Pore-Scale Geochemical Reactivity Associated with CO2 Storage: New Frontiers at the Fluid-Solid Interface.

Abstract

The reactivity of carbonate and silicate minerals is at the heart of porosity and pore geometry changes in rocks injected with CO2, which ultimately control the evolution of flow and transport properties of fluids in porous and/or fractured geological reservoirs. Modeling the dynamics of CO2-water-rock interactions is challenging because of the resulting large geochemical disequilibrium, the reservoir heterogeneities, and the large space and time scales involved in the processes. In particular, there is a lack of information about how the macroscopic properties of a reservoir, e.g., the permeability, will evolve as a result of geochemical reactions at the molecular scale. Addressing this point requires a fundamental understanding of how the microstructures influence the macroscopic properties of rocks. The pore scale, which ranges from a few nanometers to centimeters, has stood out as an essential scale of observation of geochemical processes in rocks. Transport or surface reactivity limitations due to the pore space architecture, for instance, are best described at the pore scale itself. It can be also considered as a mesoscale for aggregating and increasing the gain of fundamental understanding of microscopic interfacial processes. Here we focus on the potential application of a combination of physicochemical measurements coupled with nanoscale and microscale imaging techniques during laboratory experiments to improve our understanding of the physicochemical mechanisms that occur at the fluid-solid interface and the dynamics of the coupling between the geochemical reactions and flow and transport modifications at the pore scale. Imaging techniques such as atomic force microscopy, vertical scanning interferometry, focused ion beam transmission electron microscopy, and X-ray microtomography, are ideal for investigating the reactivity dynamics of these complex materials. Minerals and mineral assemblages, i.e., rocks, exhibit heterogeneous and anisotropic reactivity, which challenges the continuum description of porous media and assumptions required for reactive transport modeling at larger scales. The conventional approach, which consists of developing dissolution rate laws normalized to the surface area, should be revisited to account for both the anisotropic crystallographic structure of minerals and the transport of chemical species near the interface, which are responsible for the intrinsic evolution of the mineral dissolution rate as the reaction progresses. In addition, the crystal morphology and the mineral assemblage composition, texture, and structural heterogeneities are crucial in determining whether the permeability and transport properties of the reservoir will be altered drastically or maintain the sealing properties required to ensure the safe sequestration of CO2 for hundreds of years. Investigating the transport properties in nanometer- to micrometer-thick amorphous Si-rich surface layers (ASSLs), which develop at the fluid-mineral interface in silicates, provides future direction, as ASSLs may prevent contact between the dissolving solids and the pore fluid, potentially inhibiting the dissolution/carbonation process. Equally, at a larger scale, the growth of micrometer- to millimeter-thick alteration layers, which result from the difference in reactivity between silicates and carbonates, slows the transport in the vicinity of the fluid-solid interface in polymineralic rocks, thus limiting the global reactivity of the carbonate matrix. In contrast, in pure limestone, the global reactivity of the monomineralic rock decreases because the flow localization promotes the local reactivity within the forming channels, thus enhancing permeability changes compared with more homogeneous dissolution of the rock matrix. These results indicate that the transformation of the rock matrix should control the evolution of the transport properties in reservoirs injected with CO2 to the same extent as the intrinsic chemical reactivity of the minerals and the reservoir hydrodynamics. This process, which is currently not captured by large-scale modeling of reactive transport, should benefit from the increasing capabilities of noninvasive and nondestructive characterization tools for pore-scale processes, ultimately constraining reactive transport modeling and improving the reliability of predictions.