Hileman — Estimation of Fluorine Iodometrically . 383 



Art. XXXYI. — The Estimation of Fluorine Iodometrically * 

 by Albert Hileman. 



[Contributions from the Kent Chemical Laboratory of Yale University— cl.] 



It is obvious that the reaction by which fluosilicic acid liber- 

 ates iodine from a mixture of potassium iodide and potassium 

 iodate may be turned to account in the analysis of fluorides as 

 well as in the determination of fluosilicic acid provided the course 

 of action is regular. 



Upon testing the action of hydrofluosilicic acid upon the 

 iodide-iodate mixture, it was found that, while iodine is liber- 

 ated freely in the cold, a complete reaction was not obtained 

 in the course of several hours — the amount of iodine liberated 

 indicating that an acid other than fluosilicic acid, as a unit, was 

 acting. It appeared, however, that on boiling the mixture 

 nearly one equivalent of iodine is liberated for every equiva- 

 lent of fluorine present as fluosilicic of hydrofluoric acid. The 

 reaction may be written 



5KI + KI0 3 + H 2 SiF fi = 6KF + 6l + Si0 2 - H 2 



To the fluosilicic acid in a flask was added a neutral solution 

 of potassium iodide and iodate in excess and the flask closed 

 with a glass stopper fitted with a trap containing a solution of 

 potassium iodide to retain volatilized iodine. The solution in 

 the flask was heated to boiling, then cooled, and, together 

 with the contents of the trap, titrated with a standard solution 

 of thiosulphate. 



In section A of Table I are given the results obtained with 

 a solution of commercial fluosilicic acid. Titrations with sodi- 

 um hydroxide and comparative iodometric determinations are 

 given. In section B of Table I the results of similar experi- 

 ments with fluosilicic acid made by the action of hydrofluoric 

 acid on an excess of silica are given. 



The indications by the iodometric method are lower than 

 those of the alkalimetric method of titration by about 0*0008 

 grams, on the average. This fact would indicate that the 

 iodine liberated does not quite correspond to the complete 

 hydrolysis of fluosilicic acid indicated by the theoretical equa- 

 tion. 



Likewise Table II shows the results obtained by applying 

 the iodometric method to the silicon fluoride eliminated from 

 calcium fluoride according to the method proposed in a pre- 

 vious paper.* 



* This Journal, xxii, 329. 



