The velocity of DNAPL fingering in water-saturated porous media: laboratory experiments and a mobile–immobile–zone model

Dense nonaqueous phase liquids (DNAPLs) are immiscible with water and can give rise to highly fingered fluid distributions when infiltrating through water-saturated porous media. In this paper, a conceptual mobile–immobile–zone (MIZ) model is presented to describe the structure of a DNAPL finger in water-saturated porous media and the velocity of finger propagation. A finger is composed of a finger body and a tip. The finger body has a mobile core and an immobile sheath. All the DNAPL within the tip of a finger is mobile. Lab experiments utilizing image analyses of a DNAPL (PCE) penetrating into water-saturated homogeneous glass beads were carried out in a two-dimensional transparent chamber. The results show that the fingers elongated almost linearly with time. The fingers did not grow laterally after the tip of the finger had passed. The average finger diameters were between 3.9 and 5.4 mm for PCE propagation in water-saturated glass bead porous media with mean particle diameters from 0.32 to 1.36 mm. The estimated mobile core diameters were 51–60% of the average finger diameters.     

The significance of heterogeneity on mass flux from DNAPL source zones: an experimental investigation

Understanding the process of mass transfer from source zones of aquifers contaminated with organic chemicals in the form of dense non-aqueous phase liquids (DNAPL) is of importance in site management and remediation. A series of intermediate-scale tank experiments was conducted to examine the influence of aquifer heterogeneity on DNAPL mass transfer contributing to dissolved mass emission from source zone into groundwater under natural flow before and after remediation. A Tetrachloroethylene (PCE) spill was performed into six source zone models of increasing heterogeneity, and both the spatial distribution of the dissolution behavior and the net effluent mass flux were examined. Experimentally created initial PCE entrapment architecture resulting from the PCE migration was largely influenced by the coarser sand lenses and the PCE occupied between 30 and 60% of the model aquifer depth. The presence of DNAPL had no apparent effect on the bulk hydraulic conductivity of the porous media. Up to 71% of PCE mass in each of the tested source zone was removed during a series of surfactant flushes, with associated induced PCE mobilization responsible for increasing vertical DNAPL distributions. Effluent mass flux due to water dissolution was also found to increase progressively due to the increase in NAPL-water contact area even though the PCE mass was reduced. Doubling of local groundwater flow velocities showed negligible rate-limited effects at the scale of these experiments. Thus, mass transfer behavior was directly controlled by the morphology of DNAPL within each source zone. Effluent mass flux values were normalized by the up-gradient DNAPL distributions. For the suite of aquifer heterogeneities and all remedial stages, normalized flux values fell within a narrow band with mean of 0.39 and showed insensitivity to average source zone saturations.     

Manipulation of density and viscosity for the optimization of DNAPL recovery by alcohol flooding

Laboratory experiments demonstrate that in situ recovery of pooled tetrachloroethene (PCE) from porous media may be accomplished more efficiently using multiple-step alcohol floods than with single alcohol floods. To optimize flooding efficiency while maintaining a low risk of downward DNAPL mobilization, a three-step flooding process is developed employing an isobutanol preflood, a composite alcohol mainflood, and a polymer solution postflood. The density and viscosity of these solutions are manipulated to prevent the onset and propagation of viscous and gravitational fingers, while maintaining phase behavior critical for efficient miscible NAPL displacement. An aqueous partitioning preflood solution of 10% by volume (10% v) isobutanol reduces the NAPL density in situ to approximately 1.00 g/ml by swelling the NAPL prior to miscible displacement induced by the mainflood. The composite alcohol mainflood, containing 65% v ethylene glycol and 35% v 1-propanol maintains miscibility while achieving neutral buoyancy and near stable displacement of the NAPL. Aqueous solutions of xanthan gum polymer efficiently displace the mainflood, reducing viscous fingering associated with waterfloods. Two-dimensional experiments using the multiple-step technique achieve 99.8% DNAPL mass recovery using a total of 0.45 pore volumes of alcohol, illustrating greater recovery efficiency than previous alcohol flooding formulations under comparable conditions.     

