Measuring Soil-Water Density by Helium Pycnometer

Soil-water density (SWD) is the density of water retained in soil. Its value can be remarkably different from the free water density (i.e., $0.997\mathrm{g}/\mathrm{c}{\mathrm{m}}^3$). SWD as high as $1.68\mathrm{g}/\mathrm{c}{\mathrm{m}}^3$ has been reported (e.g., Martin 1960; Richards and Bouazza 2007; Zhang and Lu forthcoming). However, SWD is universally treated as a constant equal to the free water in defining fundamental soil properties of volumetric water content, specific surface area, pore-water pressure, and matric suction. Many research efforts have been devoted to measuring SWD and are critically reviewed in Martin (1960) and Zhang and Lu (Forthcoming). All previous methods are plagued with either unjustified assumptions or limited applicability within a certain water content range. Consequently, current knowledge on SWD greatly deviates on the magnitude and even the direction in abnormality. To date, there is still no reliable experimental method for measuring SWD as a function of water content.

Fundamentally, SWD abnormality can be interpreted from physical mechanisms of soil-water interaction (Zhang and Lu Forthcoming). At low water content, the soil-water interaction is dominated by van der Waals interaction and surface and cation hydration. These interactions are produced by attractive van der Waals forces or electrostatic hydration forces, compressing water molecules, and thereby producing a SWD higher than the free water. At high water content, the soil-water interaction is dominated by capillarity under unsaturated conditions, leading to negative pore water pressure and a SWD lower than the free water. Thus, SWD is intrinsically related to thermodynamic states of soil water and depends on water content.

In this paper, a humidity-controlled helium pycnometer method is reported to measure SWD as a function of water content. The soil samples with various water contents (i.e., soil-water mass) are first prepared by equilibrating in a humidity-controlled chamber with salt solutions. The helium pycnometer, conventionally used to determine soil’s specific gravity (e.g., Richards and Bouazza 2007), is then utilized to measure the soil sample volume. Assuming unchanged in solid density, the soil-water volume is the measured soil sample volume subtracting the solid volume. The SWD at various water contents is calculated by dividing the measured soil-water mass with the soil-water volume.

The proposed method is validated by comparing it with the buoyant force method (Bahramian et al. 2017) and is illustrated in Fig. 1(a) for Wyoming montmorillonite. The proposed method produces close matches to the buoyant force method. Fig. 1(b) illustrates the SWD results measured by the present method for three different types of soil. At low water content, the measured SWD can be as high as $1.37\mathrm{g}/\mathrm{c}{\mathrm{m}}^3$ and decreases with increasing water content.