•_M4 W. M. Thornton, -Jr. — Estimation of Titanium. 



Aim. XIX. — The Estimation of Titanium in t/u Pnxence 



of Iron ; by William M.. Tiiounton, Jr. 

 [Contributions from the Kent Chemical Laboratory of Yale Univ. — ccxxxiii.] 



During the author's service as chemist on the Virginia 

 Geological Survey a good many titaniferous rocks, containing 

 the minerals rutile and ilmenite in various associations, were 

 presented for analysis. Since the titanium content of these rocks 

 was (if ten high (amounting to 69 percent in a few cases) it was 

 necessary to resort to a reliahle gravimetric method. Among 

 the gravimetric methods in use at the present time that have 

 heen thoroughly tested, the one hrought out by Gooch* is per- 

 haps unequaled, and especially recommends itself because the 

 iron can be completely separated from the titanium by a single 

 treatment. This method provides not only for the separation 

 of titanium from iron but also from aluminium and phosphoric 

 acid. Since the Virginia rocks which ran highest in titanium 

 were non-aluminous (or nearly so), it occurred to the author 

 that the method might be so modified as to shorten the process 

 and eliminate some of the more difficult manipulation. 



The first step is to separate the iron. This is done by 

 adding tartaric acid in quantity equal to three times the aggre- 

 gate weight of the oxides to be held by it in solution. Hydro- 

 gen sulphide is then introduced until the iron has been reduced 

 to the ferrous condition. Ammonium hydroxide is then added 

 to slight alkalinity, and hydrogen sulphide again passed in 

 until the iron has been completely precipitated and the solu- 

 tion left faintly alkaline to litmus paper. After filtration and 

 washing of the ferrous sulphide with very dilute colorless 

 ammonium sulphide, the titanium is entirely in the iron-free 

 filtrate. The tartaric acid must next be destroyed ; for tita- 

 nium can not be precipitated by any reagent in its presence. 

 This is done by acidifying the filtrate with considerable sul- 

 phuric acid, boiling out the hydrogen sulphide thus set free, 

 and adding potassium permanganate (in the form of a strong- 

 solution) to the boiling solution until a permanent precipitate 

 of manganic hydroxide appears, which is dissolved in a little 

 sulphurous acid. The result of this reaction is that the tar- 

 taric acid is oxidized partly to formic acid and partly to car- 

 bonic acid. The titanium can then be precipitated by the 

 basic acetate process. Much manganese is thus introduced into 

 the solution and is co-precipitated in some measure with the 

 titanium — necessitating a solution of the precipitate and a sec- 

 ond precipitation of the titanium by the above mentioned 



* Proc. Am. Acad. Arts and Sci., n. s., vol. xii, p. 435. 