Natural remobilization of multicomponent DNAPL pools due to dissolution

Mixtures of dense nonaqueous phase liquids (DNAPLs) trapped in the subsurface can act as long-term sources of contamination by dissolving into flowing groundwater. If the components have different solubilities then dissolution will alter the composition of the remaining DNAPL. We theorized that a multicomponent DNAPL pool may become mobile due to the natural dissolution process. In this study, we focused on two scenarios: (1) a DNAPL losing light component(s), with the potential for downward migration; and (2) a DNAPL losing dense component(s), with the potential for upward migration following transformation into a less dense than water nonaqueous phase liquid (LNAPL). We considered three binary mixtures of common groundwater contaminants: benzene and tetrachloroethylene (PCE), PCE and dichloromethane (DCM), and DCM and toluene. A number of physical properties that control the retention and transport of DNAPL in porous media were measured for the mixtures, namely: density, interfacial tension, effective solubility, and viscosity. All properties except density exhibited nonlinear relationships with changing molar ratio of the DNAPL. To illustrate the potential for natural remobilization, we modelled the following two primary mechanisms: the reduction in pool height as mass is lost by dissolution, and the changes in fluid properties with changing molar ratio of the DNAPL. The first mechanism always reduces the capillary pressure in the pool, while the second mechanism may increase the capillary pressure or alter the direction of the driving force. The difference between the rate of change of each determines whether the potential for remobilization increases or decreases. Static conditions and horizontal layering were assumed along with a one-dimensional, compositional modelling approach. Our results indicated that for initial benzene/PCE ratios greater than 25:75, the change in density was sufficiently faster than the decline in pool height to promote DNAPL breakthrough into the adjacent porous medium. In contrast, there was no potential for natural remobilization of a PCE-DCM mixture, primarily because the densities of the components are not sufficiently different. Dissolution of a DCM-toluene mixture decreased the density, reducing the tendency for downward displacement. However, the ultimate transformation from a DNAPL to an LNAPL may induce upward displacement. These results suggest that at sites with DNAPL pools containing a mix of components of sufficiently different densities and relative solubilities, natural remobilization may be an active mechanism, with implications for site evaluation and remediation.     

Modeling field-scale cosolvent flooding for DNAPL source zone remediation

A three-dimensional, compositional, multiphase flow simulator was used to model a field-scale test of DNAPL removal by cosolvent flooding. The DNAPL at this site was tetrachloroethylene (PCE), and the flooding solution was an ethanol/water mixture, with up to 95% ethanol. The numerical model, UTCHEM accounts for the equilibrium phase behavior and multiphase flow of a ternary ethanol-PCE-water system. Simulations of enhanced cosolvent flooding using a kinetic interphase mass transfer approach show that when a very high concentration of alcohol is injected, the DNAPL/water/alcohol mixture forms a single phase and local mass transfer limitations become irrelevant. The field simulations were carried out in three steps. At the first level, a simple uncalibrated layered model is developed. This model is capable of roughly reproducing the production well concentrations of alcohol, but not of PCE. A more refined (but uncalibrated) permeability model is able to accurately simulate the breakthrough concentrations of injected alcohol from the production wells, but is unable to accurately predict the PCE removal. The final model uses a calibration of the initial PCE distribution to get good matches with the PCE effluent curves from the extraction wells. It is evident that the effectiveness of DNAPL source zone remediation is mainly affected by characteristics of the spatial heterogeneity of porous media and the variable (and unknown) DNAPL distribution. The inherent uncertainty in the DNAPL distribution at real field sites means that some form of calibration of the initial contaminant distribution will almost always be required to match contaminant effluent breakthrough curves.     

Locally-calibrated light transmission visualization methods to quantify nonaqueous phase liquid mass in porous media

