Visualization and prediction of supercritical CO2 distribution in sandstones during drainage: An in situ synchrotron X-ray micro-computed tomography study

Abstract

Pore-scale distribution of supercritical CO2 (scCO(2)) exerts significant control on a variety of key hydrologic as well as geochemical processes, including residual trapping and dissolution. Despite such importance, only a small number of experiments have directly characterized the three-dimensional distribution of scCO(2) in geologic materials during the invasion (drainage) process. We present a study which couples dynamic high-resolution synchrotron X-ray micro-computed tomography imaging of a scCO(2)/brine system at in situ pressure/temperature conditions with quantitative pore-scale modeling to allow direct validation of a pore-scale description of scCO(2) distribution. The experiment combines high-speed synchrotron radiography with tomography to characterize the brine saturated sample, the scCO(2) breakthrough process, and the partially saturated state of a sandstone sample from the Domengine Formation, a regionally extensive unit within the Sacramento Basin (California, USA). The availability of a 3D dataset allowed us to examine correlations between grains and pores morphometric parameters and the actual distribution of scCO(2) in the sample, including the examination of the role of small-scale sedimentary structure on CO2 distribution. The segmented scCO(2)/brine volume was also used to validate a simple computational model based on the local thickness concept, able to accurately simulate the distribution of scCO(2) after drainage. The same method was also used to simulate Hg capillary pressure curves with satisfactory results when compared to the measured ones. This predictive approach, requiring only a tomographic scan of the dry sample, proved to be an effective route for studying processes related to CO2 invasion structure in geological samples at the pore scale.