A Probability-Based Pore Network Model of Particle Jamming in Porous Media

Geometric straining of particles in porous media is of critical importance in a broad range of natural and industrial settings, such as contaminant transport in aquifers and the permeability decline due to pore plugging in oil reservoirs. Pore network modeling is a computationally efficient approach to simulate transport problems in porous media that has been used to simulate particle deposition and size exclusion. However, existing network models are unable to simulate particle jamming due to the simplification of geometry and the lack of capability for simulating particle–particle interactions. Here, we develop a novel pore network model for particle jamming in porous media. The jamming in pore throats is predicted by the probability of jamming, which is a function of pore/particle size ratio, particle concentration, and coefficient of friction (COF). A unified probability model is developed based on Discrete Element Method (DEM) simulations of particle jamming in porous media consisting of a single layer of spherical grains. The geometry is based on two packing extremes, those with grains arranged in a triangle and a square. The model is then implemented into the pore network model to predict jamming in porous media. This framework combines the efficiency of network modeling and the accuracy of direct simulations on particle jamming. We verify the model against CFD-DEM simulations. The results of network simulations show that the probability of jamming increases with COF. At low particle concentrations (C = 5%), the probability of jamming for large pore throats is small. There is a critical particle concentration (C = 9%) when the grain/particle size ratio is 10, above which jamming is self-reinforcing and the inlet face of the porous medium will be completely blocked (C = 11%). The effect of velocity on the probability of jamming is insignificant compared with the effects of particle concentration and pore/particle size ratio.