Five locally-calibrated light transmission visualization (LTV) methods were tested to quantify nonaqueous phase liquid (NAPL) mass and mass reduction in porous media. Tetrachloroethylene (PCE) was released into a two-dimensional laboratory flow chamber packed with water-saturated sand which was then flushed with a surfactant solution (2% Tween 80) until all of the PCE had been dissolved. In all the LTV methods employed here, the water phase was dyed, rather than the more common approach of dyeing the NAPL phase, such that the light adsorption characteristics of NAPL did not change as dissolution progressed. Also, none of the methods used here required the use of external calibration chambers. The five visualization approaches evaluated included three methods developed from previously published models, a binary method, and a novel multiple wavelength method that has the advantage of not requiring any assumptions about the intra-pore interface structure between the various phases (sand/water/NAPL). The new multiple wavelength method is also expected to be applicable to any translucent porous media containing two immiscible fluids (e.g., water-air, water-NAPL). Results from the sand-water-PCE system evaluated here showed that the model that assumes wetting media of uniform pore size (Model C of Niemet and Selker, 2001) and the multiple wavelength model with no interface structure assumptions were able to accurately quantify PCE mass reduction during surfactant flushing. The average mass recoveries from these two imaging methods were greater than 95% for domain-average NAPL saturations of approximately 2.6x10(-2), and were approximately 90% during seven cycles of surfactant flushing that sequentially reduced the average NAPL saturation to 7.5x10(-4).     

Surfactant-induced changes in gravity fingering of water through a light oil

Gravity-driven preferential flow (fingering) can greatly affect how one fluid displaces another in the subsurface. We have studied the internal properties of these preferential flow paths for water, with and without surfactants, infiltrating into oil saturated porous media using synchrotron X-rays, and miniature tensiometers to characterize fluid content and pressure relationships. We also used a light transmission technique to visualize overall flow pattern. Capillary pressure and water content decrease behind the front, similar to fingers in air-dry sand, with quantitative differences for five different surfactants with surface tensions ranging from 4–21 g/s². Using unstable flow theory, the finger widths, capillary pressure drops within the fingers, finger tip lengths, and finger splitting dynamics were scaled successfully with interfacial tension, fluid density, and the contact angle using the fingers in air–water systems as the reference.     

Intermediate-scale 2D experimental investigation of in situ chemical oxidation using potassium permanganate for remediation of complex DNAPL source zones

In situ chemical oxidation is a technology that has been applied to speed up remediation of a contaminant source zone by inducing increased mass transfer from DNAPL sources into the aqueous phase for subsequent destruction. The DNAPL source zone can consist of one or more individual sources that may be present as an interconnected pool of high saturation, as a region of disconnected ganglia at residual saturation, or as combinations of these two morphologies. Potassium permanganate (KMnO(4)) is a commonly employed oxidant that has been shown to rapidly destroy DNAPL compounds like PCE and TCE following second-order kinetics in an aqueous system. During the oxidation of a target DNAPL compound, or naturally occurring reduced species in the subsurface, manganese oxide (MnO(2)) solids are produced. Research has shown that these manganese oxide solids may result in permeability reductions in the porous media thus reducing the ability for oxidant to be transported to individual DNAPL sources. It can also occur at the DNAPL-water interface, decreasing contact of the oxidant with the DNAPL. Additionally, MnO(2) formation at the DNAPL-water interface, and/or flow-bypassing as a result of permeability reductions around the source, may alter the mass transfer from the DNAPL into the aqueous phase, potentially diminishing the magnitude of any DNAPL mass depletion rate increase induced by oxidation. An experiment was performed in a two-dimensional (2D) sand-filled tank that included several discrete DNAPL source zones. Spatial and temporal monitoring of aqueous PCE, chloride, and permanganate concentrations was used to relate changes in mass depletion of, and mass flux, from DNAPL residual and pool source zones to chemical oxidation performance and MnO(2) formation. During the experiment, permeability changes were monitored throughout the 2D tank and these were related to MnO(2) deposition as measured through post-oxidation soil coring. Under the conditions of this experiment, MnO(2) formation was found to reduce permeability in and around DNAPL source zones resulting in changes to the overall flow pattern, with the effects depending on source zone configuration. A pool with little or no residual around it, in a relatively homogeneous flow field, appeared to benefit from resulting MnO(2) pore-blocking that substantially reduced mass transfer from the pool even though there was relatively little PCE mass removed from the pool. In contrast, a pool with residual around it (in a more typical heterogeneous flow field) appeared to undergo increased mass transfer as MnO(2) reduced permeability, altering the water flow and increasing the mixing at the DNAPL-water interface. Further, the magnitude of increased PCE mass depletion during oxidation appeared to depend on the PCE source configuration (pool versus ganglia) and decreased as MnO(2) was formed and deposited at the DNAPL-water interface. Overall, the oxidation of PCE mass appeared to be rate-limited by the mass transfer from the DNAPL to aqueous phase.     

