Determination of spatially-resolved porosity, tracer distributions and diffusion coefficients in porous media using MRI measurements and numerical simulations.

This work is focused on measuring the concentration distribution of a conservative tracer in a homogeneous synthetic porous material and in heterogeneous natural sandstone using MRI techniques, and on the use of spatially resolved porosity data to define spatially variable diffusion coefficients in heterogeneous media. The measurements are made by employing SPRITE, a fast MRI method that yields quantitative, spatially-resolved tracer concentrations in porous media. Diffusion experiments involving the migration of H(2)O into D(2)O-saturated porous media are conducted. One-dimensional spatial distributions of H(2)O-tracer concentrations acquired from experiments with the homogeneous synthetic calcium silicate are fitted with the one-dimensional analytical solution of Fick's second law to confirm that the experimental method provides results that are consistent with expectations for Fickian diffusion in porous media. The MRI-measured concentration profiles match well with the solution for Fick's second law and provide a pore-water diffusion coefficient of 1.75×10(-9)m(2)s(-1). The experimental approach was then extended to evaluate diffusion in a heterogeneous natural sandstone in three dimensions. The relatively high hydraulic conductivity of the sandstone, and the contrast in fluid density between the H(2)O tracer and the D(2)O pore fluid, lead to solute transport by a combination of diffusion and density-driven advection. The MRI measurements of spatially distributed tracer concentration, combined with numerical simulations allow for the identification of the respective influences of advection and diffusion. The experimental data are interpreted with the aid of MIN3P-D - a multicomponent reactive transport code that includes the coupled processes of diffusion and density-driven advection. The model defines local diffusion coefficients as a function of spatially resolved porosity measurements. The D(e) values calculated for the heterogeneous sandstone and used to simulate diffusive and advective transport range from 5.4×10(-12) to 1.0×10(-10)m(2)s(-1). These methods have broad applicability to studies of contaminant migration in geological materials.

Transport implications resulting from internal redistribution of arsenic and iron within constructed soil aggregates.

Soils are an aggregate-based structured media that have a multitude of pore domains resulting in varying degrees of advective and diffusive solute and gas transport. Consequently, a spectrum of biogeochemical processes may function at the aggregate scale that collectively, and coupled with solute transport, determine element cycling in soils and sediments. To explore how the physical structure impacts biogeochemical processes influencing the fate and transport of As, we examined temporal changes in speciation and distribution of As and Fe within constructed aggregates through experimental measurement and reactive transport simulations. Spherical aggregates were made with As(V)-bearing ferrihydrite-coated sand inoculated with Shewanella sp. ANA-3; aerated solute flow around the aggregate was then induced. Despite the aerated aggregate exterior, where As(V) and ferrihydrite persist as the dominant species, anoxia develops within the aggregate interior. As a result, As and Fe redox gradients emerge, and the proportion of As(III) and magnetite increases toward the aggregate interior. Arsenic(III) and Fe(II) produced in the interior migrate toward the aggregated exterior and result in coaccumulation of As and Fe(III) proximal to preferential flow paths as a consequence of oxygenic precipitation. The oxidized rind of aggregates thus serves as a barrier to As release into advecting pore-water, but also leads to be a buildup of this hazardous element at preferential flow boundaries that could be released upon shifting geochemical conditions.