An investigation into the upscaling of mineral dissolution from the pore to the core scale

Studying the behavior of mineral dissolution has practical uses in Carbon Capture and Storage (CCS) and Improved Oil Recovery (IOR), and several numerical models are striving to simulate the process accurately. In this paper, we investigate the core-scale numerical model presented by Golfier et al. (J. Fluid Mecha., 2002), which uses the Darcy-Brinkman-Stokes (DBS) flow formulation. This model uses a simplified Kozeny-Carmen formulation and a simplified linear formulation to describe the evolution of permeability and mass exchange coefficient as a function of porosity at the core-scale, respectively. This assumption is equivalent to neglecting the impact of pore-scale non-uniform dissolution on the prediction of the dissolution processes overall behavior. However, recent pore-scale dissolution studies have observed many different dissolution regimes, which calls into question the accuracy of the assumptions in Golfier et al. (2002)'s model. To investigate the legitimacy of this assumption, we first used our inhouse pore-scale numerical simulator (GeoChemFoam) to observe the dissolution in the uniform and dominant wormhole regimes at the pore-scale, and then developed representative permeability (K) and mass exchange coefficient (α) relations derived from the pore-scale models and applied them to the corescale model. We observed a direct impact on the model's total permeability, the time to breakthrough and the wormhole's total porosity volume, which indicates that Golfier et al (2002)'s uniform dissolution assumption cannot be directly used for predicting the evolution of dissolution under a wide range of flow and transport conditions without investigating the relations between the pore-scale and the core-scale.