Effective parameters for two-phase flow in a porous medium with periodic heterogeneities

Computational simulations of two-phase flow in porous media are used to investigate the feasibility of replacing a porous medium containing heterogeneities with an equivalent homogeneous medium. Simulations are performed for the case of infiltration of a dense nonaqueous phase liquid (DNAPL) in a water-saturated, heterogeneous porous medium. For two specific porous media, with periodic and rather simple heterogeneity patterns, the existence of a representative elementary volume (REV) is studied. Upscaled intrinsic permeabilities and upscaled nonlinear constitutive relationships for two-phase flow systems are numerically calculated and the effects of heterogeneities are evaluated. Upscaled capillary pressure-saturation curves for drainage are found to be distinctly different from the lower-scale curves for individual regions of heterogeneity. Irreducible water saturation for the homogenized medium is found to be much larger than the corresponding lower-scale values. Numerical simulations for both heterogeneous and homogeneous representations of the considered porous media are carried out. Although the homogenized model simulates the spreading behavior of DNAPL reasonably well, it still fails to match completely the results form the heterogeneous simulations. This seems to be due, in part, to the nonlinearities inherent to multiphase flow systems. Although we have focussed on a periodic heterogeneous medium in this study, our methodology is applicable to other forms of heterogeneous media. In particular, the procedure for identification of a REV, and associated upscaled constitutive relations, can be used for randomly heterogeneous or layered media as well.     

Determining effective interfacial tension and predicting finger spacing for DNAPL penetration into water-saturated porous media

The difficulty in determining the effective interfacial tension limits the prediction of the wavelength of fingering of immiscible fluids in porous media. A method to estimate the effective interfacial tension using fractal concepts was presented by Chang et al. [Water Resour. Res. 30 (1994) 125]. We modified the method in that the macroscopic interface length was used instead of the system width. Methods to determine the macroscopic and the microscopic interface length are given. Lab experiments of dense nonaqueous phase liquid (DNAPL) penetrating into water-saturated glass beads were carried out in a two-dimensional (2-D) transparent chamber. The displacement processes were recorded using a 35-mm camera or a video camera, which was directly connected to and controlled by a computer. Unlike the method of Chang et al. (1994), the modified method used here gives a constant value of the effective interfacial tension over time. The predicted wavelengths of fingering are relatively close to those observed except for the fine beads.     

Infiltration and redistribution of perchloroethylene in partially saturated, stratified porous media

Contamination of the subsurface by nonaqueous phase liquids (NAPLs) is a widespread problem. To investigate the behavior of a nonspreading, dense NAPL (DNAPL) in the vadose zone, we conducted perchloroethylene (PCE) infiltration experiments in nominally 1- and 2-dimensional (D), stratified porous media. In addition, the usefulness and limitations of a multifluid flow simulator to describe PCE infiltration and redistribution under the experimental conditions were tested. The physical simulations were conducted in a column (1-D) and a flow container (2-D) which were packed with two distinct layers of coarse-grained sand and a fine-grained sand layer in between. Volumetric water and PCE contents were determined with a fully automated dual-energy gamma radiation system. While migrating through the drier parts of the coarse-grained sand layers, PCE appeared to wet the water–air interface rather than displacing any water. In the wetter parts of the porous medium, PCE displaced water and behaved as a true nonwetting fluid. PCE showed a limited response to gradients in capillary pressure and rather high values for the volumetric PCE content were measured in the fine-grained sand layers. This was attributed to the nonspreading nature of PCE. The multifluid flow simulator appeared to predict the initial PCE movement in the vadose zone reasonably well. However, the model was not capable of predicting the final amounts of PCE retained in either the unsaturated or saturated part of the flow domain, mainly because the simulator does not consider the nonspreading flow behavior of NAPLs.     

