Controls of sluggish, CO2-promoted, hematite and K-feldspar dissolution kinetics in sandstones

CO2 sequestration is regarded as an important strategy for reducing anthropogenic CO2 emissions. Both the nature and rate of fluid–mineral reactions in CO2–water–rock systems are crucial, yet poorly constrained, parameters in understanding the fate of CO2 injected in geological formations. This study models reactions and reaction rates in an exhumed CO2-charged aquifer where CO2-rich brines have bleached red sandstones by dissolution of hematite grain coatings. We show that the vertical movement of the reaction front is dominated by diffusion and this allows calculation of reaction rates for the dissolution of hematite grain coatings and K-feldspar. Using mineral surface areas calculated from BET measurements, we estimate K-feldspar dissolution rates of 3.4±2.0×10−13molm−2s−1; and hematite reaction rates of 6.1±5.0×10−16molm−2s−1. The rates for K-feldspar are lower than previous, experimentally derived, estimates of K-feldspar dissolution rates by 1–2 orders of magnitude, likely explained by the proximity of the natural system to equilibrium. The inferred hematite reaction rates are 5–6 orders of magnitude slower than laboratory experiments and appear to be controlled by the chemical gradients imposed by the more sluggish K-feldspar dissolution. As the majority of potentially mobile trace metals are hosted in iron-oxide grain coatings, we argue that the rate of contaminant mobilization by CO2-charged brines will be lower than suggested by laboratory experiments.

► We model dissolution rates of K-feldspar and hematite in a natural CO2 reservoir.
► Vertical transport in the system is dominated by diffusion.
► Calculated kinetic constants are lower than predicted from laboratory experiments.
► Hematite reaction rates are inhibited by the rate of silicate dissolution.
► Contaminant release from oxide grain coatings will be slower than models suggest.