Experimental insights into the geochemistry and mineralogy of a granite-hosted geothermal system injected with supercritical CO2

Supercritical CO₂ may be a viable working fluid in enhanced geothermal systems (EGS) due to its large expansivity, low viscosity, and reduced reactivity with rock as compared to water. Hydrothermal experiments are underway to evaluate the geochemical impact of using supercritical CO₂ as a working fluid in granite-hosted geothermal systems. Synthetic aqueous fluid and a model granite are reacted at 250 °C and 250 bars in a rocking autoclave and Au-Ti reaction cell for a minimum of 28 days (water:rock ratio of approximately 20:1). Subsequent injection of supercritical CO₂ increases pressure, which decays over time as the CO₂ dissolves into the aqueous fluid. Initial experiments decreased to a steady state pressure of 450 bars approximately 14 hours after injection of supercritical CO₂. Post-injection reaction is allowed to continue for at least an additional 28 days. Excess CO₂ is injected to produce a separate supercritical fluid phase (between 1.7 and 3.1 molal), ensuring aqueous CO₂ saturation for the duration of each experiment. The granite was created using mineral separates and consists of ground (75 wt%, <45 microns) and chipped (25 wt%, 0.5-1.0 cm), sub-equal portions of quartz, perthitic potassium feldspar (~ 25 wt% albite and 75 wt% potassium feldspar), oligoclase, and a minor (4 wt%) component of Fe-rich biotite. The synthetic saline water (I = 0.12 m) contains molal quantities of Na, Cl, and HCO₃ and millimolal quantities of K, SiO₂, SO₄, Ca, Al, and Mg, in order of decreasing molality. Aqueous fluids are sampled approximately 10 times over the course of each experiment and analyzed for total dissolved carbon and sulfide by coulometric titration, anions by ion chromatography, and major, minor, and trace cations by ICP-OES and -MS. Bench pH measurements are paired with aqueous analyses to calculate in-situ pH. Solid reactants are evaluated by SEM-EDS, XRD, and/or bulk chemical analysis before and after each experiment. Analytical data are reviewed alongside geochemical models to evaluate fluid-rock interactions and the capacity of theoretical models to predict the observed outcome. Data derived from this study will inform our understanding of how a real world geothermal system may respond geochemically and mineralogically given 'spontaneous' injection of CO₂, whether by an anthropogenic or natural source. Companion modeling work is also underway, which will use these experiments to calibrate EGS models for field application.