Inspired by natural occurring hydrophobic soil, researchers proposed to apply hydrophobised soil as a cover to reduce water infiltration and hence enhance the stability of earthen structures, such as slopes and landfill covers. Although the hydrological behaviour of hydrophobised soil has been widely studied, the mechanical behaviour of hydrophobised soil remains poorly understood. Thus, research is needed in not only hydrological but also mechanical behaviour of hydrophobised soil especially at unsaturated conditions. The aim of this study is to conduct a multiscale investigation on hydromechanical behaviour of unsaturated hydrophobised soils.

Five interconnected tasks were conducted in this study. In Task I, formation and evolution of water meniscus between hydrophobised particles we...[ Read more ]

Inspired by natural occurring hydrophobic soil, researchers proposed to apply hydrophobised soil as a cover to reduce water infiltration and hence enhance the stability of earthen structures, such as slopes and landfill covers. Although the hydrological behaviour of hydrophobised soil has been widely studied, the mechanical behaviour of hydrophobised soil remains poorly understood. Thus, research is needed in not only hydrological but also mechanical behaviour of hydrophobised soil especially at unsaturated conditions. The aim of this study is to conduct a multiscale investigation on hydromechanical behaviour of unsaturated hydrophobised soils.

Five interconnected tasks were conducted in this study. In Task I, formation and evolution of water meniscus between hydrophobised particles were observed upon controlled wetting and drying via an environmental scanning electron microscope (ESEM). In Task II, a new unified water retention cell was developed to measure hysteretic water retention curves (WRCs) of soils of different hydrophilicity and hydrophobicity. In Task III, flow pattern and percolation of hydrophobised hydraulic barriers and underlying soil layers were examined by soil plate and soil column experiments. In task IV, hydromechanical behaviour (compressibility, shearing and dilatancy) of hydrophobised soil was investigated by oedometer and direct-shear box tests at varying degrees of saturation and stress levels. In Task V, preliminary microscopic investigation of hydrophobised sand was conducted by the discrete element method (DEM).

In terms of hydrological behaviour, it was discovered that concave menisci can be formed between untreated-treated particles because water tend to be attracted by the untreated surfaces which have surface free energies higher than the surface tension of water rather than by the treated surfaces which have surface free energies lower than the surface tension of water, hence leading to concave shaped menisci. The concave menisci verified, for the first time, the presence of the effects of matric suction in the unsaturated 50%-treated sand. Wetting and drying paths of WRCs of hydrophobised materials were explained by the materials’ maximum, minimum contact angles (which indicate the movement of three-phase (solid, liquid, vapour) contact lines) and the Washburn equation. The difference of hysteretic WRCs between untreated and treated specimens was partially explained by a new ink-bottle conceptual model. Furthermore, a hydrophobised barrier can prevent the percolation of underlying soil when the ponding head is less than the water-entry head (WEH) of the barrier. The barrier can become ineffective when the percentage of treatment reduces or when the initial degree of saturation increases due to the reduction of WEH.

In terms of hydromechanical behaviour, experimental results reveal that (i) compressibility is unaffected when DMDCS content is less than 10%, and it is a function of degrees of saturation (S) when the percentages of treatment is less than 100% due to concave menisci; (ii) the shearing mechanism switches from strain-softening to strain-hardening beyond a threshold DMDCS content because of the switch of shearing contacts from between particles to between coatings; (iii) the shear stress mobilised in the 50%-treated sand lies in between that of the untreated and 100%-treated sands due to the presence of particle-coating contacts; (iv) the peak friction angle declines with the increase in DMDCS content and is S-dependent for 50%-treated sand due to the presence of concave menisci in untreated-treated particles; and (v) soil dilatancy decreases significantly beyond a threshold DMDCS content of 0.05% or a threshold percentage of treatment of 50%. Besides, the preliminary micromechanical investigation suggests that more data are needed in the particle-level compression and shearing experiments to obtain the friction coefficient of PDMS coatings as well as the vertical displacement of hydrophobised particles during shearing. Accordingly, the contact model and its parameters in the DEM can be reselected and recalibrated when the experimental data of hydrophobised sand are available.