Laboratory-scale characterization of slow-release permanganate gel (SRP-G) for the in-situ treatment of chlorinated-solvent groundwater plumes.

Significant challenges remain for the remediation of chlorinated-solvent plumes in groundwater, such as trichloroethene (TCE) and tetrachloroethene (PCE). A novel slow-release permanganate gel (SRP-G) technique may show promise for the in-situ treatment (remediation) of chlorinated contaminant plumes in groundwater. A series of laboratory experiments were conducted to characterize the primary physical factors that influence SRP-G gelation processes to optimize SRP-G performance for plume treatment. Specifically, experiments were conducted to quantify gel zeta potential, particle size distribution, and viscosity to determine SRP-G gelation characteristics and processes. These experiments tested various concentrations of two SRP-G amendment solutions (NaMnO4 and KMnO4) prepared with 30-wt.% and 50-wt.% colloidal silica to determine such influences on zeta potential, particle size distribution, and viscosity. The results of this study show that SRP-G solutions with low zeta potential and relatively high pH favor more rapid SRP-G gelation. The concomitant interaction of the predominantly negatively charged colloidal silica particles and the positively charged dissociated cations (Na+ and K+) in the SRP-G solution had the effect of stabilizing charge imbalance via attraction of particles and thereby inducing a greater influence on the gelation process. Gel particle size distribution and changes in viscosity had a significant influence on SRP-G solution gelation. The addition of permanganate (NaMnO4 or KMnO4) increased the average particle size distribution and the viscosity of the SRP-G solution and decreased the overall gelation time. SRP-G amendments (NaMnO4 or KMnO4) prepared with 50-wt.% colloidal silica showed more effective gelation (and reduced gelation time) compared to SRP-G amendments prepared with 30-wt.% colloidal silica. Under the conditions of these experiments, it was determined that both the 7-wt.% NaMnO4 solution and 90 mg/L KMnO4 solution using 50-wt.% colloidal silica would be the optimal injection SRP-G solution concentrations for this in-situ treatment technique.

Rethinking pump-and-treat remediation as maximizing contaminated groundwater.

For over half a century, the United States has developed water quality regulations (e.g., Safe Drinking Water Act), which has been accompanied by innumerable advances in contaminant transport and fate, site characterization, and remediation. Since the 1980s, "pump-and-treat" techniques have been the most widely used methods for groundwater contamination remediation. By 1982, pump-and-treat was included in 100 % of the U.S. Superfund groundwater remedy decisions, but applications decreased continuously after 1992. This was likely associated with the documented limitations of pump-and-treat for achieving complete remediation with site closure. Several factors can limit the effectiveness of pump-and-treat, a primary one being that contaminant mass residing in NAPL, sorbed, and low-permeability matrices is not removed in an effective or efficient manner. This ineffectiveness leads to extended cleanup times and the generation of enormous volumes of extracted groundwater, in effect creating conditions of maximizing the amount of contaminated groundwater needing treatment. We highlight a means by which to reassess our approach to remediation by recognizing that pump-and-treat, due to its well-documented limitations, often maximizes the generation of contaminated groundwater.

Global distributions, source-type dependencies, and concentration ranges of per- and polyfluoroalkyl substances in groundwater.

