Atomistic to meso-scale modeling of mineral dissolution: Methods, challenges and prospects

Dissolution and growth of minerals constitute a special interdisciplinary field covering a large variety of modern ecological and geochemical problems, for example, radioactive and toxic waste sequestration, weathering and biomineralization. The processes of mineral dissolution and growth are inherently complex. Formulation of a consistent theory applicable to arbitrary environmental conditions is a big goal for the geochemical community. This task cannot be fulfilled without a rigorous systematic methodological approach to create multi-scale models. We review current modeling approaches from atomistic to meso-scale as well as state-of-the-art approaches to connect these scales. Atomistic models provide us with molecular reaction rates and help us to explain mechanisms of bond dissociation, formation of transition state and adsorption complexes. Kinetic Monte Carlo (KMC) models incorporate an ensemble of different elementary reactions taking place at mineral surfaces. These stochastic models provide us with reaction sequences and corresponding surface topographies, as well as the geometry of reactive surface features. Experimental measurements provide important material for model verification. Environmental controls of the reaction process are incorporated into KMC models by using complementary Grand Canonical Monte Carlo (GCMC) simulations that provide statistics on charged sites and adsorbed ions. Reactive Canonical Monte Carlo constitutes an alternative approach to study mineral-fluid systems at chemical equilibrium. Analytical models of dissolution and growth based on atomic step velocities as functions on elementary rates and environmental conditions have been actively developed for decades. A novel, fast approach based on computational geometry, using Voronoi surface partitioning with non-Euclidean distance function, recently appeared. This approach allows us to simulate systems larger than what can be handled with atomistic methods and enables the possibility to upscale all-atomic meso-scale models to pore and continuum scales. The aim of this manuscript is to provide a road map of the existing vibrant field of atomistic-to-meso-scale models and computer simulations. We supplement the text with important equations quantitatively relating physical parameters involved in models at different scales. The discussion of existing methodological gaps and lack of parameters necessary for the construction of multi-scale models is provided.