(397c) Effects of Morphological Properties on the Permeability of Reconstructed Porous Media

Predicting the transport properties of porous media from their fundamental morphology is a long-standing challenge. A number of functions are available to characterize the microscopic structure of porous media, such as two-point autocorrelation, chord-length distribution, Minkowski functionals, and the burn-number distribution. In this paper we use pore-network modeling to examine their effects on the permeability of reconstructed porous media. First the reference values of these functions are extracted from digital images of real porous media, and are honored in the reconstruction process, where we employ a stochastic optimization method. Subsequent flow simulations are performed using a pore-network model in these reconstructed media to reveal: (1) the impact of the optimization parameters on microscopic morphological properties (e.g., pore-size distribution and pore coordinate number distribution); and (2) macroscopic transport properties, i.e. permeability. By selecting different combinations of the aforementioned morphological functions and comparing the results to those of the real materials, we are able to identify which function carries the most valuable information in terms of preserving the correct permeability.