David A. Reynolds

Discretizing the fracture-matrix interface to simulate solute transport

This article examines the required spatial discretization perpendicular to the fracture-matrix interface (FMI) for numerical simulation of solute transport in discretely fractured porous media. The discrete-fracture, finite-element model HydroGeoSphere (Therrien et al. 2005) and a discrete-fracture implementation of MT3DMS (Zheng 1990) were used to model solute transport in a single fracture, and the results were compared to the analytical solution of Tang et al. (1981). To match analytical results on the relatively short timescales simulated in this study, very fine grid spacing perpendicular to the FMI of the scale of the fracture aperture is necessary if advection and/or dispersion in the fracture is high compared to diffusion in the matrix. The requirement of such extremely fine spatial discretization has not been previously reported in the literature. In cases of high matrix diffusion, matching the analytical results is achieved with larger grid spacing at the FMI. Cases where matrix diffusion is lower can employ a larger grid multiplier moving away from the FMI. The very fine spatial discretization identified in this study for cases of low matrix diffusion may limit the applicability of numerical discrete-fracture models in such cases.

Electrokinetic Migration of Permanganate Through Low-Permeability Media

This research was conducted to evaluate the combination of electromigration and potassium permanganate as a potential remediation method for low-permeability media (e.g., soil and sediment) contaminated with dissolved and sorbed organic contaminants. The experimental procedure was composed of two stages: determination of migration rates of permanganate through homogeneous cores and a primarily qualitative analysis of migration in more heterogeneous, two-dimensional scenarios. Results indicated that transport of permanganate through fine-grained porous media and clays can be undertaken using electromigration, and electromigration rates were found to be at least 400% faster than diffusion alone. In addition, the use of an applied electric field in a flushing scenario was shown to result in almost 100% sweep efficiency of a domain consisting of clay blocks interspersed in a glass bead medium. The results of the study show that there is potential for this method to be able to deliver permanganate and other potential remedial agents to treat contaminated zones within heterogeneous and low-permeability porous media through in situ chemical oxidation or other processes.

Multiphase flow and transport in fractured clay/sand sequences

A numerical model (Queens University Multi-Phase Flow Simulator, QUMPFS) was used to assess the rate of trichloroethylene (TCE) dense, non-aqueous phase liquid (DNAPL) migration through fractured clay, with special attention focused on the influence of interbedded sand lenses. The presence of these sand lenses was found to increase the time required for the non-wetting phase to migrate through the full 30 m vertical extent of the clay sequence from a few days to several years. Applied vertical hydraulic gradients were found to be moderately influential in systems consisting solely of fractured clays, yet one of the dominant factors controlling speed of vertical migration when sand lenses were present. Larger displacement pressure of the sands relative to that of the fractures leads to slower DNAPL migration rates, due to the delays that occur during build-up of capillary pressures. Dissolution of DNAPL and subsequent matrix diffusion of the aqueous phase has little effect on the rate of DNAPL migration through systems consisting of fractured clay only, yet slows the rate of migration in systems containing sand lenses. In all cases examined, the rate of DNAPL loading to the lower aquifer far exceeded the rate of aqueous phase mass loading. It was also found that DNAPL reaches the lower aquifer at approximately the same time as the aqueous phase plumes even for systems experiencing downward groundwater flow due to the attenuation of the aqueous phase through matrix diffusion.

Electrokinetic in situ oxidation remediation: Assessment of parameter sensitivities and the influence of aquifer heterogeneity on remediation efficiency

A newly developed groundwater and electrokinetic (EK) flow and reactive transport numerical model was applied to simulate electrokinetic in situ chemical oxidation (EK-ISCO) remediation. Scenario simulations that considered the oxidation of a typical organic contaminant (tetrachloroethene) by permanganate were used to gain a better understanding of the key processes and parameters that control remediation efficiency. In a first step a sensitivity analysis was carried out to investigate a range of EK, hydraulic and engineering parameters on the performance of EK-ISCO. While all investigated parameters affected the remediation process to some extent, the duration and energy required for remediation were shown to be most dependent upon the applied voltage gradient, the natural oxidant demand and the concentration of the injected oxidant. Secondly, the efficacy of EK-induced oxidant transport was further examined for a heterogeneous aquifer system with random permeability fields. Oxidant migration under EK was slower in low-permeability media due to the increased oxidant consumption of competing reductants. Instead of injecting oxidant only at the cathode, locating injection wells between the electrodes greatly increased the contaminant degradation by decreasing the distance the amendment had to migrate before reaching the contaminant.

Overcoming Permanganate Stalling during Electromigration

AbstractElectrokinetic experiments were undertaken to transport permanganate (MnO4−) through a low permeability porous media. The experiments employed a one-dimensional apparatus in which MnO4− was electromigrated through a central porous media core. Two outer porous media cores separated the electrode reservoirs from the inner permanganate source and permanganate target reservoirs. By utilizing a pH-isolation technique, whereby electrolysis reactions occurring at electrodes were isolated, uniform and repeatable MnO4− electromigration was achieved. This result was compared with non-pH-isolated experiments (normal mode), which resulted in a stalled MnO4− electromigration front. The research also investigated potential stalling mechanisms, including voltage gradient nonlinearity through the central porous media core and the reduction of MnO4− to Mn2+. It was observed that the voltage gradient decreased as a result of MnO4− stalling; however, it was not considered a stalling mechanism. Results from Mn2+ anal...

Development of an Apparatus for pH-Isolated Electrokinetic In Situ Chemical Oxidation

An apparatus was designed, manufactured, and implemented to isolate pH during electrokinetic in situ chemical oxidation (EK-ISCO). H+ and OH- electromigration were used to determine the adequacy of the designed apparatus for pH isolation. A series of pH-isolation and normal-mode (no pH-isolation) experiments were undertaken and compared. It was found that pH isolation was achieved when the electrode reservoirs were separated by porous media combined with the purging of the electrode reservoir fluid. The electromigration retardation factor of H+ and OH- was calculated for the porous media using the observed pH breakthrough times. The retardation factor for H+ was also calculated by considering mass flux data. The retardation factors for H+ and OH- were found to be 28.3 and 95, respectively, when using the breakthrough time. The retardation factor for H+ was calculated to be 36.7 using the mass flux data.