Dissolution followed by precipitation is a major mechanism in the formation of secondary precipitates in most porous sediments. Secondary precipitation is of interest as a structure modifying mechanism that may also control contaminant transport in the subsurface environment. Quantification of structural change is a necessary component for the construction of predictive models for effective reaction rates at field scales.

We have employed synchrotron X-ray computed microtomography in combination with flow-column experiments to capture and quantify snapshots in time of dissolution and secondary precipitation changes in the microstructure of Hanford sediments exposed to simulated caustic waste in flow-column experiments. Careful image analysis was done to address the anticipated systematic errors. Changes accompanying a net reduction in porosity of 4% were quantified including:(1) a 25% net decrease in pores resulting from a 38% loss in the number of pores < 10⁻⁴ mm³ in volume and a 13% increase in the number of pores of larger size; and (2) a 38% decrease in the number of throats. The loss of throats resulted in decreased coordination number from pores of all sizes and significant reduction in the number of pore pathways.

A reactive flow network model was developed to simulate the evolution of the chemical species resulting from the reactions in a flow-column experiment and under batch experiment conditions. This single network flow modal incorporated both kinetic (for dissolution and precipitation of solids) and instantaneous (for equilibrium of aqueous species) reactions, as well as advection and diffusion of concentrations in the pore space. The single phase flow model incorporated channel conductances based upon more exact, pre-computed, Lattice-Boltzmann computations. The reactive network flow simulation indicates that, after initial quartz dissolution, secondary precipitation dominates in the pore space within six hours of initialization of flow resulting in eventual equilibrium of silicon ion concentration, [Si], (from quartz dissolution) and aluminum ion concentration, [Al], (carried in the invading solution). After halting further fluid input (to simulate batch reactor conditions), dissolution begins to dominate again, resulting in increased [Si] and decreased [Al].