Geological and engineered media are characterized by heterogeneity at multiple scales such that porosity is often distributed over a range of spatial scales. Pore-scale models make it possible to capture processes at the scale of individual pores but are computationally expensive for the simulation of large domains. Darcy-scale models are routinely used for efficient simulation of domains of any size.
Sergi Molins is a Research Scientist at the Lawrence Berkeley National Laboratory with extensive experience in the field of reactive transport modeling. His research focuses on elucidating the processes that affect effective reaction rates at a range of spatial scales in applications relevant to water, energy and the environment. Code development in collaborative projects is an integral part of his work having contributed to CrunchFlow, Chombo-Crunch, MIN3P, Amanzi/ATS and Alquimia. He holds a PhD from the University of British Columbia, Vancouver, and a Civil Engineering degree from the Technical University of Catalonia. Barcelona. He currently serves as Associate Editor for Water Resources Research
Reactions driven by the introduction of a fluid out of equilibrium with the native material result in the physical evolution of the media. This evolution may in turn alter the balance between the transport processes leading to emergent behavior. Pore-scale models have made significant strides in simulating solid-fluid interface as demonstrated by a recent code benchmarking effort. However, they still are challenged by the need to resolve fine-scale features and at the same time capture coarser-scale processes.
Multiscale models that combine pore- and Darcy-scale descriptions provide a reasonable compromise between the fine process-resolution of pore-scale models and the computational advantages of Darcy-scale models. In this presentation, we build on previously developed pore-scale model for the simulation of reactive transport to develop a hybrid multiscale model. The model is composed of pore-scale and Darcy-scale sub-domains that make up the entire medium. The two sub-problems are solved sequentially by exchanging concentrations and fluxes at their interface. The evolution of the Darcy-scale domain is captured with porosity changes, while that of the pore-scale domain with the evolution of the interface