Pore‐space structure and average dissolution rates: A simulation study

Abstract

We study the influence of the pore‐space geometry on sample‐averaged dissolution rates in millimeter‐scale carbonate samples undergoing reaction‐controlled mineral dissolution upon the injection of a CO2‐saturated brine. The representation of the pore space is obtained directly from micro‐CT images with a resolution of a few microns. Simulations are performed with a particle tracking approach on images of three porous rocks of increasing pore‐space complexity: a bead pack, a Ketton oolite, and an Estaillades limestone. Reactive transport is simulated with a hybrid approach that combines a Lagrangian method for transport and reaction with the Eulerian flow field obtained by solving the incompressible Navier‐Stokes equations directly on the voxels of three‐dimensional images. Particle advection is performed with a semianalytical streamline method and diffusion is simulated via a random walk. Mineral dissolution is defined in terms of the particle flux through the pore‐solid interface, which can be related analytically to the batch (intrinsic) reaction rate. The impact of the flow heterogeneity on reactive transport is illustrated in a series of simulations performed at different flow rates. The average dissolution rates depend on both the heterogeneity of the sample and on the flow rate. The most heterogeneous rock may exhibit a decrease of up to two orders of magnitude in the sample‐averaged reaction rates in comparison with the batch rate. Furthermore, we provide new insights for the dissolution regime that would be traditionally characterized as uniform. In most cases, at the pore‐scale, dissolution preferentially enlarges fast‐flow channels which greatly restricts the effective surface available for reaction.

Evolution of permeability and microstructure of tight carbonates due to numerical simulation of calcite dissolution

“…We numerically simulated calcite dissolution using a phenomenological approach. Rather than solving the advection-diffusion equation directly [e.g., Jiang and Tsuji, 2014;Pereira Nunes et al, 2016a, 2016bSoulaine et al, 2016], our model assumes the local velocity field as a proxy for solvent flux. By doing so, we implicitly ignore diffusion, which we argue is a valid simplification for a transported-limited, high-flow dissolution regime where solvents are advected through the pore geometry faster than they can change the chemistry of the fluid due to reaction with the pore wall.…”

Mineralogical and transport controls on the evolution of porous media texture using direct numerical simulation

“…As noted by Pereira Nunes et al . [], the evolution of a porous medium does not depend only on the reactive transport regime (e.g., as expressed by the Péclet and Damköhler numbers) but also on the initial pore structure of the medium. In the simulation presented here, only one of the three domains considered showed an unstable evolution (this domain initially had a relatively fast flow path).…”

Reactive Flow and Homogenization in Anisotropic Media

“…The explicit but simplified representation of void space in network models allows studying the basic effect of heterogeneity and anisotropy of the medium (Friedman & Jones, 2001;Jang et al, 2011;Steefel et al, 2013;Xiong et al, 2016), while their computational efficiency enables the use of large and representative domains and repetition over multiple realizations. This is in contrast to more accurate methods, such as direct numerical simulations, that are much more restricted by their computational demand (Liu & Mostaghimi, 2017;Molins, 2015;Pereira Nunes et al, 2016). The model comprises a square of length L of 2-D regular network of cylindrical channels and nodes (junctions) in a soluble solid.…”