Approximating heterogeneous colloidal transport by n-population filtration models

Abstract

This study evaluates the effectiveness of n-particle filtration models in approximating the behaviour of heterogeneous colloidal suspension flows in porous media. Using a numerical investigation, the study examines binary (n = 2) and ternary (n = 3) particle models to approximate breakthrough curves and retention profiles of systems with varying filtration coefficient distributions. The results show that binary and ternary models can effectively replicate the behaviour of systems with lower heterogeneity, where the ratio of maximum to minimum filtration coefficients remains moderate. However, as heterogeneity increases (higher coefficients of variation), the accuracy of lower-particle models decreases, and more particles are required to maintain model fidelity. The study highlights practical guidelines for model selection, recommending binary models for moderate heterogeneity and ternary models for high heterogeneity. It cautions against oversimplifying highly heterogeneous systems with low-particle models and provides a framework for iterative model refinement based on experimental data. This work offers a robust approach for modelling particle transport and retention in porous media, with applications in aquifer recharge, oil recovery, and environmental engineering.