4^2 ./. (i. Dinwiddie — Hydrofluoric and Fluosilicic Acids. 



In testing this method on a mixture of known concentra- 

 tion, Katz finds that the alcoholic-potassium-chloride titration 

 requires considerably less than the theoretical amount of alkali. 

 He explains this result by assuming that some of the free 

 hydrofluoric acid is absorbed by the precipitated potassium 

 fluosilicate and is thus kept from being neutralized. Using 

 normal and twice-normal alkali for titrations, lie gets results 

 which tend to indicate that, for very small amounts of fluo- 

 silicic acid, each molecule carries down with it one molecule of 

 hydrofluoric acid, while for larger amounts three molecules 

 carry down two of hydrofluoric acid. 



On this basis he makes out a sliding factor to be used accord- 

 ing to the relative proportions of the two acids present. For 

 the fluosilicic acid content, if the difference between the water 

 and the alcoholic titration is 



5$ or less of the water titration, multiply it by (-0576), 

 5-10$ of the water titration, multiply it by (-058--0595), 

 10-12$ " " " " " " " (-06- -061), 



over 12$ " " " " " " " (-0617). 



These factors are calculated on the basis of 2 N alkali. 



Since Katz uses twice-normal alkali, and uses only from 

 three to eight cubic centimeters in a number of the titrations, 

 there is ample room for very large percentage errors. 



Before attempting to go farther, it was thought advisable to 

 repeat some of Katz's work as well as to compare the titration 

 of pure fluosilicic acid in water and in alcoholic-potassium- 

 chloride solution. 



For a comparison of the two titrations of the pure acid, 

 three samples were used : No. 1 was made by suitable dilution 

 of Baker and Adamson's commercial fluosilicic acid ; Nos. 2 

 and 3 were made by treating silica and calcium fluoride with 

 concentrated sulphuric acid, passing the evolved silicon tetra- 

 fluoride through a double coil of glass tube (water-cooled) and 

 then into water where the fluoride of silicon was decomposed 

 into fluosilicic acid and silica. 



The following table shows the results of these titrations : 



Table I. 



Alkali used was approx. 033 normal NaOH. 



H 2 SiF 6 NaOH req. NaOH req. NaOH req. 



grin, of solution (in alcohol) (in water) (cc. per gram) 



cc. cc. 



f 7-6774 42-1 5'48 



■vj J 6-6227 36-3 5-481 



JN0 " 1 8-5302 15-9 1-864 



(_8-4770 15-7 1-852 



