Molecular scale modeling of mineral-fluid reactions
I have a long term interest in the kinetics and thermodynamics of mineral interactions with natural fluids. Although I have extensive expertise in experimental and analytical work in the laboratory, I am currently focused on (1) the development of kinetic models for the study of mineral growth, dissolution, nucleation, recrystallization and related processes, and (2) the use of these tools to drive the development of fundamental theories governing mineral properties and behavior. I see two current problems in this area of geochemistry: first, despite an extensive and increasing catalog of published observations from both the laboratory and the field, we have but a limited description of how processes such as growth and dissolution operate at a mechanistic, molecular level. This limitation can be addressed through vigorous development of computer-based simulations (molecular dynamics, kinetic Monte Carlo, etc.). This approach harnesses inexpensive computing power to produce simulations of basic processes over a range of time and space scales. Rather than first produce observations in the laboratory and then seek to understand these observations in the context of existing models, it is more beneficial to first produce simulations using available energetic parameterizations. The results of such simulations can then be compared with existing experimental results; a disparity between observed experimental and simulation results can then drive selection of new parameters in a recursive, feedback approach. This approach can incoporate biological mediation reactions and processes as well. Second, these molecular scale results must be linked to the complex, phenomenological, process-level observations in the field. This extrapolation across time and space scales requires a new approach that incorporates the fundamentally stochastic nature of heterogeneous reaction networks.
I am also interested in the integration of the understanding provided by the above activities to reaction-based predictive models of biogeochemical cycling. These efforts can range from short-term, ecosystem-level descriptions to long-term models that permit an understanding of past, present, and future systems. I am the primary author of MAGic , a dynamical model integrating long-term, global biogeochemical cycles over the Phanerozoic, incorporating basic reactions of silicate and carbonate weathering, the exchange of seawater with MORB and clays, precipitation and dissolution of chemical sediments, organic matter production, burial, and remineralization. A central motivation for this effort is to understand the nature of the overall linkage between major reservoirs and the role of long-term cyclic versus secular change. The latest revision of this model  explores the uncertainty in seawater-derived sedimentary dolomite fluxes, produces a baseline flux of dolomite that can then be compared with observed mass in the rock record, and concludes that variations in dolomite production may have played a significant role in terms of buffering of seawater saturation state over geologic time.