Analysis of Solute Mixing at the Pore-Scale Using Micromodels and Lattice -Boltzmann Finite Volume Modeling

102 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006.Transverse mixing has been shown to often be the controlling mechanisms for degradation of contaminants in groundwater. For a conservative contaminant, downgradient concentrations depend on initial source zone concentrations and the extent to which the contaminant mixes (i.e., dilutes) with surrounding groundwater. For a (bio)reactive contaminant, downgradient concentrations also depend on the rate of reaction. In this work I examine the effects of porous media structure (geometry) on the extent of a bi-molecular reaction utilizing a combination of pore scale modeling and high-resolution 2D micromodels. To evaluate reactive transport, two reactive substrates are introduced into a network of pores via two separate and parallel fluid streams. The substrates mix within the porous media via transverse dispersion and react. In micromodel experiments, the reaction forms a fluorescent product which is quantified utilizing fluorescent microscopy. Experimental micromodel results are compared directly with simulations from a pore scale lattice-Boltzmann (LB) finite volume model (FVM) utilizing identical pore structures and flow rates. The LB method is used to solve for interstitial pore velocities. Reactive transport is simulated by combining pore velocities from the LB model with a reactive transport FVM model. Pore scale modeling and experimental results are up-scaled to determine transverse dispersion coefficients at the continuum scale. Results from pore scale simulations and micromodel experiments indicate that (i) LB-FVM is able to capture the mechanisms controlling mixing-limited reactive transport in 2D micromodel experiments, (ii) flow focusing in high conductivity preferential flow zones enhances reactive transport, (iii) grain orientation has a significant effect on extent of mixing and product formation, (iv) grain size (as is commonly used) is not an accurate predictor of mixing, and (v) intra-particle porosity does not have a significant effect on steady-state pore-scale transverse mixing. Finally, excellent agreement between reactive micromodel experiments and forward modeled continuum predictions of reactive transport determined from up-scaled conservative micromodel experimental results was found. This study suggests that sub-continuum effects such as grain orientation, flow focusing, and effective interfacial plume area can play an important role in overall extent of mixing and reaction in groundwater flow, and hence may need to be considered when evaluating reactive transport at the continuum scale.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD