Pore size distribution and swelling behavior of compacted bentonite/claystone and bentonite/sand mixtures

The swelling behavior of clayey materials has been studied for many years, and the influence of pore size distribution and suction has been found to be significant. Swelling tests with suction control have been commonly and successfully performed on highly expansive clayey materials, but such work has not always included materials with small percentages of expansive clay minerals. In this study, compacted bentonite (Be), clayey material (C) from claystone, and mixtures of bentonite with clayey material (BeC) and sandy material (BeS) were subjected to swelling tests by the constant volume method in both conventional oedometers with water hydration (water wetting) and oedometers with suction control by relative humidity imposed by NaCl solutions (vapor wetting). The C samples at the optimal moisture content showed no swelling pressure by the vapor wetting method, even for an initial suction of 40,000 kPa. The BeC mixtures showed a lower swelling pressure with the vapor wetting method than with the water wetting method for samples with less than 70% bentonite, whereas the BeS mixtures showed similar swelling pressures in both methods. The BeC and BeS samples had a maximum swelling pressure that was directly proportional to the percentage of bentonite added when tested by the water wetting method. Mixtures with the same percentage of bentonite showed an increase in the swelling pressure that was directly proportional to the initial suction imposed before the tests. Pore size distribution (PSD) curves generated before the swelling tests showed that the greater the initial interaggregate porosity difference among the BeC and BeS mixtures, the greater the swelling pressure difference among the BeC mixtures tested by the water wetting method and the BeS mixtures tested by the vapor wetting method for the same percentages of bentonite. The PSD analysis also showed that the mixtures with higher volumes of entrapped air had higher swelling pressures, mainly during testing by the water wetting procedure. Therefore, the results of the swelling tests are highly dependent on the initial sample suction and the clay mineralogy, but the swelling tests with the vapor wetting method may lead to measurements lower than actual swelling pressures when clayed materials with reduced percentages of expansive clays are tested.