Geochemical modelling of water-rock interaction

Abstract

CO2 geological storage is one of the most promising technologies for reducing atmospheric emissions of greenhouse gas. In this work we present and discuss a new approach geochemical modelling for evaluating the effects of short-medium term CO2 disposal in deep geologic formations that has been tested in the Weyburn test site (Saskatchewan, Canada), where since September 2000 5000 t/day of supercritical CO2 are injected. The geochemical modeling has been performed by using the code PRHEEQC (V2.11) software package, via thermodynamic corrections to the code default database. First, we reconstructed the in-situ reservoir (62°C and 0.1 MPa) chemical composition, including pH, by the chemical equilibrium among the various phases, and we evaluated the boundary conditions (e.g. PCO2 , PH2S), which are necessary for the implementation of reaction path modeling. This is the starting point to assess the geochemical impact of CO2 into the oil reservoir and, as main target, to quantify water-gas-rock reactions. Furthermore, we identified possible compositions of the initially reservoir liquid phases by assuming the equilibrium conditions for the mineral assemblage with respect to a Na-Cl water (Cl/Na=1.2). Then we computed the kinetic evolution of the CO2-rich Weyburn brines interacting with the host-rock minerals, performed over 100 years after injection. Results of reaction path modeling suggest that, in this period, CO2 can be neutralized by solubility (as CO2 (aq)) and mineral trapping through Dawsonite precipitation. In order to validate our geochemical model we have simulated the geochemical impact of three years of CO2 injection (September 2000-2003) by kinetically controlled reactions and we have compared the computed and measured data. The calculated chemical composition after the CO2 injection is consistent with the analytical data of samples collected in 2003 with an error within 5 % for most analytical species, with the exception of the Ca and Mg contents (error > 90%), likely due to the complexation effect of carboxilic acid.