Effects of Pore-Size Distribution on the Gas Diffusion Coefficient and Gas Permeability of Compacted Manufactured Sand Tailing–Bentonite Mixtures

Gas permeability ($k_a$) and gas diffusion coefficient ($D_p$) are vital for controlling gas transport through earthen landfill cover overlying municipal solid wastes or acid mine tailings. Soil clods generated during the construction of landfill cover can result in various soil microstructures. However, the effects of soil microstructure on $D_p$ and $k_a$ are unclear. This study aimed to investigate and compare effects of soil microstructure on measured $k_a$ and $D_p$ of compacted manufactured sand tailings (MST)-bentonite mixtures, considering the effects of specimen preparation method, compaction water content ($w_{\mathrm{comp}}$), and wetting path. The related pore size distributions (PSDs) were measured by mercury intrusion porosimetry. The measurements show that the average $k_a$ of as-compacted specimens increased and then decreased as $w_{\mathrm{comp}}$ increased (i.e., a lower soil air content $\epsilon$—gas-filled volume per unit soil volume), while $D_p$ reduced monotonically. It was because the compacted MST-bentonite mixtures exhibited a multimodal PSD, in which macropores significantly increased the average advective flow velocity of gas. However, the average diffusive flow velocity of gas was independent of pore size, according to the kinetic theory of gas molecules. Conversely, $k_a$ and $D_p$ both decreased as $\epsilon$ decreased when the water content of the as-compacted specimen was increased along the wetting path by spraying water, due to the wetting-induced collapse of macropores. These illustrated that $D_p$ was mainly controlled by the $\epsilon$ and tortuosity of gas-filled pores, while $k_a$ was dominated by well-connected gas-filled macropores. Moreover, the PSD of specimens was affected by the specimen preparation method, bentonite content, and sieving treatment, thus affecting $k_a$ and $D_p$.