A meta-analysis was conducted of published literature reporting concentrations of per- and polyfluoroalkyl substances (PFAS) in groundwater for sites distributed in 20 countries across the globe. Data for >35 PFAS were aggregated from 96 reports published from 1999 to 2021. The final data set comprises approximately 21,000 data points after removal of time-series and duplicate samples as well as non-detects. The reported concentrations range over many orders of magnitude, from ng/L to mg/L levels. Distinct differences in concentration ranges are observed between sites located within or near sources versus those that are not. Perfluorooctanoic acid (PFOA), ranging from <0.03 ng/L to ~7 mg/L, and perfluorooctanesulfonic acid (PFOS), ranging from 0.01 ng/L to ~5 mg/L, were the two most reported PFAS. The highest PFAS concentration in groundwater is ~15 mg/L reported for the replacement-PFAS 6:2 fluorotelomer sulfonate (6:2 FTS). Maximum reported groundwater concentrations for PFOA and PFOS were compared to concentrations reported for soils, surface waters, marine waters, and precipitation. Soil concentrations are generally significantly higher than those reported for the other media. This accrues to soil being the primary entry point for PFAS release into the environment for many sites, as well as the generally significantly greater retention capacity of soil compared to the other media. The presence of PFAS has been reported for all media in all regions tested, including areas that are far removed from specific PFAS sources. This gives rise to the existence of a "background" concentration of PFAS that must be accounted for in both regional and site-specific risk assessments. The presence of this background is a reflection of the large-scale use of PFAS, their general recalcitrance, and the action of long-range transport processes that distribute PFAS across regional and global scales. This ubiquitous distribution has the potential to significantly impact the quality and availability of water resources in many regions. In addition, the pervasive presence of PFAS in the environment engenders concerns for impacts to ecosystem and human health.

Groundwater level modeling framework by combining the wavelet transform with a long short-term memory data-driven model.

Developing models that can accurately simulate groundwater level is important for water resource management and aquifer protection. In particular, machine learning tools provide a new and promising approach to efficiently forecast long-term groundwater table fluctuations without the computational burden of building a detailed flow model. This study proposes a multistep modeling framework for simulating groundwater levels by combining the wavelet transform (WT) with the long short-term memory (LSTM) network; the framework is named the combined WT-multivariate LSTM (WT-MLSTM) method. First, the WT decomposes the groundwater level time series (i.e., the training stage) into a self-control term and a set of external-control terms. Second, Pearson correlation analysis reveals the correlations between the influencing factors (i.e., river stage) and the groundwater table, and the multivariate LSTM model incorporating external factors is built to simulate the external-control terms. Third, the spatiotemporal evolution of the groundwater level is modeled by reconstructing the sequence of each term of the groundwater level time series. Methodological applications in the Liangshui River Basin, Beijing, China and the Cibola National Wildlife Refuge along the lower Colorado River, United States, show that the combined WT-MLSTM model has a higher simulation accuracy than the standard LSTM, MLSTM, and WT-LSTM models. A comparison between the combined WT-MLSTM model and support vector machine (SVM) also demonstrates the advantage of the proposed model. Additional comparison between model forecasts and observed groundwater levels shows the model predictability for short-term time series. Further analysis reveals that the applicability of the combined WT-MLSTM model decreases with increasing distance between the groundwater well and adjacent river channel, or with the increasing complexity of the changing groundwater level patterns, which may be driven by additional controlling factors. This study therefore provides a new methodology/approach for the rapid and accurate simulation and prediction of groundwater level.

Applicability of time fractional derivative models for simulating the dynamics and mitigation scenarios of COVID-19.

Fractional calculus provides a promising tool for modeling fractional dynamics in computational biology, and this study tests the applicability of fractional-derivative equations (FDEs) for modeling the dynamics and mitigation scenarios of the novel coronavirus for the first time. The coronavirus disease 2019 (COVID-19) pandemic radically impacts our lives, while the evolution dynamics of COVID-19 remain obscure. A time-dependent Susceptible, Exposed, Infectious, and Recovered (SEIR) model was proposed and applied to fit and then predict the time series of COVID-19 evolution observed over the last three months (up to 3/22/2020) in China. The model results revealed that 1) the transmission, infection and recovery dynamics follow the integral-order SEIR model with significant spatiotemporal variations in the recovery rate, likely due to the continuous improvement of screening techniques and public hospital systems, as well as full city lockdowns in China, and 2) the evolution of number of deaths follows the time FDE, likely due to the time memory in the death toll. The validated SEIR model was then applied to predict COVID-19 evolution in the United States, Italy, Japan, and South Korea. In addition, a time FDE model based on the random walk particle tracking scheme, analogous to a mixing-limited bimolecular reaction model, was developed to evaluate non-pharmaceutical strategies to mitigate COVID-19 spread. Preliminary tests using the FDE model showed that self-quarantine may not be as efficient as strict social distancing in slowing COVID-19 spread. Therefore, caution is needed when applying FDEs to model the coronavirus outbreak, since specific COVID-19 kinetics may not exhibit nonlocal behavior. Particularly, the spread of COVID-19 may be affected by the rapid improvement of health care systems which may remove the memory impact in COVID-19 dynamics (resulting in a short-tailed recovery curve), while the death toll and mitigation of COVID-19 can be captured by the time FDEs due to the nonlocal, memory impact in fatality and human activities.

