Surfactant-enhanced Aquifer Remediation Research Articles

Effects of 3D microstructure of porous media on DNAPL migration and remediation by surface active agents in groundwater

Aquifers composed of porous granular media are important to human beings because they are capable of storing a large amount of groundwater. Contaminant migration and remediation in subsurface environments are strongly influenced by three-dimensional (3D) microstructures of porous media. In this study, fractal models are developed to investigate contaminant transport and surfactant-enhanced aquifer remediation (SEAR) for the regular tetrahedron microstructure (RTM) and right square pyramid microstructure (RSPM). The relationships of permeability and entry pressure are derived for these two kinds of 3D microstructures of granular porous media. Afterward, the difference in perchloroethylene (PCE) migration and SEAR efficiency between RTM and RSPM is investigated by the numerical simulation based on a synthetic heterogeneous granular aquifer. Results indicate that PCE penetrates faster and spreads farther in RSPM-based aquifers compared with RTM-based aquifers. Further, SEAR in RTM-based aquifers can achieve remediation efficiencies of 66.129%–92.214% with a mean of 84.324%, which is clearly lower than the SEAR efficiency of 70.149%–94.773% (with a mean of 89.122%) in RSPM-based aquifers. Findings are significant for understanding the 3D microstructure of porous media and how the microstructure of porous media affects macroscopic contaminant behaviors and remediation.

Sustainable lindane waste remediation: Surfactant-driven residual DNAPL extraction and oxidation in a real landfill (LIFE SURFING)

The LIFE SURFING Project was carried out at the Bailin Landfill in Sabiñánigo, Spain (2020-2022), applying Surfactant Enhanced Aquifer Remediation (SEAR) and In Situ Chemical Oxidation (S-ISCO) in a 60-meter test cell beneath the old landfill, to remediate a contaminated aquifer with dense non-aqueous phase liquid (DNAPL) from nearby lindane production. The project overcame traditional extraction limitations, successfully preventing groundwater pollution from reaching the river.In spring 2022, two SEAR interventions involved the injection of 9.3 m3 (SEAR-1) and 6 m3 (SEAR-2) of aqueous solutions containing 20 g/L of the non-ionic surfactant E-Mulse 3®, with bromide (around 150 mg/L) serving as a conservative tracer. 7.1 and 6.0 m3 were extracted in SEAR-1 and SEAR-2, respectively, recovered 60–70 % of the injected bromide and 30–40 % of the surfactant, confirming surfactant adsorption by the soil. Approximately 130 kg of DNAPL were removed, with over 90 % mobilized and 10 % solubilized. A surfactant-to-DNAPL recovery mass ratio of 2.6 was obtained, a successful value for a fractured aquifer. In September 2022, the S-ISCO phase entailed injecting 22 m3 of a solution containing persulfate (40 g/L), E-Mulse 3® (4 g/L), and NaOH (8.75 g/L) in pulses over 48 h, oxidizing around 20 kg of DNAPL and ensuring low toxicity levels after that.Preceding the SEAR and S-ISCO trials, 2020 and 2021 were dedicated to detailed groundwater flow characterizations, including hydrological and tracer studies. These preliminary investigations allowed the design of a barrier zone between 317 and 557 m from the test cell and the river, situated 900 m away. This zone, integrating alkali dosing, aeration, vapor extraction, and oxidant injection, effectively prevented the escape of fluids to the river. Neither surfactants nor contaminants were detected in river waters post-treatment. The absence of residual phase in test cell wells and reduction of chlorinated compound levels in groundwater were noticed till one year after S-ISCO.

Sustainable application of surfactants in soil remediation: Selective pollutants adsorption and hydrogen peroxide-driven adsorbent regeneration

The uncontrolled disposal of the liquid lindane wastes have led to the formation of dense non-aqueous phase liquids (DNAPL), consisting of 28 chlorinated organic compounds (COCs), contaminating soil and groundwater. Surfactant-enhanced aquifer remediation is proposed as technology to treat these sites. However, the polluted emulsion generated must be manged on-site. In this work a two-step process is applied to treat emulsion composed of E-Mulse® 3 (4 g·L−1) as surfactant and a DNAPL (2 gCOCs·L−1). In the first, the COCs were selectively adsorbed in a granular activated carbon (GAC) column with Fe (II) previously adsorbed (10–20mg·g-1) onto the carbon surface, recovering an aqueous phase with surfactant for their reuse. In the second step, the spent GAC was regenerated with a 40 g·L−1 solution of hydrogen peroxide fed to the column at 2 mL·min−1 to promote the oxidation of the COCs adsorbed in the GAC. The kinetic and adsorption model in a multisolute (surfactant and DNAPL) system has been proposed. Five successive cycles of regeneration/adsorption have been successfully applied in the column process. About 50 % of the COCs were retained from the emulsion, and more than 70 % of the surfactant was recovered. The consumption of unproductive oxidants decreased with the number of regeneration cycles. The water effluent obtained after regeneration of GAC did not present chlorinated compounds desorbed and nontoxic by-products generated, such as short-chain acids.

Selective removal of chlorinated organic compounds from soil flushing emulsions: Adsorbent regeneration with thermal-activated persulfate and surfactant recovery

Hydrophobic organic compounds (HOCs) released into the environment can form nonaqueous phase liquids, contaminating soil and groundwater. Surfactant-enhanced aquifer remediation, followed by extracting a highly contaminated emulsion comprising surfactants and contaminants, is a promising technology for remediating polluted sites. However, the sustainable treatment of the emulsion is crucial. This study presents a two-step method for treating emulsions containing HOCs and surfactants. Firstly, pollutants are selectively adsorbed onto granular activated carbon (GAC) in a column. Secondly, spent GAC is regenerated by oxidizing adsorbed pollutants using persulfate activated at 60 °C. The emulsion was obtained from surfactant-enhanced aquifer remediation at a polluted site with lindane production waste, consisting of E-Mulse® 3 surfactant (12 g L−1) and 28 chlorinated organic compounds (COCs) at a concentration of 9.1 gCOCs L−1. Selective adsorption of COCs was observed since the adsorption kinetic constant was much higher than the E3, being 31.60 and 1.76gGAC−W·mgj−1·h−1 respectively. The adsorption-regeneration cycle was repeated three times. It was found that the GAC adsorbed 90 mgCOCs·gGAC−1 were constant after four cycles (70 % of the COCs adsorption capacity in the first step). The surfactant adsorption decreased by approximately 80 % from 108 to 26 mgE3 gGAC−1 after the third cycle, enabling surfactant recovery from the emulsion. In addition, the remaining amount of persulfate increased from 24 to 39 %, reducing oxidant consumption. Finally, the water effluent obtained after regeneration showed reduced toxicity due to the generation of nontoxic by-products, such as short-chain acids and sulphates. The results obtained proved the GAC-based adsorption/regeneration process was effective and stable over multiple cycles.

Solubilization mechanism and mass-transfer model of anionic-nonionic gemini surfactants for chlorinated hydrocarbons

Accurate prediction and modeling of the quantitative recovery of chlorinated hydrocarbons in an aquifer by anionic-nonionic gemini surfactants requires estimates of their mass-transfer models and capture mechanisms. In this work, a series of synthesized anionic-nonionic gemini surfactants (GEOnS-m) with different molecular structures were used to solubilize different chlorinated hydrocarbons that contaminate the aquifer. GEOnS-m aggregate showed vesicle, rodlike micelle, and spherical micelle as the ethylene oxide content (EO) number increases. Non-polar chlorinated hydrocarbons (PCE, CT) mainly solubilized in the hydrophobic interior of micelles, and polar chlorinated hydrocarbons (TCE, MCB) mainly solubilized in the “palisade layer” of micelles. Solubilization processes were followed employing first-order and second-order kinetics. The solubilization rate increases with the decrease of EO number of GEOnS-m molecules. In essence, lower interfacial tension reduces the mass-transfer resistance of chlorinated hydrocarbons at the two-phase interface. The diameter of vesicles and spherical micelles increases, and rodlike micelles evenly changed to vesicles as chlorinated hydrocarbons transfer into micelles. The solubilization process of GEOnS-m for chlorinated hydrocarbons conforms to a novel “oil diffusion & micelle reformation model” that represents the coupling to two pre-existing models. And the solubilization mass-transfer mathematical model demonstrated the solubilization mass-transfer coefficient is affected by the combination of surfactant and solubilized organic properties. This study advances the application of solubilization for chlorinated hydrocarbons using anionic-nonionic gemini surfactants in surfactant-enhanced aquifer remediation (SEAR) design.

