332 Hileman — Alkalimetric Estimation of Silicon Fluoride. 



It is obvious that the process of titrating fluosilicic acid in 

 water solution yields uniform indications, both with potassium 

 hydroxide and sodium hydroxide, but that the values for fluo- 

 rine are very much below those of the titrations in alcoholic 

 solution. And this would be the case if the solution of fluo- 

 silicic acid contains hydrofluoric acid as is natural. 



In the analytical process in which silicon fluoride is passed 

 into the alcoholic solution of potassium chloride the forma- 

 tion of hydrofluoric acid is likely to be at a minimum and so 

 the titration of the hydrochloric acid set free in this meta- 

 thesis should indicate closely the actual amount of fluorine 

 present. If, however, the silicon fluoride is passed into water 

 instead of into this alcoholic solution of potassium chloride, it 

 is necessary to titrate together the products of action, fluo- 

 silicic acid and hydrofluoric acid, to the point of formation of 

 the alkali fluoride, in order that the indication may be correct. 



The action of ammonium hydroxide upon fluosilicic acid in 

 water solution proves to be comparable with that of sodium 

 hydroxide, and inferentially with that of potassium hydroxide, 

 though the hydrolysis of the fluosilicate appears to be not 

 quite so complete. In the table are given the results of com- 

 parative titrations. 









Table 



IV. 





HoSiF (; 





NH 4 OH 







Fluorine 



Fluorine foun 



taken 



[1cm 3 



found by NH 4 OH 



NaOH 



cm 3 . 





cm 3 . 







grm. 



grm. 



25 





31-04 







0-2030 



0-2057 



25 





31-00 







0-2030 



0-2057 



10 





12-39 







0-0811 



0-0823 



10 





12-42 







0-0816 



0-0823 



10 





12-42 







0-081 L 



0-0823 



The following expression represents the reaction : 



6NH 4 OH + H s SiF e = 6NH 4 F + Si(OH) 4 + 2H 2 



When silicon fluoride is passed into water containing ammo- 

 nium hydroxide, as is suggested in Liversidge's method for the 

 analysis of fluorides, it is obvious that nearly complete hydrol- 

 ysis must take place, with formation of silicic acid precipi- 

 tated or colloidal, according to the equation 



SiF 4 + 4NH a OH = 4NH 4 F + Si(OH) 4 . 



According to Liversidge, potassium fluosilicate" may be thrown 

 down by addition of potassium chloride and alcohol, after dis- 

 solving the precipitated silica by heating the ammoniacal 

 liquid. In my experience, it has never been possible to thus 

 dissolve all the silica, and the precipitate obtained by adding 

 potassium chloride, with or without alcohol, appears to be 

 *Chem. News xxiv, 266. 



