HR: 10:50h
AN: H51H-03    [PDF]
TI: The role of Peclet number on the alteration of variable aperture fractures by dissolution: A comparison of physical experiments with computational simulations
AU: * Detwiler, R L
EM: detwiler1@llnl.gov
AF: University of California, Lawrence Livermore National Laboratory PO Box 808, L-201, Livermore, CA 94551 United States
AU: Rajaram, H
EM: hari@colorado.edu
AF: University of Colorado, UCB-428, Boulder, CO 80309-0428 United States
AU: Cheung, W W
EM: cheung.wendy@epamail.epa.gov
AF: United States Environmental Protection Agency, 999 18th St., Denver, CO 80202-2466 United States
AB: In a fracture in which the fluid and rock surfaces are not in chemical equilibrium, reactions between the flowing fluid and host rock can lead to dissolution and/or precipitation of minerals along the fracture surfaces. The resulting changes in fracture aperture can lead to alteration of fracture permeability. Local rates of precipitation/dissolution are controlled by surface reaction rates and the transport of reactants by advection and diffusion within the fracture. For the case of fast surface reactions, the local Peclet number (Pe = V$\langle$b$\rangle$/D$_m$, where V is the mean fluid velocity, $\langle$b$\rangle$ is the mean fracture aperture, and D$_m$ is the molecular diffusion coefficient of dissolved minerals ), which is the ratio of the time scales of advection and diffusion, will control the fracture-scale patterns of mineral dissolution and precipitation. We compare results from a pair of physical dissolution experiments at different initial values of Pe (Pe$_o$) to dissolution simulations using a coupled reactive transport model based upon the two-dimensional Reynolds equation for flow through variable aperture fractures. We fabricated transparent analog fractures by mating a 9.9 x 15.2 cm textured glass plate with flat potassium-dihydrogen-phospate (KDP) crystals. Because only the KDP surface dissolved during the experiments, it was possible to replicate the initial conditions for each experiment. In addition, the transparent fracture allowed us to use light transmission techniques to directly measure the fracture aperture during experiments at high-resolution (0.083 x 0.083 mm pixels). For Pe$_o$=54, distinct dissolution channels formed, while for Pe$_o$=216, we measured minimal channeling and a reduction in short-wavelength aperture variability. Computational simulations using the initial measured aperture fields yield qualitatively similar dissolution patterns. In both the experiments and simulations, the nature of the dissolution patterns strongly influenced the relative increase of fracture permeability. The experiments demonstrated that a 110% increase in mean fracture aperture resulted in estimated permeability increases of 440% and 640% for the Pe$_o$=54 and Pe$_o$=216 experiments, respectively. The simulations are in close agreement with the experiments up to about 50% growth in mean aperture after which the simulations predict faster increases in permeability than we observed experimentally.
DE: 1829 Groundwater hydrology
DE: 1832 Groundwater transport
DE: 5104 Fracture and flow
DE: 5114 Permeability and porosity
SC: Hydrology [H]
MN: 2003 Fall Meeting