Pore-scale investigation of surfactant-enhanced DNAPL mobilization and solubilization

Surfactant-enhanced aquifer remediation has been proved successful to remove dense non-aqueous phase liquids (DNAPLs) from contaminated sites. However, the underlying mechanisms of the DNAPL mobilization and solubilization at the pore scale remains to be addressed for efficient application to the field remediation system. In this work, the emerging microfluidic and imaging technologies are applied to investigate the dynamics of DNAPL remediation. Visualized experiments of the evolution of DNAPL remediation are performed to study the role of surfactant type, concentration and injection rate. The DNAPL remediation is dominated by mobilization followed by solubilization for most surfactants. Mobilization occurs as soon as surfactants and DNAPL are in contact until forming a new stable phase structure, and the solubilization continues until the end of injection. We observe the breakup behavior of long droplets and ganglia during the mobilization, which is attributed to the surfactant-reduced interfacial tension and thus expedites DNAPL mobilization and redistribution. During the solubilization, the formation of micelles incorporating DNAPL fractions increases the DNAPL concentration gradient and thus enhances the mass transfer, but the rate-limited diffusion of micelles reduces the mass transfer rate coefficient. Increasing the surfactant content and decreasing the injection rate can promote mobilization and solubilization. The DNAPL mobilization ability of the surfactants SDS and SDBS is stronger than SAOS and Tween 80 regardless of the injection rates. Tween 80 may be considered an ideal surfactant of only solubilization but not mobilization is desired. This work elucidates the pore-scale mechanisms during surfactant-enhanced DNAPL remediation, which are beneficial for upscaling studies, predictive modeling, and operation optimization of DNAPL remediation in the field.

The Effects of Spill Pressure on the Migration and Remediation of Dense Non-Aqueous Phase Liquids in Homogeneous and Heterogeneous Aquifers

The spill pressure of the contaminant source is an important factor affecting the amount, location, form, and behavior of the dense non-aqueous phase liquids (DNAPLs) that plume in a contaminated subsurface environment. In this study, perchloroethylene (PCE) infiltration, distribution and, remediation via a surfactant-enhanced aquifer remediation (SEAR) technique for a PCE spill event are simulated to evaluate the effects of the spill pressure of the contaminant source on the DNAPLs’ behavior in two-dimensional homogeneous and heterogeneous aquifers. Five scenarios with different spill pressures of contamination sources are considered to perform the simulations. The results indicate that the spill pressure of the contaminant source has an obvious influence on the distribution of DNAPLs and the associated efficiency of remediation in homogeneous and heterogeneous aquifers. As the spill pressure increases, more and more contaminants come into the aquifer and the spread range of contamination becomes wider and wider. Simultaneously, the remediation efficiency of contamination also decreases from 93.49% to 65.90% as the spill pressure increases from 33.0 kPa to 41.0 kPa for a heterogeneous aquifer with 200 realizations. The simulation results in both homogeneous and heterogeneous aquifers show the same influence of the spill pressure of the contaminant source on PCE behaviors in the two-dimensional model. This study indicates that the consideration of the spill pressure of the contaminant sources (such as underground petrol tanks, underground oil storage, underground pipeline, and landfill leakage) is essential for the disposal of contaminant leakage in the subsurface environment. Otherwise, it is impossible to accurately predict the migration and distribution of DNAPLs and determine the efficient scheme for the removal of contaminant spills in groundwater systems.