Effects of ethanol addition on micellar solubilization and plume migration during surfactant enhanced recovery of tetrachloroethene

Alcohol addition has been suggested for use in combination with surfactant flushing to enhance solubilization kinetics and permit density control of dense non-aqueous phase liquid (DNAPL)-laden surfactant plumes. This study examined the effects of adding ethanol (EtOH) to a 4% Tween 80 (polyoxyethylene (20) sorbitan monooleate) solution used to flush tetrachloroethene (PCE)-contaminated porous media. The influence of EtOH concentration, subsurface layering and scale on flushing solution delivery and PCE recovery was investigated through a combination of experimental and mathematical modeling studies. Results of batch experiments demonstrated that the addition of 2.5%, 5% and 10% (wt.) EtOH incrementally increased the PCE solubilization capacity and viscosity of the surfactant solution, while reducing solution density from 1.002 to 0.986 g/cm3. Effluent concentration data obtained from one-dimensional (1-D) column experiments were used to characterize rate-limited micellar solubilization of residual PCE, which was strongly dependent upon flow velocity and weakly dependent upon EtOH concentration. Two-dimensional (2-D) box studies illustrated that minor differences (0.008 g/cm3) between flushing and resident solution density can strongly influence surfactant front propagation. A two-dimensional multiphase simulator, MISER, was used to model the influence of EtOH composition on the aqueous flow field and PCE mass recovery. The ability of the numerical simulator to predict effluent concentrations and front propagation was demonstrated for both 1-D columns and 2-D boxes flushed with EtOH-amended Tween 80 solutions. Results of this study quantify the potential influence of alcohol addition on surfactant solution properties and solubilization capacity, and demonstrate the importance of considering small density variations in remedial design.     

A three-layer diffusion-cell to examine bio-enhanced dissolution of chloroethene dense non-aqueous phase liquid

Microbial reductive dechlorination of trichloroethene (TCE) and perchloroethene (PCE) in the vicinity of their dense non-aqueous phase liquid (DNAPL) has been shown to accelerate DNAPL dissolution. A three-layer diffusion-cell was developed to quantify this bio-enhanced dissolution and to measure the conditions near the DNAPL interface. The 12 cm long diffusion-cell setup consists of a 5.5 cm central porous layer (sand), a lower 3.5 cm DNAPL layer and a top 3 cm water layer. The water layer is frequently refreshed to remove chloroethenes at the upper boundary of the porous layer, while the DNAPL layer maintains the saturated chloroethene concentration at the lower boundary. Two abiotic and two biotic diffusion-cells with TCE DNAPL were tested. In the abiotic diffusion-cells, a linear steady state TCE concentration profile between the DNAPL and the water layer developed beyond 21 d. In the biotic diffusion-cells, TCE was completely converted into cis-dichloroethene (cis-DCE) at 2.5 cm distance of the DNAPL. Dechlorination was likely inhibited up to a distance of 1.5 cm from the DNAPL, as in this part the TCE concentration exceeded the culture’s maximum tolerable concentration (2.5 mM). The DNAPL dissolution fluxes were calculated from the TCE concentration gradient, measured at the interface of the DNAPL layer and the porous layer. Biotic fluxes were a factor 2.4 (standard deviation 0.2) larger than abiotic dissolution fluxes. This diffusion-cell setup can be used to study the factors affecting the bio-enhanced dissolution of DNAPL and to assess bioaugmentation, pH buffer addition and donor delivery strategies for source zones.     

Variations in expression of carbon isotope fractionation of chlorinated ethenes during biologically enhanced PCE dissolution close to a source zone

The stable carbon isotope values of tetrachloroethene (PCE) and its degradation products were monitored during studies of biologically enhanced dissolution of PCE dense nonaqueous phase liquid (DNAPL) to determine the effect of PCE dissolution on observed isotope values. The degradation of PCE was monitored in a 2-dimensional model aquifer and in a pilot test cell (PTC) at Dover Air Force Base, both with emplaced PCE DNAPL sources. Within the plume down gradient from the source, the isotopic fractionation of dissolved PCE and its degradation products were consistent with those observed in biodegradation laboratory studies. However, close to the source zone significant shifts in the isotope values of dissolved PCE were not observed in either the model aquifer or PTC due to the constant input of newly dissolved, non fractionated PCE, and the small isotopic fractionation associated with PCE reductive dechlorination by the mixed microbial culture used. Therefore the identification of reductive dechlorination in the presence of PCE DNAPL was based upon the appearance of daughter products and the isotope values of those daughter products. An isotope model was developed to simulate isotope values of PCE during the dissolution and degradation of PCE adjacent to a DNAPL source zone. With the exception of very high degradation rate constants (>1/day) stable carbon isotope values of PCE estimated by the model remained within error of the isotope value of the PCE DNAPL, consistent with measured isotope values in the model aquifer and in the PTC.     

