Scaling of geochemical reaction rates using pore-scale network modeling: An investigation of anorthite and kaolinite reaction kinetics

Geochemical reaction kinetics are often measured using crushed minerals in well-mixed laboratory systems. These systems can be significantly different from natural porous media where inherent heterogeneities exist. Such differences may lead to the scaling problem, namely the applicability of lab-measured geochemical reaction kinetics in natural porous media. This study develops a methodology to explore these differences, examines the mechanisms that lead to the scaling effects, and identifies the conditions under which the scaling effects are significant.; Pore-scale network models were constructed to represent sandstones with anorthite and kaolinite as reactive minerals, and quartz as the non-reactive mineral. Reactive transport processes were simulated at the pore scale, accounting for the heterogeneities of pore properties. The continuum-scale reaction rates were calculated by either incorporating pore-scale heterogeneities or ignoring them, with their differences being measures of the scaling effects.; The study focuses on high pressure and salinity conditions relevant to geological CO2 sequestration. Simulation results show that the scaling effects are significant under many circumstances. The degree of the scaling effects depends on hydrogeological conditions, including mineral spatial distributions, reactive mineral abundance, hydrodynamic conditions, and boundary conditions. The observed scaling effects are attributed to the concentration variations at the pore scale. Conditions that increase the concentration variations lead to relatively large scaling effects, especially when small amounts of reactive mineral are distributed in large clusters or in layers parallel to the main flow direction, and are subject to highly acidic boundary flow with slow advection rate. Under such conditions, the anorthite dissolution rates can be overestimated by several times; for kaolinite precipitation, its rates can be underestimated by orders of magnitude, and can be predicted as dissolution. Scaling effects also depend on the form of rate laws; with highly nonlinear rate laws, such as that of kaolinite precipitation, the scaling effects are relatively large.; These findings have potentially important implications for representing geochemical kinetics in reactive transport forward modeling, and for interpreting concentration data measured from groundwater samples in inverse modeling. The established methodology can potentially be used to upscale rates of other reactions of environmental significance.