H53H-1779
QUANTIFYING CONDITIONS FOR FAULT ‘SELF-SEALING’ IN GEOLOGIC CARBON SEQUESTRATION

Friday, 18 December 2015
Poster Hall (Moscone South)
Vivek Patil, Energy and Geoscience Institute, Salt Lake City, UT, United States
Abstract:
Injecting anthropogenic CO2 into a subsurface reservoir for sequestration will impact the reservoir significantly, including its geochemistry, porosity and permeability. If a fault or fracture penetrates the reservoir, CO2-laden brine may migrate into that fault, eventually sealing it via precipitation or opening it up via dissolution. The goal of this study was to identify and quantify such conditions of fault self-sealing or self-enhancing. We found that the dimensionless Damköhler number (Da), the ratio of reaction rate to advection rate, provides a meaningful framework for characterizing the propensity of (fault) systems to seal or open up. We developed our own framework wherein Damköhler numbers evolve spatiotemporally as opposed to the traditional single Da value approach. Our approach enables us to use the Damköhler for characterization of complex multiphase and multimineral reactive transport problems. We applied this framework to 1D fault models with eight conditions derived from four geologic compositions and two reservoir conditions. The four­ geologic compositions were chosen such that three out of them were representative of distinct geologic end-members (sandstone, mudstone and dolomitic limestone) and one was a mixed composition based on an average of three end-member compositions. The two sets of P-T conditions chosen included one set corresponding to CO2 in a gaseous phase (“shallow conditions”) and the other corresponding to supercritical phase CO2 (“deep conditions”). Simulation results suggest that fault sealing via carbonate precipitation was a possibility for shallow conditions within limestone and mixed composition settings. The concentration of cations in the water was found to be an important control on the carbonate precipitation. The deep conditions models did not forecast self-sealing via carbonates. Sealing via clay precipitation is a likely possibility, but the 1000 year time-frame may be short for such. Model results indicated a range of Da values within which substantial reductions of fault porosity (meaning self-sealing) could be expected. A key conclusion suggested by the results of this study is that carbonate precipitation in the near-surface (top ~50-100 m) depths of a fault is the most likely mechanism of “self-sealing” for most geological settings.