1,4-Dioxane cosolvency impacts on trichloroethene dissolution and sorption.

Solvent stabilizer 1,4-dioxane, an emerging recalcitrant groundwater contaminant, was commonly added to chlorinated solvents such as trichloroethene (TCE), and the impact of co-disposal on contaminant transport processes remains uncertain. A series of batch equilibrium experiments was conducted with variations of 1,4-dioxane and TCE composition to evaluate aqueous dissolution of the two components and their sorption to aquifer sediments. The solubility of TCE increased with increasing amounts of 1,4-dioxane, indicating that 1,4-dioxane acts as a cosolvent causing solubility enhancement of co-contaminants. The solubilization results compared favorably with predictions using the log-linear cosolvency model. Equilibrium sorption coefficients (Kd and Kf) were also measured for different 1,4-dioxane and TCE compositions, and the findings indicate that both contaminants adsorb to aquifer sediments and TCE Kd values increased with increasing organic matter content. However, the Kd for TCE decreased with increases in 1,4-dioxane concentration, which was attributed to cosolvency impacts on TCE solubility. These findings further advance our understanding of the mass-transfer processes controlling groundwater plumes containing 1,4-dioxane, and also have implications for the remediation of 1,4-dioxane contamination.

Experimental comparison of agent-enhanced flushing for the recovery of crude oil from saturated porous media.

The subsurface remediation of nonaqueous liquid (NAPL) has proven to be challenging even when implementing more aggressive enhanced-flushing techniques. The objective of this study was to evaluate the effectiveness of a combination of alkaline- and surfactant-based enhanced flushing for the removal of crude oil (medium fraction) from saturated porous media. Synchrotron X-ray microtomography (SXM) was used to perform pore-scale examination of NAPL fragmentation and changes in blob morphology, and recovery using three different advective flushing methods: surface-active agent (surfactant) flushing, alkaline flushing, and sequential alkaline-surfactant flushing. This set of experiments was conducted to understand effects on such processes (fragmentation and recovery) as a function of media composition (geochemical/mineralogical) and pH alterations due to calcium-carbonate fraction. Results showed that the sequential flushing technique (alkaline→ surfactant) yielded the highest recovery, 32% after 5 pore volumes (PV) of flushing. The crude oil (NAPL) distribution varied due to differences in porous medium mixture composition and type of fluid (i.e. surfactant vs. alkaline) used for flushing. The results of this study can be used to aid in the understanding of physical and chemical parameters/properties that control mobilization of crude oil in saturated porous media. This can help reduce time and cost during remediation of contaminated sites that contain crude oil or less dense NAPL derivatives consistent with fuel-type petroleum hydrocarbons.

A pore-scale investigation of heavy crude oil trapping and removal during surfactant-enhanced remediation.

