It is a well known fact that the quantitative absorption of carbon dioxide in aqueous solutions of baryta either requires a very large excess of baryta or the use of complicate absorption vessels, or both. A survey of the difficulties and of different types of apparatus has been given by PARTRIDGE AND SCHROEDER~, LINDNER~ and recently in this journal by PIETERS*. Both ways of meeting the difficulty are rather unattractive. In volumetric determinations the large excess of baryta, which has to be titrated, gives us the carbon dioxide as the small difference between two large quantities, and the complicate absorption vessels are not only expensive but moreover liable to breakage and difficult to cleanse.
We won't here go into the subject of the origin of the trouble. It seems to be related to the slow rate of formation of carbonate and bicarbonate ions from carbon dioxide and watePq and to the rather low solubility of this gas in water. Now we know that in a great many organic solvents the solubility of CO, is considerably greater than in waterb, and one could hope that in some of them the velocity constants were more favorable, although as far as we know not a single indication in this direction has ever been given in literature.
Indeed, after some preliminary experiments, we founa that in a mixture of aniline, ethanol and. aqueous baryta the absorption of CO, is far more complete than in aqueous baryta alone. The addition of ethanol only serves the purpose of obtaining a homogeneous solution, When a slow current of CO,, carried as usual by air at x-z bubbles per second, is passed through an ordinary washing bottle containing 5 ml of pure, freshly distilled aniline, x5 ml of ethanol (96%) and 7 ml of N/IO aqueous baryta, it is quantitatively retained, provided that its amount does not exceed about ,807~ of the theoretical absorbing capacity of the baryta used.
As of course nothing is known about the dissociation constants in this organic medium it is useless to speculate on the choice of the best indicator. We had to find it experimentally. To this end we started from a well kept stock solution of baryta in water (0.0970 IV) and of hydrochloric acid (0.1034 N) ; 7.00 ml of this baryta is aequivalent to 6.56 ml of HCl when titrated in the ordinary way with &fcrences p. 70. VOL. 5 (rggx) ABSORITION OF CARBON DIOXIDE 69