Enhanced solubilization of strongly adsorbed organic pollutants using synthetic and natural surfactants in soil flushing: Column experiment simulation

This study investigates the performance of two non-ionic sugar-based surfactants, belonging to the families of synthetic alkylpoliglucosides and biological rhamnolipids, for enhancing the solubilization and mass transfer of strongly adsorbed hydrophobic organic compounds (HOCs) in porous media. Using toluene and perchloroethylene (PCE) as reference pollutants, a continuous configuration test (column experiment) was performed to simulate a soil flushing process under controlled laboratory conditions. The potential applicability of the selected surfactants in remediation processes was evaluated within the context of Surfactant Enhanced Aquifer Remediation (SEAR) technology. Experimental results showed that both surfactants effectively increased the solubilization of contaminants adsorbed on porous media. Compared to H2O flushing, synthetic surfactant increased the solubilization of toluene by 52% and the solubilization of PCE by 81%, while biosurfactant increased solubilization ability by 73% and 92% for toluene and PCE, respectively. Biosurfactant exhibited the greatest solubilization ability, with 92% of adsorbed toluene and 98% of adsorbed PCE solubilized. This research highlights the potential of green surfactants as efficient candidates for remediation processes due to their excellent solubilization ability and environmentally friendly nature.

Treatment of a Complex Emulsion of a Surfactant with Chlorinated Organic Compounds from Lindane Wastes under Alkaline Conditions by Air Stripping.

Surfactant-enhanced aquifer remediation is commonly applied in polluted sites with dense non-aqueous phase liquids (DNAPLs). This technique transfers the contamination from subsoil to an extracted emulsion, which requires further treatment. This work investigated the treatment of a complex emulsion composed of a nonionic surfactant and real DNAPL formed of chlorinated organic compounds (COCs) and generated as a lindane production waste by air stripping under alkaline conditions. The influence of the surfactant (1.5-15 g·L-1), COC concentrations (2.3-46.9 mmol·L-1), and temperature (30-60 °C) on the COC volatilization was studied and modeled in terms of an apparent constant of Henry at pH > 12. In addition, the surfactant stability was studied as a function of temperature (20-60 °C) and surfactant (2-10 g·L-1), COC (0-70.3 mmol·L-1), and NaOH (0-4 g·L-1) concentrations. A kinetic model was successfully proposed to explain the loss of surfactant capacity (SCL). The results showed that alkali and temperature caused the SCL by hydrolysis of the surfactant molecule. The increasing surfactant concentration decreased the COC volatility, whereas the temperature improved the COC volatilization. Finally, the volatilization of COCs in alkaline emulsions by air stripping (3 L·h-1) was performed to evaluate the treatment of an emulsion composed of the COCs (17.6 mmol·kg-1) and surfactant (3.5 and 7 g·L-1). The air stripping was successfully applied to remove COCs (>90%), reaching an SCL of 80% at 60 °C after 8 h. Volatilization can remove COCs from emulsions and break them, enhancing their further disposal.

Feasibility evaluation of novel anionic-nonionic gemini surfactants for surfactant-enhanced aquifer remediation

