Coupled geomechanical-geochemical aspects of CO2-sequestration in peridotites

Mineralization of carbon dioxide by subsurface reaction with olivine-rich peridotite has recently been proposed as a potential CO2-sequestration method. The abundance of olivine rocks at and near the Earth’s surface and the stability of the product mineral phases, notably magnesite, make this an interesting option, especially as the in-situ temperatures and CO2-injection pressures will strongly enhance olivine dissolution and hence carbonation rates. However, geological sequestration of CO2 in dense, crystalline peridotites involves considerable challenges. First, to trap CO2 by injection and in-situ mineralization, fracture porosity and permeability must be created by fracturing the host rock. Second, the reaction between CO2-rich hydrous fluids and olivine, to form magnesite and silica, involves a significant increase in solid volume (~84%). Mineralization of CO2 may therefore result in the filling and sealing of the fractures, which limits the volume of CO2 that can be injected. In addition, ongoing dissolution of olivine from fracture surfaces may lead to a decrease in fracture surface reactivity as less reactive minerals (serpentine, pyroxene) begin to dominate the surface composition. To maintain high mineralization rates, then, new fracture surface area needs to be created continuously. A key question here is whether reaction driven fracturing will occur as the result of the solid volume expansion associated with the precipitation of product phases such as MgCO3, through the development of a force of crystallization. We report experimental research that characterizes the evolution of dissolving peridotite (fracture) surfaces and the development of a force of crystallization due to product phase precipitation. Our findings for mesh-serpentinized peridotite show that rapid dissolution rates at fracture surfaces are maintained as the serpentine mesh, rather than inhibiting dissolution, provides fluid pathways into the peridotite rock that promote olivine dissolution in an advancing front beyond the fracture surface. To assess whether the in-situ stress and tensile strength of the rock can be overcome by stress development, as was predicted by thermodynamic modeling, we conducted experiments under confined conditions to measure the force of crystallization resulting from olivine carbonation. In most of our experiments, no force of crystallization developed, as precipitation led to pore occlusion, grain boundary cementation and the inhibition of CO2-transport within the peridotite. This implies that slow diffusive transport of CO2 through the fracture network to crack tips will be the limiting factor in subsurface mineralization. Estimates from our work suggest a resulting reaction-driven fracture-propagation rate of 1m per 300-6000 year, which is too slow for sequestration purposes.