Abstract

In most conceptual models of dissolution trapping of CO2, it is assumed that mixing of dissolved supercritical CO2 and formation brine occurs through density-driven convection. In our previous modeling study, we showed that the presence of continuous low-permeability shale layers in the formations causes convective shutdown through disruption of fingers, which impacts the effectiveness of mixing and trapping processes. However, these layers are naturally heterogeneous due to variations in compositional and textural properties. In the present study, we investigate the potential effects of heterogeneity present within semi-confining low-permeability layers on the overall mixing and trapping of dissolved CO2. Since accurate field experimentation in deep geologic formations is difficult due to inability to adequately characterize geology and define boundary conditions of the formation, we designed well-controlled experiments using NaBr solution and water under laboratory conditions. We conducted intermediate-scale 3D laboratory experiments under two homogeneous and one heterogeneous multilayered sand packing configurations. The results of these experiments show that connectivity of relatively higher permeability material within the semi-confining low-permeability layers contributes to mixing through (1) brine leakage between upper and lower aquifers (reverse convection), and (2) trapping through diffusion and back diffusion of initially trapped mass due to reversed concentration gradients in the long term. These findings suggest that when estimating the effectiveness of dissolution trapping in CO2 sequestration one needs to consider possible deviations from the traditional convective mixing theory in homogeneous media. Under some conditions of natural formation heterogeneity, diffusion can contribute to trapping; in some others, it may not.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call