Surfactant-enhanced aquifer remediation (SEAR) can effectively remove residual dense non-aqueous phase liquids. In this work, a series of environment-friendly anionic-nonionic gemini surfactants, namely, ethylene glycol succinate polyoxyethylene sulfonate (GEOnS-m, n = 3, 5 and 7; m = 10 and 12), were synthesized and evaluated for application in SEAR. The critical micelle concentrations of GEOnS-m were 2.14 × 10−3–1.62 × 10−2 mM, and the molar solubilization ratios of GEOnS-m for tetrachloroethylene were 1.93–3.44, which were 1.5–5 times higher than traditional surfactants. The GEOnS-m solutions remained stable at 5–25 °C, 50 mM salt, and pH = 5.0–9.0. The micelle size decreased as the temperature decreased, which reduced the solubilization capacity (24.9%–37.0%) of the GEOnS-m solutions. Inorganic salts could improve the solubilization capacity of the GEOnS-m solutions by promoting micelle polymerization and increasing the hydrophobic core volume for solubilizing tetrachloroethylene. This work found that 0.80%–4.00% of GEOnS-m adsorbed on the media during simulated column flushing, which was less than the traditional surfactants due to stronger electrostatic repulsion between the micelles and the media. Meanwhile, the solubilization capacity of GEOnS-m was positively correlated with the hydrophobic carbon chain length, while, the adsorption amount in the media and application cost of GEOnS-m were positively correlated with the EO number and negatively correlated with the hydrophobic carbon chain length. Comprehensive evaluation results revealed that GEO3S-12 was preferred for SEAR. This work not only developed suitable surfactants for SEAR but also attempted to further the rational design of surfactants for diverse applications by combining different functional groups into one molecule.

Micellar solubilization of binary organic liquid mixtures for surfactant enhanced aquifer remediation

AbstractAlthough non‐aqueous phase liquids (NAPLs) are typically released to the environment as complex mixtures, most investigations of subsurface remediation focus on single contaminants and ignore the potential effects of co‐constituents on mass recovery and groundwater plume evolution. In this study, we investigate the dissolution and micellar solubilization of binary NAPL mixtures in completely mixed batch reactors and heterogeneous aquifer cells using a commercially‐available nonionic surfactant, Tween® 80. Micellar solubilization of trichloroethene (TCE), tetrachloroethene (PCE), decane, and dodecane measured in batch studies containing binary mixtures of TCE/PCE or decane/dodecane exhibited nonideal behavior that could not be predicted using Raoult's law. For a mixed PCE/decane NAPL, micellar solubilization of PCE was approximately 40% less than predicted, while decane was over 100% greater, which was attributed to expansion of the micelle core. In two aquifer cells containing different size fractions of quartz sand and low‐permeability lenses, the initial NAPL (1:1 TCE:PCE) saturation distributions resulted in “pool” fractions (PFs) of 0.88 and 0.36. During the initial water flood, the greater aqueous solubility of TCE relative to PCE resulted in preferential removal of TCE from the source zone. However, when a 4% (wt) solution of Tween® 80 was introduced, preferential micellar solubilization of PCE relative to TCE resulted in enhanced removal of PCE, with TCE mass discharge reduced from 97% and 90% and PCE mass discharge reduced from 79% and 53%. The observed relationships between mass discharge and mass removal indicating that plume evolution is strongly influenced by the initial NAPL saturation distribution, changes in the mole fraction of NAPL constituents, and regions of high NAPL saturation that persist over time.

Non-Ionic Surfactant Recovery in Surfactant Enhancement Aquifer Remediation Effluent with Chlorobenzenes by Semivolatile Chlorinated Organic Compounds Volatilization.

Surfactant enhanced aquifer remediation is a common treatment to remediate polluted sites with the inconvenience that the effluent generated must be treated. In this work, a complex mixture of chlorobenzene and dichlorobenzenes in a non-ionic surfactant emulsion has been carried out by volatilization. Since this techhnique is strongly affected by the presence of the surfactant, modifying the vapour pressure, and activity coefficient, , a correlation between and surfactant concentration and temperature was proposed for each compound, employing the Surface Response Methodology (RSM). Volatilization experiments were carried out at different temperatures and gas flow rates. A good agreement between experimental and predicted remaining SVCOCs during the air stripping process was obtained, validating the thermodynamic parameters obtained with RSM. Regarding the results of volatilization, at 60 °C 80% of SVCOCs were removed after 6 h, and the surfactant capacity was almost completely recovered so the solution can be recycled in soil flushing.