PCE dissolution and simultaneous dechlorination by nanoscale zero-valent iron particles in a DNAPL source zone

While the capability of nanoscale zero-valent iron (NZVI) to dechlorinate organic compounds in aqueous solutions has been demonstrated, the ability of NZVI to remove dense non-aqueous phase liquid (DNAPL) from source zones under flow-through conditions similar to a field scale application has not yet been thoroughly investigated. To gain insight on simultaneous DNAPL dissolution and NZVI-mediated dechlorination reactions after direct placement of NZVI into a DNAPL source zone, a combined experimental and modeling study was performed. First, a DNAPL tetrachloroethene (PCE) source zone with emplaced NZVI was built inside a small custom-made flow cell and the effluent PCE and dechlorination byproducts were monitored over time. Second, a model for rate-limited DNAPL dissolution and NZVI-mediated dechlorination of PCE to its three main reaction byproducts with a possibility for partitioning of these byproducts back into the DNAPL was formulated. The coupled processes occurring in the flow cell were simulated and analyzed using a detailed three-dimensional numerical model. It was found that subsurface emplacement of NZVI did not markedly accelerate DNAPL dissolution or the DNAPL mass-depletion rate, when NZVI at a particle concentration of 10g/L was directly emplaced in the DNAPL source zone. To react with NZVI the DNAPL PCE must first dissolve into the groundwater and the rate of dissolution controls the longevity of the DNAPL source. The modeling study further indicated that faster reacting particles would decrease aqueous contaminant concentrations but there is a limit to how much the mass removal rate can be increased by increasing the dechlorination reaction rate. To ensure reduction of aqueous contaminant concentrations, remediation of DNAPL contaminants with NZVI should include emplacement in a capture zone down-gradient of the DNAPL source.     

Refinement of the density-modified displacement method for efficient treatment of tetrachloroethene source zones

A novel method to remediate dense nonaqueous phase liquid (DNAPL) source zones that incorporates in situ density conversion of DNAPL via alcohol partitioning followed by displacement with a low interfacial tension (IFT) surfactant flood has been developed. Previous studies demonstrated the ability of the density-modified displacement (DMD) method to recover chlorobenzene (CB) and trichloroethene (TCE) from heterogeneous porous media without downward migration of the dissolved plume or free product. However, the extent of alcohol (n-butanol) partitioning required for in situ density conversion of high-density NAPLs, such as tetrachloroethene (PCE), could limit the utility of the DMD method. Hence, the objective of this study was to compare the efficacy of two n-butanol delivery approaches: an aqueous solution of 6% (wt) n-butanol and a surfactant-stabilized macroemulsion containing 15% (vol) n-butanol in water, to achieve density reduction of PCE-NAPL in two-dimensional (2-D) aquifer cells. Results of liquid-liquid equilibrium studies indicated that density conversion of PCE relative to water occurred at an n-butanol mole fraction of 0.56, equivalent to approximately 5 ml n-butanol per 1 ml of PCE when in equilibrium with an aqueous solution. In 2-D aquifer cell studies, density conversion of PCE was realized using both n-butanol preflood solutions, with effluent NAPL samples exhibiting density reductions ranging from 0.51 to 0.70 g/ml. Although the overall PCE mass recoveries were similar (91% and 93%) regardless of the n-butanol delivery method, the surfactant-stabilized macroemulsion preflood removed approximately 50% of the PCE mass. In addition, only 1.2 pore volumes of the macroemulsion solution were required to achieve in situ density conversion of PCE, compared to 6.4 pore volumes of the 6% (wt) n-butanol solution. These findings demonstrate that use of the DMD method with a surfactant-stabilized macroemulsion containing n-butanol holds promise as an effective source zone remediation technology, allowing for efficient recovery of PCE-DNAPL while mitigating downward migration of the dissolved plume and free product.     

