Experimental and modeling study of water-retention behavior of fine-grained soils with dual-porosity structures

Dual-porosity structures of fine-grained soils can noticeably affect their ability to retain water. This work jointly employs axis translation technique, filter paper method, and vapor equilibrium technique to study the soil–water retention curve (SWRC) over a wide suction range of Nanyang expansive soil characterized by double porosity. Mercury intrusion porosimetry tests are carried out to investigate the correlations between the aforementioned water-retention response and underlying pore structure characteristics. The test data show that dual-porosity distribution leads to bimodal SWRC. The change in void ratio mainly affects the median size of the inter-aggregate pores and consequently the portion of SWRC at low suction range. Based on these experimental observations, this work presents an SWRC equation for fine-grained soils with dual-porosity structures. Attracting water through capillary and adsorptive processes is explicitly distinguished. The capillary water is described by a relation that includes the characteristics of both inter- and intra-aggregate pore size distributions as parameters for representing bimodal characteristics. The adsorbed water is modeled by a relation that considers capillary condensation within intra-aggregate pores and allows for the decoupling between adsorptive water-retention mechanism and void ratio change. The latter feature is the foundation for the model to include the void ratio effect on SWRC in a way consistent with how it affects the pore structures of soils. By simulating test data in this work and in the literature, the proposed model is shown to be capable of representing the water-retention behavior of fine-grained soils with dual-porosity structures under different void ratios. To include the aforementioned key factors that influence the SWRC of fine-grained soils, seven parameters are required in the proposed model. This feature can reduce the practical applicability of the model. Future directions to enhance this aspect are discussed.