Synthetic and Natural Surfactants for Potential Application in Mobilization of Organic Contaminants: Characterization and Batch Study

In this paper, we investigated the abilities of five sugar-based synthetic surfactants and biosurfactants from three different families (i.e., alkyl polyglycoside (APG), sophorolipid (SL), and rhamnolipid (RL)) to dissolve and mobilize non-aqueous phase liquid (NAPL) components, i.e., toluene and perchloroethylene (PCE), adsorbed on porous matrices. The objective of this study was to establish a benchmark for the selection of suitable surfactants for the flushing aquifer remediation technique. The study involved a physicochemical characterization of the surfactants to determine the critical micelle concentration (CMCs) and interfacial properties. Subsequently, a batch study, through the construction of adsorption isotherms, made it possible to evaluate the surfactants’ capacities in contaminant mobilization via the reduction of their adsorptions onto a reference adsorbent material, a pine wood biochar (PWB). The results indicate that a synthetic surfactant from the APG family with a long fatty acid chain and a di-rhamnolipid biosurfactant with a shorter hydrophobic group offered the highest efficiency values; they reduced water surface tension by up to 54.7% and 52%, respectively. These two surfactants had very low critical micelle concentrations (CMCs), 0.0071 wt% and 0.0173 wt%, respectively; this is critical from an economical point of view. The batch experiments showed that these two surfactants, at concentrations just five times their CMCs, were able to reduce the adsorption of toluene on PWB by up to 74% and 65%, and of PCE with APG and RL by up to 65% and 86%, respectively. In general, these results clearly suggest the possibility of using these two surfactants in surfactant-enhanced aquifer remediation technology.

Deep learning based optimization under uncertainty for surfactant-enhanced DNAPL remediation in highly heterogeneous aquifers

The optimization for surfactant-enhanced aquifer remediation (SEAR) of dense non-aqueous phase liquid (DNAPL)-contaminated aquifers is usually accompanied by uncertainties, which may arise from the characterization of complex aquifer heterogeneity and DNAPL source zone architecture (SZA) due to measurement sparsity. Optimization under uncertainty is computationally expensive as it involves an enormous number of model runs. The huge computational burden can be alleviated by utilizing a surrogate model for repeated model evaluations. However, most of the developed surrogates are often limited to low-dimensional optimization problems that only consider simplified aquifer heterogeneity. In this study, we developed a multi-objective simulation-optimization framework to optimize the SEAR schemes considering the characterization uncertainties from both highly heterogeneous aquifer permeability and complex SZA. A fast-to-run convolutional neural network (CNN)-based surrogate model was developed to approximate the high-dimensional and highly complex input-output mapping of the DNAPL multiphase flow simulation model. We first used the rejection sampling strategy to generate random realizations of permeability and SZA conditioning on their limited measurements and then formulated a multi-objective optimization under uncertainty problem based on these realizations. The developed 3-D CNN was trained and used as the surrogate for repeated model runs in optimization to identify the optimal SEAR schemes under uncertainty. A 3-D numerical experiment was used to test the performance of the CNN-based simulation-optimization framework. Comprehensive analysis on the obtained Pareto fronts demonstrates that the proposed framework can efficiently identify reliable Pareto-optimal solutions with a 99.8% speedup compared to the traditional optimization coupled with the forward model. Moreover, the optimization considering multiple realizations enables us to perform the risk assessment to locate the risk zone where the NAPL phase possibly exists after remediation, which provides useful information for decision-making.

