The rock matrix of granites is expected to be an important buffer against the dispersion of contaminants, e.g. radionuclides, and against the ingress of oxygenated glacial meltwater. The influence of matrix heterogeneity on O2 diffusive transport is assessed here by means of numerical experiments based on a micro-Discrete Fracture Network (micro-DFN) representation of the diffusion-available pore space along with random realisations of idealized biotite grains, to simulate the heterogeneous nature of granitic rocks. A homogeneous-based analytical solution is also presented and used to assess possible deviations of the numerical experiments from the assumption of homogeneity. The analytical solution is also used to test upscaled values of mineral surface area. The numerical experiments show that the matrix behaves as a composite system, with the coexistence of fast and slow diffusive pathways. This behavior is more evident at low Damköhler numbers. Our interpretation of the numerical experiments points out the importance to properly characterise the heterogeneity of the rock matrix.
Keywords: High performance computing; Mineralogical heterogeneity; Oxygen transport; micro-DFN.
Copyright © 2018 The Authors. Published by Elsevier B.V. All rights reserved.