Kuzirian — Correcting Silica for Included Sails. 63 



As is well known, sulphates other than alkali sulphates, lose 

 the acidic oxide, more or less according to the conditions, 

 when submitted to ignition with the blast-lamp. While pro- 

 longed blast-ignition may bring about a complete transforma- 

 tion of the sulphates of iron, aluminium, chromium, and 

 titanium to the respective oxides, the refractory alkali sulphates, 

 as well as the sulphates of magnesium, calcium, and barium in 

 large degree, will remain in the condition of sulphates (though 

 the alkali sulphates may volatilize appreciably), and the correc- 

 tion for silica as ordinarily applied will be in error by an 

 amount approximately equal to that of the sulphur trioxide 

 combined in the sulphates. 



The source of error, inherent in the usual method of apply- 

 ing corrections to the ignited silica, may be largely avoided by 

 the introduction of a slight modification of the treatment. 

 This modification consists essentially in treating the ignited 

 silica with sulphuric acid, gently igniting again before weigh- 

 ing, and in igniting under exactly the same conditions the sul- 

 phate residues left after the removal of the silica in the usual 

 way. That is to say, the included impurities of the silica must 

 be transformed before weighing into the condition in which 

 they will be left when the silica is removed by sulphuric acid 

 and hydrofluoric acid. This can be accomplished by adding a 

 few drops of dilute sulphuric acid to the well ignited impure 

 silica, evaporating the excess of the acid slowly over a radiator 

 and igniting the residue by means of a Bunsen burner, before 

 weighing the residue. Then, after the removal of the silica 

 in the usual manner, the sulphate residue left is ignited, at the 

 same temperature and for the same duration of time as was 

 the silica, and then weighed. From the weight of the silica 

 plus the impurity before the treatment with sulphuric acid 

 and hydrofluoric acids and that of the residue after that treat- 

 ment the weight of the silica is found by difference. 



Table III contains the details of experiments in which silica 

 was fused with six times its weight of sodium carbonate, the 

 melt treated with hydrochloric acid ; the mixture evaporated ; 

 the residue desiccated (at 110° (A) or, in presence of acetic 

 anhydride, at 137° (B) ), and extracted as usual with hydro- 

 chloric acid ; the precipitate filtered off and washed ; and the 

 filtrate treated again like the original solution of the melt for 

 the recovery of silica soluble in the former operation. The 

 residues (first, second, and total) and the error which results 

 from calling these residues oxide instead of sulphate and sub- 

 tracting their weight from that of the ignited silica, according 

 to the usual method of making the correction, are shown. 



