426 J. G. Dinwiddle — Hydrofluoric and Fluosilicic Acids. 



of this liltrate should he just two-thirds of the alkali required 

 in the reaction, 



II,SiF 6 + 6NaOH = 6NaF + Si(OII), + 2H 8 0, 



and so the amount of fluosilicic acid could he calculated. 



Titrations made in this manner all tended to run low to 

 ahout the same extent as those made after simply adding potas- 

 sium chloride and alcohol to the mixture of the two acids. 

 The explanation of this is prohahly that an insolnhle compound 

 of potassium fluosilicate and hydrofluoric acid is formed even 

 in the presence of soluhle calcium chloride. 



In a second series of experiments, to a solution of the mixed 

 acids calcium chloride was first added, and to an aliquot part of 

 the clear liquid was added potassium sulphate and alcohol. 

 The potassium ion should form potassium fluosilicate, and the 

 sulphate ion calcium sulphate which with 50 per cent alcohol 

 is sufficiently insoluble not to decompose the potassium fluosili- 

 cate when the solution is made neutral with alkali. The results 

 here were low to about the same extent as in the former series 

 of experiments. 



This method of making a mixture containing a precipitate 

 up to a definite volume and using an aliquot part of the clear 

 liquid is not to be recommended unless the precipitate is very 

 small and it is impossible to filter and wash it ; therefore, other 

 insoluble fluorides such as lead fluoride were tried, but the 

 precipitates were just as hard to filter and no better results 

 were obtained. 



Since other methods failed, and since calcium fluoride seemed 

 to be more suitable for the removal of fluorine than other 

 fluorides because of its greater insolubility, it was determined 

 to find some method of precipitating this compound in a form 

 capable of being filtered. 



The expedient of Hose, which is to precipitate calcium car- 

 bonate together with calcium fluoride, was obviously of no 

 value since the carbonate would decompose the fluosilicate 

 and form calcium fluoride. Calcium sulphate, while being 

 rather insoluble in water, is more soluble than calcium fluoride. 

 Therefore it was thought that if an excess of pure powdered 

 calcium sulphate were added to a solution of sodium fluoride 

 the more insoluble calcium fluoride would continue to form at 

 the expense of the sulphate until all of the fluoride was pre- 

 cipitated as calcium fluoride. There should then remain a pre- 

 cipitate of calcium fluoride mixed with the excess of calcium 

 sulphate. 



Upon trial, the precipitate formed in the above manner by 

 the use of calcium sulphate was found to be quite easy to filter 

 and wash, obtaining clear filtrates. A solution containing 





