DISCREPANCIES BETWEEN LABORATORY AND FIELD-BASED MINERAL-FLUID REACTION RATES: IMPLICATIONS FOR CO2 SEQUESTRATION BASED on REACTIVE TRANSPORT STUDIES OF NATURAL SYSTEMS

Over the last three decades it has been increasingly recognized that the rates of fluid-mineral reactions measured in natural systems are generally several orders of magnitude (10² to 10⁵) slower than those determined experimentally. The extreme variability in natural reaction rates has posed a long-standing problem in aqueous geochemistry as it creates substantial uncertainty in the kinetic parameters incorporated into interpretive and predictive reactive transport models, such as those used to assess the consequences of CO₂ injection in the subsurface. Growing evidence suggests that the apparent rate discrepancy may be due to the non-linear behavior of mineral dissolution rates near equilibrium, combined with pervasive transport-control of reaction rates. Transport-control suggests that the fluid reaches equilibrium with the solid phase over the length scale under consideration, such that aqueous transport is the rate-limiting factor. If the reactive transport system is transport-controlled, mineral surface areas and kinetic rate constants determine the geometry of the reaction front, but these parameters do not have a substantial influence on the propagation of the reaction front or the net transformation rate. However, factors that affect thermodynamic departure from equilibrium of the dissolving phase(s), such as the rates of secondary mineral precipitation, the effect of secondary mineral coatings on dissolution of primary minerals, and the evolution of pH, flow rate, reactive surface area, and porosity/permeability will determine the rate at which reaction fronts propagate. A reactive transport analysis of the reaction between aqueous fluids equilibrated with supercritical CO₂ and both arkosic sandstone and ultramafic rocks suggests that the nature of mineral-fluid reactions is likely to range from surface-reaction controlled to transport-controlled. We apply these results to coupled experimental studies of a Mg-silicate—H₂O—CO₂ system and a natural analog of CO₂ sequestration in serpentinized ultramafic rocks. In these rocks, carbonate mineralization appears as both metasomatic replacements and massive vein fillings, producing an economic mineral deposit of Mg-carbonate at the Red Mountain Magnesite District in the Del Puerto Ophiolite of California.