T. B. Osborne — Quantitative determination of Niobium. 335 



For the determination of titanium in the presence of niobium 

 the method proposed by A. Weller (Berichte, xv, 1882, 2, p. 

 2592), appeared most suitable. It depends upon a comparison 

 of the color produced by hydrogen dioxide in an acid solution 

 of titanic acid, with the color of a solution containing a known 

 amount of titanium. A. Weller, in describing his method, calls 

 attention to its applicability to the determination of titanium 

 in the presence of tantalum and niobium, but says he has not 

 tried it in this case. I therefore undertook to determine the 

 best manner of applying it. I found that it was necessary to 

 use hydrofluoric or hydrochloric acid in order to keep the tan- 

 talum and niobium in solution when sufficiently diluted. I 

 found that the first of these acids even in small amounts entirely 

 prevented the formation of the color of titanium with hydrogen 

 dioxide and that hydrochloric acid deepened it very greatly in 

 proportion to the amount added. It is necessary, therefore, to 

 have the solution free from fluorides, and to have the same 

 amount of acid in the standard solution as in the solution to be 

 analyzed. It is very difficult to obtain these conditions and I 

 was entirely unsuccessful in my attempts to determine the 

 amount of titanium in a solution containing much free acid. 

 I succeeded in obtaining quite satisfactory results, however, by 

 precipitating the strongly acid solution with a small excess of 

 ammonia and then redissolving the precipitate with as little sul- 

 phuric acid as possible, the freshly precipitated niobic and tan- 

 talic acids being readily soluble in dilute acid. 



In order to test the accuracy of this method a mixture of 

 tantalic, niobic and titanic acids containing a known amount of 

 each acid was dissolved in hydrofluoric acid in a platinum 

 dish and the excess of acid evaporated off on the water bath. 

 The fluorides thus obtained were dissolved in concentrated 

 hydrochloric acid and poured into a glass flask of about 100 c. c. 

 capacity, the platinum dish being rinsed out with concentrated 

 hydrochloric acid. The solution and rinsings amounted to 50 

 c. c. Amalgamated zinc was then added and a piece of plati- 

 num and the solution allowed to reduce in a stream of carbonic 

 acid for three-quarters of an hour at a temperature of about 

 80° C. During this process the greater part of the hydrofluoric 

 acid was removed from the solution, going off as silicon fluoride. 

 After cooling thoroughly the reduced solution was poured into 

 a beaker and diluted to 350 c. c. with distilled water freshly 

 boiled and perfectly cold. A standard solution of potassium 

 permanganate was added till the solution, at first nearly black, 

 became perfectly clear and a distinct pink color was produced 

 by the addition of a single drop. To the solution containing 

 the tantalum, niobium and titanium, ammonia was added in 

 slight excess and then sulphuric acid till the precipitate pro- 

 duced by the ammonia dissolved entirely. In this way a solu- 



