Prediction of colloid sticking efficiency at pore-scale and macroscale using a pore network model

Abstract

The sticking efficiency (α) is a vital parameter to predict the transport and deposition of colloids in porous media. The value of α depends on various factors such as the interaction energy between colloids and the solid-water interface (SWI), kinetic energy fluctuations of diffusing colloids, and the hydrodynamics of the flow field in the pore structure. However, α is usually assumed to be spatially constant and fitted from experimental data. In this study, a theoretical method is proposed to predict the distribution of the sticking efficiency at pore scale (αt) using a pore network model (PNM) and the macroscale sticking efficiency of the whole porous media (αM) by means of upscaling. A PNM is established to simulate the pore structure of porous media and to obtain the flow field, and energy and torque balances are calculated throughout the spatial pore structure to determine the distribution of αt under different physiochemical conditions. The value of αM is determined through breakthrough curves provided by the PNM under favorable and unfavorable conditions. Results show the distribution of αt is sensitive to various factors including colloid characteristics, pore surface features, geochemical environment and hydrodynamic conditions. Colloid characteristics like colloid surface zeta potential, geochemical environment such as solution ionic strength, and pore surface characteristics including nanoscale roughness and especially charge heterogeneity have a controlling influence on αt for nanoparticles (<100 nm) due to their impact on interaction energies. Hydrodynamic conditions played an increasingly important and eventually dominant role in determining αt for larger colloids by changing the sticking of weakly associated colloids (e.g., at secondary minima). When hydrodynamic torques are weak, the influence of colloid size on αt can be non-monotonic due to the combined influence of interaction energy and hydrodynamic torques. As a result, higher values of αM occurred for smaller colloid sizes, lower flow velocities, larger pore sizes, and in the presence of microscopic roughness. The method proposed by this study can be used to predict the sticking efficiency under different solution and solid phase chemistries, nanoscale heterogeneities, microscopic roughness, flow velocities, and pore structure conditions.