The presence of nonaqueous phase liquid (NAPL) in the subsurface presents significant challenges for soil and groundwater remediation. In particular, heavy crude oil, coal tar and/or bitumen present unique difficulties for removal and cleanup due to associated high viscosities, low aqueous solubilities, and limited mobility extraction potential. Although surfactant-enhanced aquifer remediation (SEAR) techniques have shown some promise for source removal, overall remediation (mobilization) performance will depend significantly on interfacial effects between the fluid and solid phases. A pore-scale study, implementing synchrotron X-ray microtomography (SXM), was conducted to understand and quantify the trapping and mobilization mechanisms and in-situ emulsification processes of heavy crude oil distributed within increasing complexity (i.e. physical heterogeneity) unconsolidated sands during surfactant flushing events. Pore-scale imaging analyses were conducted to quantify the changes in oil blob morphology before and after surfactant flushing events to assess the primary factors controlling the recovery. Results showed relatively low (10%) net recovery from the homogeneous sand after 5 pore volumes (PVs) of surfactant flushing and may be, in part, due to the more connected ganglia (i.e. single continuous) oil-phase. Such a condition may have limited the surfactant/oil contact resulting in relatively low interfacial activity and correspondingly inefficient oil mobilization and recovery. Negligible net oil recovery was achieved from the mildly-heterogeneous-sand and is likely due to the lower associated permeability of this particular porous medium. Furthermore, the oil-phase distribution within this medium primarily consisted of small disconnected blobs more readily exposed (in contact with) the surfactant solution. For the highly-heterogeneous-sand experiments, an average of 20% heavy-oil recovery resulted after each flushing event (total of ~37% after 5 PVs) and was attributed to more efficient reduction of interfacial tension associated with the increased surfactant-oil contact. The associated higher pH sand/fine­carbonate system may have aided in maintaining a water-wet porous medium, a condition more conducive to higher oil recovery and displacement efficiency.

In situ stabilization of NAPL contaminant source-zones as a remediation technique to reduce mass discharge and flux to groundwater.

Widely used flushing and in-situ destruction based remediation techniques (i.e. pump-and treat, enhanced-solubilization, and chemical oxidation/reduction) for sites contaminated by nonaqueous phase liquid (NAPL) contaminant sources have been shown to be ineffective at complete mass removal and reducing aqueous-phase contaminant of concern (COC) concentrations to levels suitable for site closure. A remediation method was developed to reduce the aqueous solubility and mass-flux of COCs within NAPL through the in-situ creation of a NAPL mixture source-zone. In contrast to remediation techniques that rely on the rapid removal of contaminant mass, this technique relies on the stabilization of difficult-to-access NAPL sources to reduce COC mass flux to groundwater. A specific amount (volume) of relatively insoluble n-hexadecane (HEXDEC) or vegetable oil (VO) was injected into a trichloroethene (TCE) contaminant source-zone through a bench-scale flow cell port (i.e. well) to form a NAPL mixture of targeted mole fraction (TCE:HEXDEC or TCE:VO). NAPL-aqueous phase batch tests were conducted prior to the flow-cell experiments to evaluate the effects of various NAPL mixture ratios on equilibrium aqueous-phase concentrations of TCE to design optimal NAPL (HEXDEC or VO) injection volumes for the flow-cell experiments. The NAPL-stabilization flow-cell experiments initiated and sustained significant reductions in COC concentration and mass flux due to a combination of both reduced relative permeability (increased NAPL-saturation) and via modification of NAPL composition (decreased TCE mole fraction). Variations in remediation performance (i.e. impacts on TCE concentration and mass flux reduction) between the different HEXDEC injection volumes were relatively minor, and therefore inconsistent with Raoult's Law predictions. This phenomenon likely resulted from non-uniform mixing of the injected HEXDEC with TCE in the source-zone. VO injection caused TCE concentrations and mass-flux to decrease more rapidly than with HEXDEC injections. This phenomenon occurred because the injected VO was observed to mix more uniformly with TCE in the source-zone due to a lower mobilization potential. The relative lower density differences (buoyancy effects) between VO and the flushing solution (water) was the primary factor contributing to the lower mobilization potential for VO. Overall, this study indicated that the delivery of HEXDEC or VO into the toxic TCE source-zone was effective in significantly reducing contaminant aqueous-phase concentration and mass-flux. However, the effectiveness of this in-situ NAPL stabilization technique depends on source delivery, uniform mixing of amendment, and that the amendment remains immobilized within and around the NAPL contaminant source.