Dissolution and remobilization of NAPL in surfactant-enhanced aquifer remediation from microscopic scale simulations

In this paper, the dissolution and mobilization of non-aqueous phase liquid (NAPL) blobs in the Surfactant-Enhanced Aquifer Remediation (SEAR) process were upscaled using dynamic pore network modeling (PNM) of three-dimensional and unstructured networks. We considered corner flow and micro-flow mechanisms including snap-off and piston-like movement for two-phase flow. Moreover, NAPL entrapment and remobilization were evaluated using force analysis to develop the capillary desaturation curve (CDC) and predict the onset of remobilization. The corner diffusion mechanism was also applied in the modeling of interphase mass transfer to represent NAPL dissolution as the dominant mass transfer process. In addition, the effect of pore-scale heterogeneity on mass transfer rate coefficient and recovered residual NAPL was considered in the simulations. Sodium dodecyl sulfate (SDS) and Triton X-100 were used as the surfactant for the SEAR process. The results indicate that although surfactants enhance NAPL recovery during two-phase flow, surfactant-enhanced remediation of residual NAPL through dissolution is highly dependent on surfactant type. When SDS ─as a surfactant with high critical micelle concentration (CMC) and low micelle partition coefficient (Km)─ was injected into a NAPL contaminated site, the mass transfer rate coefficient decreased (due to considerable changes in interface chemical potentials) which leads to a significant reduction in NAPL recovery after the end of two-phase flow. In contrast, Triton X-100 (with low CMC and high Km) improved NAPL recovery, by enhancing solubility at surfactant concentrations greater than CMC which overcompensates the interphase mass transfer reduction.

Optimized stacking, a new method for constructing ensemble surrogate models applied to DNAPL-contaminated aquifer remediation

Surfactant-enhanced aquifer remediation (SEAR) is an appropriate method for DNAPL-contaminated aquifer remediation; However, due to the high cost of the SEAR method, finding the optimal remediation scenario is usually essential. Embedding numerical simulation models of DNAPL remediation within the optimization routines are computationally expensive, and in this situation, using surrogate models instead of numerical models is a proper alternative. Ensemble methods are also utilized to enhance the accuracy of surrogate models, and in this study, the Stacking ensemble method was applied and compared with conventional methods. First, Six machine learning methods were used as surrogate models, and various feature scaling techniques were employed, and their impact on the models' performance was evaluated. Also, Bagging and Boosting homogeneous ensemble methods were used to improve the base models' accuracy. A total of six stand-alone surrogate models and 12 homogeneous ensemble models were used as the base input models of the Stacking ensemble model. Due to the large size of the Stacking model, Bayesian hyper-parameter optimization method was used to find its optimal hyper-parameters. The results showed that the Bayesian hyper-parameter optimization method had better performance than common methods such as random search and grid search. The artificial neural network model, whose input data was scaled by the power transformer method, had the best performance with a cross-validation RMSE of 0.065. The Boosting method increased the base models' accuracy more than other homogeneous methods, and the best Boosting model had a test RMSE of 0.039. The Stacking ensemble method significantly increased the base models' accuracy and performed better than other ensemble methods. The best ensemble surrogate model constructed with Stacking had a cross-validation RMSE of 0.016. Finally, a differential evolution optimization model was used by substituting the Stacking ensemble model with the numerical model, and the optimal remediation strategy was obtained at a total cost of $ 72,706.

Optimizing surfactant-enhanced aquifer remediation based on Gaussian process surrogate model in DNAPL-contaminated sites considering different wells patterns

Surfactant-enhanced aquifer remediation (SEAR) is an appropriate method for Dense non-aqueous phase liquids (DNAPLs) remediation. However, due to the high cost of chemicals used, choosing the suitable wells pattern and the optimal pumping scenario is necessary. In this study, the SEAR method performance for Regular (convergent) and Inverted (divergent) patterns with different wells numbers have been evaluated. The performance of 5 categories of patterns, including 35 different sub-patterns, was evaluated in a PCE-contaminated aquifer. The results show that the uniformity and appropriate surfactant distribution in the contaminated area significantly improves remediation performance. The distribution of surfactants in Regular patterns was better than Inverted patterns, and Regular patterns had lower remediation duration and cost. The best patterns that achieved a 95 % removal rate at the lowest cost were Regular. To find the optimal pumping scenario, a simulation-optimization model based on the Gaussian process regressor (GPR), as a surrogate model, has been used to reduce the optimization model's computational burden. Nine different kernels were applied and evaluated to find the best GPR. Also, the Bayesian hyperparameter optimization (BHO) method was used to optimize the surrogate model, and its performance was compared with the conventional grid search method. The results showed that the use of the Chi2 kernel and the BHO method are the best choices. A BHO-optimized multi-kernel Gaussian process (BHOMK-GP) model has also been developed, and its performance has been compared with single-kernel GPR surrogate models. The BHOMK-GP model's accuracy was significantly higher than single-kernel GPR models. The test and cross-validation RMSE of the BHOMK-GP model were 0.0385 and 0.0435, respectively. Finally, the optimal remediation scenario has been obtained by substituting the BHOMK-GP model as a surrogate model instead of the SEAR simulation model. The cost of remediation in the optimal strategy was $ 77,575.

