Thin Water Films Enable Low-Temperature Magnesite Growth Under Conditions Relevant to Geologic Carbon Sequestration.

Injecting supercritical CO2 (scCO2) into basalt formations for long-term storage is a promising strategy for mitigating CO2 emissions. Mineral carbonation can result in permanent entrapment of CO2; however, carbonation kinetics in thin H2O films in humidified scCO2 is not well understood. We investigated forsterite (Mg2SiO4) carbonation to magnesite (MgCO3) via amorphous magnesium carbonate (AMC; MgCO3·xH2O, 0.5 < x < 1), with the goal to establish the fundamental controls on magnesite growth rates at low H2O activity and temperature. Experiments were conducted at 25, 40, and 50 °C in 90 bar CO2 with a H2O film thickness on forsterite that averaged 1.78 ± 0.05 monolayers. In situ infrared spectroscopy was used to monitor forsterite dissolution and the growth of AMC, magnesite, and amorphous SiO2 as a function of time. Geochemical kinetic modeling showed that magnesite was supersaturated by 2 to 3 orders of magnitude and grew according to a zero-order rate law. The results indicate that the main drivers for magnesite growth are sustained high supersaturation coupled with low H2O activity, a combination of thermodynamic conditions not attainable in bulk aqueous solution. This improved understanding of reaction kinetics can inform subsurface reactive transport models for better predictions of CO2 fate and transport.

Quantification of CO2 Mineralization at the Wallula Basalt Pilot Project.

In 2013, the Pacific Northwest National Laboratory led a geologic carbon sequestration field demonstration where ∼1000 tonnes of CO2 was injected into several deep Columbia River Basalt zones near Wallula, Washington. Rock core samples extracted from the injection zone two years after CO2 injection revealed nascent carbonate mineralization that was qualitatively consistent with expectations from laboratory experiments and reactive transport modeling. Here, we report on a new detailed analysis of the 2012 pre-injection and 2015 post-injection hydrologic tests that capitalizes on the difference in fluid properties between scCO2 and water to assess changes in near-field, wellbore, and reservoir conditions that are apparent approximately two years following the end of injection. This comparative hydrologic test analysis method provides a new way to quantify the amount of injected CO2 that was mineralized in the field test. Modeling results indicate that approximately 60% of the injected CO2 was sequestered via mineralization within two years, with the resulting carbonates occupying ∼4% of the available reservoir pore space. The method presented here provides a new monitoring tool to assess the fate of CO2 injected into chemically reactive basalt formations but could also be adapted for long-term monitoring and verification within more traditional subsurface carbon storage reservoirs.