PHOSPHORIC ACID 

 DURING FUSION WITH AMMONIUM FLUORIDE. 



By WILLIAM A. DAVIS and JAMES ARTHUR PRESCOTT. 



{Rothamsted Experimental Station.) 



In the analysis of soils and rocks, treatment with hydrofluoric acid 

 or ammonium fluoride is frequently prescribed for the estimation of 

 constituents other than silica (compare for example Wiley's Agricultural 

 Analysis, Vol. i, p. 352). Recent results obtained by the writers show 

 that this method fails to give reliable results in the case of phosphoric 

 acid owing to a loss of this constituent during the volatilisation of the 

 silica as siUcon tetrafluoride. The exact cause of the loss was not 

 investigated, but apparently in presence of an excess of ammonium 

 fluoride or hydrofluoric acid some of the phosphorus may volatilise 

 in the form of phosphorus trifluoride or phosphorus pentafluoride. 

 The loss appears to be least with phosphates of the alkali metals, the 

 highly basic nature of the metal enabling it better to retain the phosphate 

 radicle. The loss from phosphates of the alkahne earths (hme) during 

 treatment with the fluoride is considerably greater and it is very marked, 

 and may rise to over 50 % in the case of minerals such as apatite. 



This loss was first observed during the determination of phosphoric 

 acid in a sample of pottery waste. The ordinary hydrochloric acid 

 digestion process showed this material to contain 11-54 % of total P2O5. 

 When 1 grm. of the sample was fused in the usual way with 8 grms. 

 of pure ammonium fluoride so as to volatiUse the whole of the sihca, 

 the residue being gently ignited twice with sulphuric acid, then dissolved 

 in dilute hydrochloric acid or nitric acid, and made up to a known 

 volume (200 c.c), very variable and low results were obtained. Thus 

 in three successive experiments the sample appeared to contain only 

 7-37, 7-96 and 6-85 % of P2O5. 



Special experiments show that this loss of phosphoric acid does not 

 occur during ignition with sulphuric acid as carried out in the later part 

 of the experiment. It was only obtained when there had been previous 

 treatment with ammonium fluoride. 



