J. G. Dinwiddie — Hydrofluoric and Flicosilicic Acidj. 427 



hydrofluoric acid and fluosilicic acid was now treated with cal- 

 cium sulphate, and, after stirring well and allowing to stand, 

 the precipitate was filtered and washed. To the filtrate was 

 added alcohol and potassium chloride, and then the free acid 

 was neutralized with standard sodium hydroxide. The results 

 here were low, just as in the previous attempts, and the explana- 

 tion is very probably that calcium fluoride is sufficiently solu- 

 ble in the free acid present to dissolve partially and to form in 

 the filtrate free hydrofluoric acid upon addition of the alcohol 

 and this would then be absorbed by the potassium fluosilicate. 

 If sodium acetate is added in order to decrease the acidity of the 

 solution before addition of calcium sulphate, then the filtrate 

 continues to require alkali very much beyond the amount 

 required by theory. It is probable that in very weakly acid 

 solutions the calcium sulphate acts on the fluosilicic acid to 

 form calcium fluoride, and this would account for the excess 

 of alkali required as follows : 



H 2 SiF 6 + 3CaS0 4 + 4H 2 >Si(OH) 4 + 3H 2 SO t + 3CaF 2 . 



Finally the method of Stolba" x ' for determining the water of 

 crystallization in crystallized fluosilicates was tried. This 

 method was to heat a weighed amount of the material, e. g. 

 CuSiF .XH 2 O with a weighed amount of magnesium oxide. 

 The reaction is as follows : 



CuSiF 6 + XH 8 + 3MgO = CuO + 3MgF 2 + SiO 2 + XH 2 0, 



and the loss in weight, therefore, represents the water of crys- 

 tallization. 



To a known weight of magnesium oxide in a platinum 

 crucible was added a solution containing a known amount of 

 fluosilicic acid and the mixture was well-stirred. The reaction 

 should be 



(a) H 2 SiF 6 + 3MgO = 3MgF, + SiO s + H 2 0, 



and upon evaporating to dryness and heating to redness, there 

 should remain (according to Stolba) only 3MgF 2 -I- SiO\ 

 With the same weight of fluorine in the form of hydrofluoric 

 acid only 3MgF, would remain according to the equation 



(b) 6HF + 3MgO = 3MgF, = 3H 2 0, 



knowing the total amount of fluorine that in the form of 

 hydrofluoric acid and of fluosilicic acid in a mixture could be 

 calculated by the increase in weight of the magnesium oxide 

 used. However, upon heating the mixture formed as in equa- 

 tion (a) to a temperature somewhat below redness, the increase 

 in weight was considerably more than required by theory, 

 * Stolba, Jour, prakt. Chem., cii, 2. 



