Colloid Retention in Porous Media:  Mechanistic Confirmation of Wedging and Retention in Zones of Flow Stagnation

Abstract

A three-dimensional particle tracking model for colloid transport in porous media was developed that predicts colloid retention in porous media in the presence of an energy barrier via two mechanisms: (1) wedging of colloids within grain to grain contacts; (2) retention of colloids (without attachment) in flow stagnation zones. The model integrates forces experienced by colloids during transport in porous media, i.e., fluid drag, gravity, diffusion, and colloid-surface Derjaguin-Landau-Verwey-Overbeek interactions. The model was implemented for a fluid flow field that explicitly represented grain to grain contacts. The model utilized a variable time stepping routine to allow finer time steps in zones of rapid change in fluid velocity and colloid-surface interaction forces. Wedging was favored by colloid: collector ratios greater than about 0.005, with this threshold ratio increasing with decreasing fluid velocity. Retention in flow stagnation zones was demonstrated for colloid: collector ratios less than about 0.005, with this threshold decreasing with increasing fluid velocity. Both wedging and retention in flow stagnation zones were sensitive to colloid-surface interaction forces (energy barrier height and secondary energy minimum depth). The model provides a mechanistic basis for colloid retention in the presence of an energy barrier via processes that were recently hypothesized to explain experimental observations.

Comment on “Transport and fate of bacteria in porous media: Coupled effects of chemical conditions and pore space geometry” by Saeed Torkzaban et al.

“…The issue we raise is not whether retention of secondary energy minimum associated colloids occurs, but rather, how it occurs. Our own simulations using a traditional force and torque balance for secondary energy minimum associated colloids show that colloids may be retained in zones of low fluid drag at the pore scale even without adhesive contact with the surface [Johnson et al, 2007]. In our model, retention without attachment occurred in rear stagnation zones, although these represent just one possible type of zone of low fluid drag.…”

“…There are two important points here: (1) retention of secondary energy minimum associated colloids can be simulated in a system where colloid spinning and translation are allowed, and (2) spinning of the colloid in response to fluid shear does not necessarily lead to reentrainment. These two points, demonstrated by Johnson et al [2007] for secondary energy minimum associated colloids, contrast with the assumptions used by Torkzaban et al [2007Torkzaban et al [ , 2008 that (1) colloids may be immobilized by secondary energy minimum interaction; that is, secondary energy minimum interaction constitutes adhesive contact with the surface, and (2) the initiation of rolling can be equated to reentrainment.…”

“…The latter is built into the force balance by way of hydrodynamic retardation coefficients [e.g., Rajagopalan and Tien, 1976;Johnson et al, 2007]. This approach is taken with the expectation that the energy barrier between the surface and the secondary energy minimum prevents physical contact between the colloid and stationary surface, such that the friction resisting colloid motion emanates from the fluid viscosity rather than emanating from adhesive contact between the colloid and the surface (Figure 3).…”

Using the DEM-CFD method to predict Brownian particle deposition in a constricted tube

“…Deposition of inert colloids is also significant in water disinfection, since microbes and inert colloids exhibit important similarities in saturated porous granular media, as stated by Johnson et al (2007) [1], who observed that hydrodynamic drag mitigates deposition and drives re-entrainment of both biological and non-biological colloids.…”

“…Regarding the implementation of the particle mechanics immersed in water, the model described does not include hydraulic retardation between particles, which would be required for high particle concentrations. The implementation of surface interaction, surface roughness and heterogeneity has not been taken into account, although it was argued to have significant effect on particle deposition [24,25].…”