Abstract

This study aims to assess the impact of relative permeability hysteresis and capillary pressure on the long-term sequestration of CO2 in sandstone aquifers. The scope includes evaluating the evolution of CO2 trapping mechanisms, including dissolution, residual trapping, mineralization, and their variations over time. We employ a fine-scale numerical model based on a hypothetical siliciclastic reservoir to simulate 1000 years, including ten years of active injection followed by a monitoring phase. The effect of relative permeability hysteresis and capillary pressure on CO2 trapping is analyzed for different time periods. The results reveal that capillary pressure is the most significant parameter that significantly increases the desired trapping mechanisms, such as residual, dissolution, and mineralization. At the end of 1000 years, a negligible fraction of CO2 exists in the mobile phase when capillary pressure is included in the simulation. However, almost 50% of the injected CO2 remains in the mobile phase when only relative permeability hysteresis is included in the model. Over the long term, dissolution trapping becomes the most dominant trapping mechanism, with mineralization contributing only a small fraction of sequestered CO2. The impact of relative permeability hysteresis and capillary pressure on different trapping mechanisms is quantified, emphasizing their importance in predicting CO2 behavior. This study provides critical aspects of CO2 storage behavior and the impact of critical factors like relative permeability hysteresis and capillary pressure. The findings highlight the complexities of CO2 trapping mechanisms over extended timeframes and demonstrate the importance of these factors in optimizing carbon capture and storage strategies. This research offers new information that can benefit geoscientists in improving the accuracy of CO2 storage predictions and the safety of geological carbon sequestration.

This content is only available via PDF.
You can access this article if you purchase or spend a download.