An image- and BET-based Monte-Carlo approach to determine mineral accessible surface areas in sandstones

Abstract

Accessible surface areas (ASAs) of individual rock-forming minerals exert a fundamental control on how minerals react with formation fluids. However, due to lacking adequate quantification methods, determining the ASAs of specific minerals in a multi-mineral rock at the appropriate scale still remains difficult. Whole-rock Brunauer-Emmett-Teller (BET) measurements at atomic scales cannot account for the variability in ASAs of individual minerals, while image-based methods are inherently limited by the image pixel/voxel resolution. Here, we present a novel joint method that overcomes the aforementioned limitations by appropriately downscaling individual image-based ASAs with the support of a Monte-Carlo algorithm and BET measurements. This joint method consists of three steps: 1) segmentation of pore/mineral phases, 2) calculation of image-based contour surface areas, and 3) determination of a resolution scaling factor (SF). Superior to conventional segmentation methods, which are based on scanning electron microscopy (SEM) images alone, here, the segmentation threshold is independently constrained by both pore size distribution measurements and rock chemical composition analyses. Most importantly, the introduction of an SF, obtained by probability mapping, using a Monte-Carlo algorithm, accounts for mineral surface roughness variations, provided by the BET measurements during the determination of the image-based ASAs. We apply this joint method to a sandstone specimen and confirm the validity and accuracy of the obtained results with our reactive flow-through experiment, reported in Ma et al., 2019. We conclude that our novel method can effectively downscale the image-based ASAs to the atomic BET resolution with minimum assumptions, providing a valuable tool to improve the calculation of fluid-mineral reactions.