HR: 16:30h
AN: H44D-03    [Abstracts]
TI: Lattice Boltzmann Simulation of Fluid Flow and Solute Transport in Porous Media at the Pore Scale and Upscaling to the Continuum Scale
AU: * Kang, Q
EM: qkang@lanl.gov
AF: Hydrology and Geochemistry Group, Los Alamos National Laboratory, Los Alamos, NM 87545,
AU: Lichtner, P C
EM: lichtner@lanl.gov
AF: Hydrology and Geochemistry Group, Los Alamos National Laboratory, Los Alamos, NM 87545,
AU: Wang, M
EM: mmwang@ucdavis.edu
AF: Nanomaterials in the Environment, Agriculture and Technology, University of California, Davis, CA 95616,
AB: In continuum modeling approaches of multiphase flow and reaction in porous media, spatial heterogeneity at the pore scale is unresolved, which may be important on the observed behavior at the larger scale. Therefore, understanding multiphase flow, transport, and reaction processes at the pore scale and subsequently upscaling to the larger scale will provide valuable insight into the effects of pore scale heterogeneity on the emergent behavior at the field scale. In this study, we investigate such effects by simulating leaching of a non-reactive tracer from a three dimensional (3D) porous structure and then upscaling to the continuum scale. The 3D structure is generated using a random structure generation-growth method, termed the quartet structure generation set (QSGS), which can reproduce porous morphological features that closely resemble the formation process of many real porous media. The generated 3D structure includes multiscale features with dead-end pores connected to the primary pores through diffusion pathways. An incompressible lattice Boltzmann (LB) model is used to simulate fluid flow, and a 3D, 6- speed (D3Q6) LB model is used to simulate solute transport at the pore scale. The leaching process is also simulated using single and dual-continuum models (SCM and DCM), based on the macroscopic parameters derived from the upscaled LB results and directly from the pore structure. Breakthrough curves obtained from the three methods are compared with each other. It is shown that the long tail of the breakthrough curve caused by the dead-end pores is reasonably captured by the LB and DCM, but not by the SCM, which predicts a breakthrough time much too early.
DE: 1832 Groundwater transport
SC: Hydrology [H]
MN: 2007 Fall Meeting