Chemical structure influence on NAPL mixture nonideality evolution, rate-limited dissolution, and contaminant mass flux.

The influence of chemical structure on NAPL mixture nonideality evolution, rate-limited dissolution, and contaminant mass flux was examined. The variability of measured and UNIFAC modeled NAPL activity coefficients as a function of mole fraction was compared for two NAPL mixtures containing structurally-different contaminants of concern including toluene (TOL) or trichloroethene (TCE) within a hexadecane (HEXDEC) matrix. The results showed that dissolution from the NAPL mixtures transitioned from ideality for mole fractions >0.05 to nonideality as mole fractions decreased. In particular, the TCE generally exhibited more ideal dissolution behavior except at lower mole fractions, and may indicate greater structural/polarity similarity between the two compounds. Raoult's Law and UNIFAC generally under-predicted the batch experiment results for TOL:HEXDEC mixtures especially for mole fractions ≤0.05. The dissolution rate coefficients were similar for both TOL and TCE over all mole fractions tested. Mass flux reduction (MFR) analysis showed that more efficient removal behavior occurred for TOL and TCE with larger mole fractions compared to the lower initial mole fraction mixtures (i.e. <0.2). However, compared to TOL, TCE generally exhibited more efficient removal behavior over all mole fractions tested and may have been the result of structural and molecular property differences between the compounds. Activity coefficient variability as a function of mole fraction was quantified through regression analysis and incorporated into dissolution modeling analyses for the dynamic flushing experiments. TOL elution concentrations were modeled (predicted) reasonable well using ideal and equilibrium assumptions, but the TCE elution concentrations could not be predicted using the ideal model. Rather, the dissolution modeling demonstrated that TCE elution was better described by the nonideal model whereby NAPL-phase activity coefficient varied as a function of COC mole fraction. For dynamic column flushing experiments, dissolution rate kinetics can vary significantly with changes in NAPL volume and surface area. However, under conditions whereby NAPL volume and area are not significantly altered during dissolution, mixture nonideality effects may have a greater relative control on dissolution (elution) and MFR behavior compared to kinetic rate limitations.

Peclet number as affected by molecular diffusion controls transient anomalous transport in alluvial aquifer-aquitard complexes.

This study evaluates the role of the Peclet number as affected by molecular diffusion in transient anomalous transport, which is one of the major knowledge gaps in anomalous transport, by combining Monte Carlo simulations and stochastic model analysis. Two alluvial settings containing either short- or long-connected hydrofacies are generated and used as media for flow and transport modeling. Numerical experiments show that 1) the Peclet number affects both the duration of the power-law segment of tracer breakthrough curves (BTCs) and the transition rate from anomalous to Fickian transport by determining the solute residence time for a given low-permeability layer, 2) mechanical dispersion has a limited contribution to the anomalous characteristics of late-time transport as compared to molecular diffusion due to an almost negligible velocity in floodplain deposits, and 3) the initial source dimensions only enhance the power-law tail of the BTCs at short travel distances. A tempered stable stochastic (TSS) model is then applied to analyze the modeled transport. Applications show that the time-nonlocal parameters in the TSS model relate to the Peclet number, Pe. In particular, the truncation parameter in the TSS model increases nonlinearly with a decrease in Pe due to the decrease of the mean residence time, and the capacity coefficient increases with an increase in molecular diffusion which is probably due to the increase in the number of immobile particles. The above numerical experiments and stochastic analysis therefore reveal that the Peclet number as affected by molecular diffusion controls transient anomalous transport in alluvial aquifer-aquitard complexes.

Understanding the sources and fate of nitrate in a highly developed aquifer system.

