A Probabilistic Formulation of the Diffusion Coefficient in Porous Media as Function of Porosity

TL;DR

This paper introduces a probabilistic approach to model the diffusion coefficient in porous media based on pore structure variability, emphasizing the importance of PDFs over deterministic functions for accurate transport predictions.

Contribution

It presents a novel statistics-based upscaling method using PDFs of porosity to better predict macroscopic transport properties from pore-scale data.

Findings

Transport ability PDFs capture pore structure effects.

Scatter in transport ability converges to a non-zero constant.

Using PDFs improves the description of transport parameters.

Abstract

We investigate the upscaling of diffusive transport parameters as function of pore scale material structure using a stochastic framework. We focus on sub-REV (representative elementary volume) scale where the complexity of pore space geometry leads to a significant scatter of transport observations. We study a large data set of sub-REV measurements on porosity and transport ability being a dimensionless parameter representing the ratio of diffusive flow through the porous volume and through an empty volume. We characterize transport ability as probability distribution functions (PDFs) of porosity capturing the effect of pore structure differences among samples. We then investigate domain size effects and predict the REV scale. While scatter in porosity observation decrease linearly with increasing sample size, the observed scatter in transport ability converges towards a constant value larger zero. Our results confirm that differences in pore structure topology impact transport parameters at all scales. Consequently, the use of PDFs to describe the relationship of effective transport coefficients to porosity is advantageous to deterministic semi-empirical functions. We discuss the consequences and advocate the use of PDFs for effective parameters in both continuum equations and data interpretation of experimental or computational work. We believe that the presented statistics-based upscaling technique of sub-REV microscopy data provides a new tool in understanding, describing and predicting macroscopic transport behavior of micro-porous media.