Density driven flow in CO2 storage sites: A new formulation for heterogeneous layered porous media

Increasing levels of greenhouse gases (GHG) in the atmosphere have led to the need for effective carbon capture and storage (CCS) technologies. This method involves capturing CO2 at key emission sites and injecting it into suitable geological formations such as aquifers and depleted reservoirs for permanent storage. However, the efficiency of storage is strongly influenced by the behavior of the CO2, the complexity of the geological porous media and the contributions of the different trapping mechanisms. In this study, the phenomenon of density-driven convective mixing of CO2 in heterogeneous porous media is investigated. A new formulation was presented that includes heterogeneity and anisotropy in the vertical and lateral directions. The stream function was also modified accordingly, and the evolution of convective fingers was incorporated into the proposed formulations. Data from the Cook Formation, a potential storage site of the Northern Lights Project in Norway, was used given its heterogeneity and the variation of permeability in the upper and lower sections. Six scenarios were considered to better understand the impact of heterogeneity, anisotropy and layering effects on the onset time and behavior of convection currents. It was revealed that the presence of a low permeable layer above a higher permeability layer significantly increases the onset time and decreases the amount of dissolved CO2 compared to the homogeneous case. It was also observed that permeability variations significantly influence the formation and stability of CO2 convection patterns and affect the velocity, size, and direction of the fingers. For instance, the amount of dissolved CO2 in the heterogeneous and anisotropic case was 10.019 tons, while the value for the homogeneous and isotropic case was 27.570 tonnes. The results of this work have potential implications for the optimization of CCS strategies in different geological settings. © 2024 Elsevier Ltd