Understanding the processes affecting the transport and fate of nitrate in coastal aquifers has become of great interest in recent years due to concerns of nutrient loading to coastal waters. Novel dual isotopic methods have shown promise for identifying sources and fate of nitrate in shallow groundwater. However, in relatively deep dynamic aquifer systems, the isotopic signatures may be overprinted by mixing of different end-member waters and biogeochemical processes. In this study, δ(15)N and δ(18)O of groundwater nitrate are coupled with other forensic geochemistry methods such as Cl/Br, SO4/Cl, and Cl/NO3 mass ratios and land use analysis in order to constrain the isotope correlations and better understand contaminant sources and biogeochemical processes. Most δ(15)NNO3 values were within ranges expected for nitrate formed by ammonia nitrification in soil. Furthermore, the persistent presence of nitrate in concentrations above background levels (median 2.3 mg/L) and the relatively low δ(15)NNO3 and δ(18)ONO3 (median: 4.5±0.2‰ AIR and 5.2±0.5‰ VSMOW, respectively) indicate no direct evidence of denitrification. However, denitrification was inferred for a few samples whereby more enriched δ(15)NNO3 and δ(18)ONO3 values coupled with an increase in SO4/Cl and Cl/NO3 ratios were observed. Finally, mixing trends were identified for a few of the samples as indicated by δ(15)NO3 and δ(18)ONO3 mixing ratios and were consistent with the study area's land-use/land-cover distribution. The combination of methods utilized in this study revealed that in some cases mass ratios were better diagnostics in elucidating the impact of denitrification, mixing processes, and source identification within dynamic aquifer systems than the dual-isotope technique.

A pore scale investigation of crude oil distribution and removal from homogeneous porous media during surfactant-induced remediation.

A pore-scale study was conducted to understand interfacial processes contributing to the removal of crude oils from a homogeneous porous medium during surfactant-induced remediation. Synchrotron X-ray microtomography (SXM) was used to obtain high-resolution three-dimensional images of the two-fluid-phase oil/water system, and quantify temporal changes in oil blob distribution, blob morphology, and blob surface area before and after sequential surfactant flooding events. The reduction of interfacial tension in conjunction with the sufficient increase in viscous forces as a result of surfactant flushing was most likely responsible for mobilization and recovery of the two lighter oil fractions. However, corresponding increases in viscous forces as a result of a reduction of interfacial tension were insufficient to initiate and maintain the displacement (recovery) of the heavy crude oil fraction during surfactant flushing. In contrast to the heavy oil system, changes in trapping number for the lighter fraction crude oils were sufficient to initiate mobilization as a result of surfactant flushing. Both light and medium oil fractions showed an increase in the number of blobs and total blob surface area, and a reduction in the total volume after 2 pore volumes (PVs) of surfactant flooding. This increase in surface area was attributed to the change in blob morphology from spherical to more complex non-spherical ganglia shape characteristics. Moreover, the increase in the number of oil blobs from larger to smaller particles after surfactant flushing may have contributed to the greater cumulative oil surface area. Complete recovery of light and medium oil fractions resulted after 5 PVs of surfactant flooding, whereas the displacement efficiency of heavy-oil fraction was severely limited, even after extended periods of flushing. The results of these experiments demonstrate the utility of SXM for quantifying pore-scale interfacial characteristics for specific crude-oil-fraction/porous-medium systems, critical for understanding mobilization/removal relationships in which surfactant-enhanced remediation techniques will be most successful.

Characterization of flow dynamics and vulnerability in a coastal aquifer system.

