Quantifying CO2 capillary heterogeneity trapping through macroscopic percolation simulation

Abstract

Capillary heterogeneity trapping caused by mesoscale heterogeneity at the millimeter scale has been shown to have a great potential in immobilizing a significant amount of CO₂ in CO₂ geologic storage. The goal of this study is to develop and apply a macroscopic percolation simulator to better understand how post-imbibition CO₂ capillary heterogeneity trapping varies with different types and degrees of mesoscale heterogeneity as well as to quantify and compare its contribution to the pore-scale snap-off/bypass residual trapping mechanism. Using invasion percolation during drainage and ordinary percolation with macroscopic trapping during imbibition, the macroscopic percolation simulator can rapidly simulate both post-drainage and post-imbibition CO₂ saturation fields under capillary-gravity equilibrium, producing results that are validated by a full-physics reservoir simulator with a speedup of 10 to 100 times. The macroscopic percolation simulation results have shown that for domains that are completely uncorrelated or with nonleaky downstream capillary barriers, CO₂ heterogeneity trapping contribution always increases with the degree of heterogeneity of the domain. This is consistent with previous experimental results. However, the relationship between the CO₂ residual trapping ability of a domain also depends on the spatial structure of the heterogeneity and trapping ability does not always increase linearly with the degree of heterogeneity as previously shown. Additionally, when the capillary barrier is leaky, then regardless of the degree of heterogeneity, no extra capillary heterogeneity trapping can be achieved after long-term imbibition. Finally, this study demonstrates how residual trapping relationships may not be scale-invariant and calls for more sophisticated upscaling methods such as the macroscopic percolation simulation method presented here.