Laboratory evidence of natural remobilization of multicomponent DNAPL pools due to dissolution

Mixtures of dense non-aqueous phase liquids (DNAPLs) trapped in the subsurface can act as long-term sources of contamination by dissolving into flowing groundwater. In general, the components of higher solubility are removed more quickly, thus altering the composition of the remaining DNAPL, and possibly leading to changes in its physical properties. Through the development of a simple compositional model, Roy et al. [J. Contam. Hydrol. 2002 (59) 163] showed that preferential dissolution of a mixed DNAPL could potentially result in changes in density and interfacial tension that could subsequently lead to remobilization of an initially static DNAPL pool. The laboratory experiments presented in this next paper provide a proof-of-concept for the previously presented theory, demonstrating and quantifying this process of remobilization. In addition, the experiments provide a data set for evaluation of the model presented by Roy et al. [J. Contam. Hydrol. 2002 (59) 163]. In the four experiments, a DNAPL pool comprised of tetrachloroethene and benzene was created as an open pool overlying glass beads within a water-saturated 2-D flow box. Experiments included rectangular and triangular pools. In each of the experiments, remobilization (as breakthrough) was observed more than 2 weeks after formation of the initial pool. During each experiment, the pool height declined as mass was lost by dissolution, while sampling indicated a decrease in the mole fraction of benzene, the more soluble component. Small protuberances formed along the bottom of the pool as its composition changed with time and the displacement pressure was achieved for various pore throats. Eventually one of the protuberances extended further, forming a finger (breakthrough). In general, the pool emptied as the finger proceeded further into the beads. It was also shown theoretically and experimentally that remobilization will occur sooner for pools with a triangular (pointing down), rather than rectangular, shape. The experimental results were simulated using the model developed by Roy et al. [J. Contam. Hydrol. 2002 (59) 163]. The model matched the observations well, suggesting that it accurately represents the primary mechanisms involved with natural remobilization under the conditions of the study.     

Dip-angle influence on areal DNAPL recovery by co-solvent flooding with and without pre-flooding

A two-dimensional (2D) laboratory model was used to study effects of gravity on areal recovery of a representative dense non-aqueous phase liquid (DNAPL) contaminant by an alcohol pre-flood and co-solvent flood in dipping aquifers. Recent studies have demonstrated that injection of alcohol and co-solvent solutions can be used to reduce in-situ the density of DNAPL globules and displace the contaminant from the source zone. However, contact with aqueous alcohol reduces interfacial tension and causes DNAPL swelling, thus facilitating risk of uncontrolled downward DNAPL migration. The 2D laboratory model was operated with constant background gradient flow and a DNAPL spill was simulated using tetrachloroethene (PCE). The spill was dispersed to a trapped, immobile PCE saturation by a water flood. Areal PCE recovery was studied using a double-triangle well pattern to simulate a remediation scheme consisting of an alcohol pre-flood using aqueous isobutanol ( approximately 10% vol.) followed by a co-solvent flood using a solution of ethylene glycol (65%) and 1-propanol (35%). Experiments were conducted with the 2D model oriented in the horizontal plane and compared to experiments at the 15 degrees and 30 degrees dip-angle orientations. Injection was applied either in the downward or upward direction of flow. Experimental results were compared to theoretical predictions for flood front stability and used to evaluate effects of gravity on areal PCE recovery. Sensitivity experiments were performed to evaluate effects of the alcohol pre-flood on PCE areal recovery. For experiments conducted with the alcohol pre-flood and the 2D model oriented in the horizontal plane, results indicate that 89-93% of source zone PCE was recovered. With injection oriented downward, results indicate that areal PCE recovery was 70-77% for a 15 degrees dip angle and 57-59% for a 30 degrees dip angle. With injection oriented upward, results indicate that areal PCE recovery was 57-60% at the 30 degrees dip angle, which was similar to PCE recovery for injection in the downward flow direction. Lower areal PCE recovery at greater dip angles in either direction of flow was attributed to DNAPL swelling and migration, flood front instabilities and bypassing of the displaced fluid past the extraction wells during the alcohol pre-flood. Additional results demonstrate that the use of an alcohol pre-flood can be beneficial in improving DNAPL recovery in the horizontal orientation, but pre-flooding may reduce areal recovery efficiency in dip-angle orientations. This study also demonstrates the use of theoretical perturbation (fingering) analysis in predicting NAPL recovery efficiency for flooding processes in remediating aquifers with dip angles.     

