Kuzirian — Correcting Silica for Included Salts. 61 



Aet. V. — A Modification of the Usual Method of Correct- 

 ing Silica for included Salts ; by S. B. Kuzirian. 



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



When silica, liberated by the action of hydrochloric acid 

 upon the product of fusion of a silicate with an alkali carbon- 

 ate, is made insoluble and separated by the usual subsequent 

 treatment, — viz., evaporation, desiccation at 110°, extraction 

 with dilute hydrochloric acid and water, filtration, and repeated 

 washing with distilled water, — it invariably contains foreign 

 material which after strong ignition consists essentially of 

 oxides, free or combined with the silica. If the silica is vola- 

 tilized by treatment with hydrofluoric acid and sulphuric acid, 

 the included material is composed of sulphates which remain 

 as such or are changed to oxides in a subsequent ignition. If 

 the constituents of this residue were definite and correctly 

 determinable, it would be possible to calculate the weight of 

 the oxides as they exist in the strongly ignited silica, and to 

 make the proper correction ; but it is a difficult task to make a 

 correct analysis of a residue which sometimes does not exceed 

 two to three milligrams and may contain iron, aluminum, man- 

 ganese, titanium, magnesium, sodium, potassium, etc. In order 

 to avoid making the quantitative analysis of this small residue 

 of sulphates and oxides, it has been an approved procedure to 

 blast the silica to constant weight and then to assume that after 

 the prolonged blast ignition the impurities are left in the form 

 of oxides. Then, after the removal of the silica with sulphuric 

 and hydrofluoric acids, the remaining sulphate residue is also 

 blasted to constant weight, and the assumption is made that 

 this latter step will transform the sulphates of the residues into 

 their oxides. The weight of silica is then found by difference. 

 It will be shown in the following account of experimental work 

 that the residue obtained after treatment of the blasted silica 

 with sulphuric and hydrofluoric acids and blasting does not 

 accurately represent the amount of included material as it is 

 ordinarily weighed with the silica. 



In the presence of moisture from the Bunsen burner or the 

 blast-lamp, chlorides are slowly decomposed with evolution of 

 hydrochloric acid and the simultaneous formation of metallic 

 oxides or, in the case of the alkali metals, hydroxides, which 

 may then enter into combination with any suitable non-metallic 

 oxide with which they are in contact. Silica containing traces 

 of soluble alkali chlorides may, on strong ignition, gradually 

 combine with the metallic oxides formed, as has been shown in 

 a previous paper.* The assumption that the impurity in the 

 silica, contaminated by chlorides when it is separated, con- 

 sists after the ignition exclusively of oxides is, therefore, well 

 founded. 



*This Journal, xxxvi, 598, 1913. 



