124 II. E. Merwin — Peroxidized Titanium Solutions. 



The full lines of fig. 1, A were thus plotted. Suppose 

 this ratio r in a particular case is 142, and that there is *5 CC of 

 acid in the test solution, then the corresponding amount of 

 fluorine is '0006 g. The formula expressing this relation is 



— = g. of fluorine. With 3*5 CC of acid in the test solution 

 70,000 & 



the formula is = ff. of fluorine. By the conditions of 



22,000 & J 



the first formula amounts of fluorine between -00005 g. and 



•001 g. can be estimated accurately within *00005 of a gram ; 



by the last formula amounts between "001 g. and *004 g. can be 



estimated within '00015 of a gram. By doubling the amounts 



of titanium and acid in the test solution and making it up to 



100 ec , *01 g. of fluorine can be estimated. 



The above formulas can not be used in rock analysis because 

 of the disturbing effects of the alkali sulphates necessarily 

 present in the test solution. In order to make it possible to 

 know accurately the composition of the solution in which 

 fluorine is to be determined in rock analysis, the following 

 method has been worked out. 



Two grams of the rock powder are fused with 8 grams of 

 mixed sodium and potassium carbonates and the fusion is taken 

 up with hot water. When leached, and without the necessity 

 for filtering, there are added 3 or 4 grams of powdered ammo- 

 nium carbonate ; the mixture is warmed for a few minutes, 

 and then heated on the water bath till the ammonium carbon- 

 ate is destroyed, and the bulk of the liquid is small. In this 

 way the silica, which otherwise might render the final solution 

 turbid, is thrown down together with the disturbing alumina 

 and ferric oxide. The destruction of the ammonium carbonate 

 is necessary because ammonium sulphate bleaches the final 

 solution. After filtering there is added to the filtrate — which 

 should not exceed 75 cc in volume — 3 or 4 CC of hydrogen per- 

 oxide, and then cautiously 10 cc of standard titanium solution* 

 (containing -01 g. Ti0 2 ). Including the acid in the titanium 

 solution, about 4 CC of strong sulphuric acid are required to 

 neutralize the alkali carbonates. As soon as neutrality is 

 reached the solution acquires a light orange color. Neutrality is 

 tested by adding a little sodium carbonate solution to discharge 

 the color, and then a drop or two of acid to restore it. The 

 further treatment depends upon the amount of fluorine 

 expected. In the vast majority of cases this amount is less than 

 •0025 g. ('125 per cent of the sample). For such amounts there 

 is added to the neutralized solution 3 CC of concentrated sulphuric 

 acid, and the solution is made up to 100 cc . After being cooled 



* The hydrogen peroxide prevents the precipitation of the titanium by the 

 alkali carbonate. 