Traditional aquifer vulnerability techniques primarily rely on spatial property data for a region and are limited by their ability to directly or indirectly assess flow and transport processes occurring from the surface to depth within an aquifer system. The main objective of this study was to investigate groundwater vulnerability in terms of aquifer interconnectivity and flow dynamics. A combination of stable isotopes, groundwater age-dating (radiocarbon), and geomorphic/geogenic spatial analyses was applied to a regional, highly developed coastal aquifer to explain the presence of nitrate at depth. The average δ(13) C value (-17.3 ± 2‰ VPDB, n = 27) is characteristic of groundwater originating from locally infiltrated precipitation through extensively cultivated soils. The average δ(18) O and δD values (-4.0 ± 0.1‰ VSMOW, n = 27; δD: -19.3 ± 1‰ VSMOW, n = 27, respectively) are similar to precipitation water derived from maritime sources feeding the region's surface water and groundwater. Stable and radioactive isotopes reveal significant mixing between shallow and deep aquifers due to high velocities, hydraulic connection, and input of local recharge water to depths. Groundwater overdevelopment enhances deeper and faster modern water downward flux, amplifying aquifer vulnerability. Therefore, aquifer vulnerability is a variable, dependent on the type and degree of stress conditions experienced by a groundwater system as well as the geospatial properties at the near surface.

Remediation of NAPL source zones: lessons learned from field studies at Hill and Dover AFB.

Innovative remediation studies were conducted between 1994 and 2004 at sites contaminated by nonaqueous phase liquids (NAPLs) at Hill and Dover AFB, and included technologies that mobilize, solubilize, and volatilize NAPL: air sparging (AS), surfactant flushing, cosolvent flooding, and flushing with a complexing-sugar solution. The experiments proved that aggressive remedial efforts tailored to the contaminant can remove more than 90% of the NAPL-phase contaminant mass. Site-characterization methods were tested as part of these field efforts, including partitioning tracer tests, biotracer tests, and mass-flux measurements. A significant reduction in the groundwater contaminant mass flux was achieved despite incomplete removal of the source. The effectiveness of soil, groundwater, and tracer based characterization methods may be site and technology specific. Employing multiple methods can improve characterization. The studies elucidated the importance of small-scale heterogeneities on remediation effectiveness, and fomented research on enhanced-delivery methods. Most contaminant removal occurs in hydraulically accessible zones, and complete removal is limited by contaminant mass stored in inaccessible zones. These studies illustrated the importance of understanding the fluid dynamics and interfacial behavior of injected fluids on remediation design and implementation. The importance of understanding the dynamics of NAPL-mixture dissolution and removal was highlighted. The results from these studies helped researchers better understand what processes and scales are most important to include in mathematical models used for design and data analysis. Finally, the work at these sites emphasized the importance and feasibility of recycling and reusing chemical agents, and enabled the implementation and success of follow-on full-scale efforts.

Investigation of small-scale preferential flow with a forced-gradient tracer test.

A new tracer experiment (referred to as MADE-5) was conducted at the well-known Macrodispersion Experiment (MADE) site to investigate the influence of small-scale mass-transfer and dispersion processes on well-to-well transport. The test was performed under dipole forced-gradient flow conditions and concentrations were monitored in an extraction well and in two multilevel sampler (MLS) wells located at 6, 1.5, and 3.75 m from the source, respectively. The shape of the breakthrough curve (BTC) measured at the extraction well is strongly asymmetric showing a rapidly arriving peak and an extensive late-time tail. The BTCs measured at seven different depths in the two MLSs are radically different from one another in terms of shape, arrival times, and magnitude of the concentration peaks. All of these characteristics indicate the presence of a complex network of preferential flow pathways controlling solute transport at the test site. Field-experimental data were also used to evaluate two transport models: a stochastic advection-dispersion model (ADM) based on conditional multivariate Gaussian realizations of the hydraulic conductivity field and a dual-domain single-rate (DDSR) mass-transfer model based on a deterministic reconstruction of the aquifer heterogeneity. Unlike the stochastic ADM realizations, the DDSR accurately predicted the magnitude of the concentration peak and its arrival time (within a 1.5% error). For the multilevel BTCs between the injection and extraction wells, neither model reproduced the observed values, indicating that a high-resolution characterization of the aquifer heterogeneity at the subdecimeter scale would be needed to fully capture 3D transport details.

