Density of Water sorbed on Montmorillonite

Abstract

IN a recent communication, Anderson and Low¹ have described some investigations indicating that the water present in Wyoming bentonite (mont-morillonite) pastes has a density lower than that of liquid water. It should be noted, however, that this does not preclude the water sorbed in the primary process having a close-packed structure around the ions², and consequently having a much higher density. For example, for monovalent or divalent ions the amount of such water would only be some 4 or 12 molecules, respectively, per ion, and if the additional sorbed water were considered to have, say, the hexagonal-net structure³, the density of which would be about 0.9, the value for the density of the sorbed water would drop rapidly at water contents above about 7–10 per cent, and for pastes with 60 per cent water¹ could easily be less than unity. Such a decrease in density of sorbed water is well shown by the results of De Wit and Arens⁴, who obtained values as high as 1.41 at 11.6 per cent moisture, dropping to 0.997 above 30 per cent. Somewhat similar results were obtained in this Institute some time ago using a density-bottle technique with a specially devised de-airing procedure and ‘Dekalin’ as the non-sorbed liquid. Values ranging from about 1.38 at 9 per cent water to about 1.06 at 20 per cent water were obtained—in substantial agreement with the results of De Wit and Arens⁴.