Download files
Abstract
Reactive transport is of great importance in chemical science, hydrogeological and environmental applications. A scalable numerical framework for modelling reactive transport is developed. The migration of solid particles released due to dissolution is included. It is validated by comparing reactive flow simulations with previously published results and dynamic imaging experimental observations. It is used to predict the evolution of pore structure, hydrological and mechanical properties. The impact of pore structure and mineralogical heterogeneity as well as microporosity on reactive transport is also investigated.
The impact of pore structure on reactive flow is investigated by simulating reaction on micro-CT images of rocks with different levels of heterogeneity. The results show the more heterogeneous the pore structure is, the faster the permeability increases during dissolution. However, the correlations between pore space heterogeneity and reaction regimes have not yet been quantitatively illustrated. Thus, we quantify the pore geometry with correlation length and conduct reaction simulations in pore-scale correlated porous media. Unstructured irregular pore spaces with different correlation lengths are generated. A conceptual diagram is suggested to relate the reactive transport with the correlation length. Variations of mechanical properties in porous media during reactive flow are studied. Stress load cases on porous media are simulated and maps of deformation in rocks are compared. Dependency of Young’s modulus and Poisson’s ratio on Péclet and Damköhler numbers are demonstrated. The effect of mineralogical heterogeneity is also explored. The mineral map of sandstones was acquired using QEMSCAN SEM-EDS. Pore-scale simulations of multi-mineral reaction were performed directly on rock images. Numerical results show that mineralogical heterogeneity can cause significant errors in permeability predictions. Reactive transport in dual porosity porous media is explored in a fractured medium and carbonates. The Stokes-Brinkman equation is solved for fluid flow in rocks with microporosity. Large discrepancy in dissolution patterns is observed if microporosity is ignored in the reaction simulation. Based on the fluid-solid reaction model by adding the module of fluid-fluid reaction, a fluid-fluid-solid reactive transport model has been developed.