Chenghao Zhang, Adam Bezuijen
Infiltration experiments were carried out to elucidate the dominant factors governing the initial fast infiltration of bentonite slurry in a porous medium. Different combinations of two slurries, two sands, two pressures, and two cylindrical setups were tested. Pressure sensors placed at specific locations on the cylinders monitored variations in pore water pressure within sand columns throughout the process of infiltration. A digital balance recorded the real-time amount of discharged water. The experimental outcomes indicated that during bentonite slurry penetration, the yield stress of bentonite slurry had a more pronounced effect on the pressure drop across the slurry infiltration zone than its viscosity. This was evidenced by a comparable pressure drop over a given slurry infiltration distance, which remained hardly affected by fluctuations in infiltration velocity. A greater pressure drop across the slurry-infiltrated zone led to less driving pressure being available to displace water in the unpenetrated sand since the overall pressure drop remains constant. The infiltration velocity of the slurry is determined by this driving pressure in combination with the permeability and length of the unpenetrated sand. Lower driving pressure, reduced sand permeability, or a longer sand length all result in a decreased infiltration velocity. Furthermore, it can be concluded that the yield stress of the slurry in combination with the sand properties determines the final infiltration distance of the bentonite slurry. In the absence of external filter cake formation, this final infiltration distance can be estimated by balancing the pressure drop across the slurry infiltration zone with the applied pressure on the bentonite slurry. An analytical model incorporating the yield stress of the slurry and the equivalent pore radius of the sand was developed to estimate the pressure drop over the slurry infiltration zone. Subsequently, by analyzing the pressure distribution over the testing system, analytical models for determining infiltration velocity and infiltration distance were formulated. Both models demonstrated a high degree of correlation with the experimental data.