Pilot-scale demonstration of cyclodextrin as a solubility-enhancement agent for remediation of a tetrachloroethene-contaminated aquifer.

The limitations associated with conventional pump and treat technology have generated interest in using enhanced in-situ flushing as an alternative for remediating source zones contaminated with immiscible liquid. This research investigates the effectiveness of cyclodextrin as a solubility-enhancement agent to enhance the removal of tetrachloroethene (PCE) from a physically isolated section of an aquifer. An important component of this project was the implementation of reagent recovery and reuse. This field experiment presented the rare opportunity, under strict regulatory guidance, to inject PCE into the surficial aquifer cell created with two sets of sheet piles driven into an underlying clay unit. The well-controlled conditions specific to this experiment allowed quantification of mass balances, which is problematic for many contaminated field sites. The fact that mass balances can be obtained provides the ability to determine remediation effectiveness with unusual accuracy for a field project. The saturated zone within the test cell was flushed with a 15 wt % cyclodextrin solution. The cyclodextrin solution increased the aqueous concentration of PCE in the extraction-well effluent to as much as 22 times the concentrations obtained during the water flush conducted prior to the complexing sugar flush (CSF). The seven pore-volume CSF removed the equivalent of approximately 33 L of PCE from the subsurface. This equates to 48% of the total initial mass, based on the volume of PCE present prior to the CSF (68.6 L). Conversely, the seven pore-volume water flush conducted prior to the CSF removed the equivalent of 2.7 L of PCE. The use of cyclodextrin as a flushing agent, especially in a recycling configuration, appears to hold promise for successful remediation of chlorinated-solvent-contaminated source zones.

Dissolution of nonuniformly distributed immiscible liquid: intermediate-scale experiments and mathematical modeling.

The purpose of this work is to examine the effect of nonuniform distributions of immiscible organic liquid on dissolution behavior, with a specific focus on the condition dependency of dissolution (i.e., mass transfer) rate coefficients associated with applying mathematical models of differing complexities to measured data. Dissolution experiments were conducted using intermediate-scale flow cells packed with sand in which well-characterized zones of residual trichloroethene (TCE) and 1,2-dichloroethane (DCA) saturation were emplaced. A dual-energy gamma radiation system was used for in-situ measurement of NAPL saturation. Aqueous concentrations of TCE and DCA measured in the flow-cell effluent were significantly less than solubility, due primarily to dilution associated with the nonuniform immiscible-liquid distribution and bypass flow effects associated with physical heterogeneity. A quantitative analysis of flow and transport was conducted using a three-dimensional mathematical model wherein immiscible-liquid distribution, permeability variability, and sampling effects were explicitly considered. Independent values for the initial dissolution rate coefficients were obtained from dissolution experiments conducted using homogeneously packed columns. The independent predictions obtained from the model provided good representations of NAPL dissolution behavior and of total TCE/DCA mass removed, signifying model robustness. This indicates that for the complex three-dimensional model, explicit consideration of the larger scale factors that influenced immiscible-liquid dissolution in the flow cells allowed the use of a dissolution rate coefficient that represents only local-scale mass transfer processes. Conversely, the use of simpler models that did not explicitly consider the nonuniform immiscible-liquid distribution required the use of dissolution rate coefficients that are approximately 3 orders of magnitude smaller than the values obtained from the column experiments. The rate coefficients associated with the simpler models represent composite or lumped coefficients that incorporate the effects of the larger scale dissolution processes associated with the nonuniform immiscible-liquid distribution, which are not explicitly represented in the simpler models, as well as local-scale mass transfer. These results demonstrate that local-scale dissolution rate coefficients, such as those obtained from column experiments, can be used in models to successfully predict dissolution and transport of immiscible-liquid constituents at larger scales when the larger scale factors influencing dissolution behavior are explicitly accounted for in the model.