Improving the extraction of tetrachloroethylene from soil columns using surfactant gradient systems

In this work, we extend the recently developed gradient approach for surfactant-enhanced remediation of dense non-aqueous phase liquid (DNAPL)-impacted sites. The goal of the gradient approach is to maximize the DNAPL solubilization capacity in swollen micelles (Type I aqueous microemulsions) while at the same time minimizing the potential for DNAPL mobilization. In this work, we introduce a modified version of the capillary/trapping curve that we refer to as the gradient curve to help interpret and/or design the gradient approach. The gradient curve presents the residual DNAPL saturation as a function of interfacial tension and microemulsion viscosity. This approach demonstrates that keeping a low viscosity of the microemulsion phase is not only important for keeping a low head loss during surfactant flooding but also to prevent oil mobilization. Eight microemulsion systems were evaluated in this research; these systems were evaluated based on their tetrachloroethylene (PCE) solubilization capacity, interfacial tension (IFT), viscosity, density, and coalescence kinetics. Two of these systems were chosen for evaluation in site-specific column tests using an increasing electrolyte gradient to produce a decreasing IFT/increasing solubilization gradient system. The column studies were conducted with media from Dover Air Force Base in Dover, DE. Both solubilized and mobilized DNAPL were quantified. During the column studies, we observed that substantial PCE was mobilized when the residual level of PCE in the column was significantly higher than the steady-state residual saturation level being approach (as predicted from the gradient curve). Four column studies were performed, three of which were used to asses the validity of the gradient curve in predicting the residual saturation after each gradient step. From these tests we observed that starting IFTs of less than 1 mN/m all produced the same mobilization potential. In the last column, we used an additional gradient step with an initial IFT above 1 mN/m to dramatically reduce the amount of PCE mobilize. Based on the good agreement between column results and projections based on the gradient curve, we propose this as a preferred method for designing gradient surfactant flushing systems.     

Removal of DNAPL contamination from the saturated zone by the combined effect of vertical upward flushing and density reduction

A common aspect of innovative remediation techniques is that they tend to reduce the interfacial tension between the aqueous and non-aqueous phase liquids, resulting in mobilization of the organic contaminant. This complicates the remediation of aquifers, contaminated with Dense Non-Aqueous Phase Liquids (DNAPLs), as they are likely to migrate downwards, deeper into the aquifer and into finer layers. A possible solution is the use of swelling alcohols, which tend to reduce the density difference between the aqueous phase and the DNAPL. To avoid premature mobilization upon the initial contact between the DNAPL and the alcohol, several researchers have proposed the use of vertical upward flow of the alcohol. In this paper, we present an equation, which describes the upward mobilization of both continuous and discontinuous DNAPLs and so the important parameters governing the upward controlled mobilization of the DNAPL. The need and required magnitude of this specific discharge was investigated by conducting four column experiments in which the initial density of the DNAPL and the permeability was varied. It was shown that the required flow velocities increase with the permeability of the porous medium and the initial density difference between the aqueous phase and the DNAPL. Whenever the specific discharge falls below the critical value, the DNAPL moves downward. A second set of column experiments looked at the impact of permeability of porous medium on the solubilization and mobilization of DNAPL during alcohol flooding. Columns, packed with coarse or fine sand, containing a residual trichloroethylene (TCE) or perchloroethylene (PCE) saturation were flushed with the alcohol mixture at a fixed specific discharge rate. The induced pressure gradients in the aqueous phase, which were higher in the fine sand, resulted for this porous medium in extensive mobilization of the DNAPL against the direction of the buoyancy force. The density of the first NAPL coming out of the top of the fine sand was close to that of the pure DNAPL. In the coarser sand, the pressure gradients were sufficient to prevent downward migration of the DNAPL, but upward mobilization was minimal. The predominant removal mechanism in this case was the much slower solubilization.