Reaction dynamics, molecular clusters, and aqueous geochemistry

Earth scientists will come to rely heavily upon computational chemistry because our environments are so difficult to sample. Essential geochemical information can often be better acquired by simulation than by either experiment or field sampling. No reactions are more essential to geochem istry than those involving water. Considerable advances are being made in understanding these hydrolytic reactions by close coupling of simulation to experiments on 1-5-nm-sized metal-hydroxide clusters. These clusters are sufficiently small that experiments can identify reaction properties at specific metal-oxygen sites that can then be treated using high-level methods of simulation. This twofold approach shows that some reaction classes at mineral surfaces are surprisingly robust and treatable using kinetic information in hand. Pathways for other reactions, such as dissociation of the oxygens that bridge metals and that hold the structure together, are enormously sensitive to details that neither experiment, nor simulation, alone can fully uncover. New efforts are needed to synthesize 1-5 nm experimental molecules to predict rate parameters for these aqueous reactions.