Synthesis of equilibrated geochemical systems using extended Debye-Huckel and Pitzer activity models for enhanced CO2 storage modelling

Abstract

Carbon capture and storage is a critical technology for reducing greenhouse gas emissions and mitigating climate change. Ensuring the safe, long-term CO₂ storage in geological formations requires accurate modelling of geochemical reactions between CO₂-saturated water and rock-forming minerals. Reactive-transport simulators represent these processes over extended timescales, but geochemical equilibrium must first be established, analogous to gravitational equilibrium in pressure initialization. This study presents a practical workflow for synthesizing equilibrated CO₂-rock-water systems, demonstrated for the Bunter Sandstone Formation. To ensure realistic initial pressure distributions that govern pressure-dependent trapping processes, gravitational equilibrium was first established. The mineralogy was then engineered to maintain a non-negative degree of freedom for chemically consistent equilibrium calculations. Long-term batch simulations using ideal, extended Debye-Huckel, and Pitzer activity models revealed significant discrepancies between activity-model-based equilibrium concentrations and short-term laboratory values, even though predictions of salinity and pH were consistent. These discrepancies highlight the importance of deriving equilibrium concentrations from long-term simulations for chemical initialization, as short-term laboratory measurements may not reflect true equilibrium conditions. The Pitzer model provided the most accurate predictions under high salinity but increased simulation time by over 100%, whereas the extended Debye-Huckel model required only 30% additional time but neglected short-range ionic interactions. The reduced-salinity scenario decreased equilibrium concentrations by approximately 20–100%, enhancing CO₂ dissolution and promoting mineral dissolution, thereby influencing structural, solubility, and mineral trapping mechanisms. These findings underscore the importance of careful activity model selection and accurate salinity characterization to balance computational efficiency with predictive accuracy and improve confidence in equilibrium predictions.