Surfactant flooding makes a comeback: Results of a full-scale, field implementation to recover mobilized NAPL

Non-aqueous phase liquid (NAPL) remediation techniques using surfactants, such as enhanced pump and treat (also known as Surfactant–Enhanced Aquifer Remediation, “SEAR”) and micellar flooding provide a faster and more efficient way to recover NAPL from the subsurface. Micellar flooding is a recovery technique that relies on the ability of surfactants to mobilize the NAPL phase by reducing the interfacial tension between the aqueous phase and the NAPL. The application of micellar flooding for NAPL recovery has been limited to laboratory studies and some pilot–scale field applications primarily due to concerns that the technology might lead to uncontrolled movement of NAPL. This paper presents results from a full-scale field application of the micellar flood process designed to mobilize and recover an LNAPL (Jet fuel) from a surficial sandy aquifer located at a tank facility in western Jutland, Denmark. Phase behavior and flow experiments were conducted with field samples to identify suitable surfactant formulations. A field–scale simulation model was developed that indicated that a line–drive pattern with hydraulic control wells would be optimal for NAPL recovery. In addition to monitoring during the field implementation, monitoring was conducted immediately after and for a period of >1 year. The field implementation resulted in >90% recovery (approximately 36,000 Kg of LNAPL) based on the mass balance using laser–induced fluorescence (LIF) and chemical soil analysis (total petroleum hydrocarbon or TPH and BTEX) data. Post–surfactant flood site monitoring consisted of sampling water from multi–levels and from recovery wells. Groundwater samples were analyzed for total petroleum hydrocarbon (TPH) and benzene, toluene, ethylbenzene and xylene (BTEX). The pre–treatment and post–treatment mass discharges were also monitored, which led to a relationship between mass discharge with the mass reduction in the source zone. Also, the mass discharge Γ–model commonly used for DNAPL modeling was successfully implemented for LNAPL remediation. Studies of field applications of surfactant remediation processes are not readily available; it is especially rare to present a study of micellar flooding implementation for full-scale remediation processes.

Remediation of trapped DNAPL enhanced by SDS surfactant and silica nanoparticles in heterogeneous porous media: experimental data and empirical models.

The remediation of nonaqueous phase liquids (NAPLs) enhanced by surfactant and nanoparticles (NP) has been investigated in numerous studies. However, the role of NP-assisted surfactants in the dissolution process is still not well discussed. Besides, there is a lack of empirical dissolution models considering the effects of initial residual saturation Strap, NAPL distribution, and surfactant concentration in NAPL-aqueous phase systems. In this work, micromodel experiments are conducted to quantify mass transfer coefficients for different injected aqueous phases including deionized water, SDS surfactant solutions, and NP-assisted solutions with different levels of concentrations and flow rates. Observations reveal that silica nanoparticles (SNP) can significantly enhance interphase mass transfer, while SDS surfactant reduces the mass transfer coefficient. In addition, Strap and intrinsic interfacial area ai, as an indicator of dense nonaqueous phase liquids (DNAPL) distribution, influence the interphase mass transfer. The ai is also independent of DNAPL saturation SNAPL except for SNAPL < 7% when ganglia breakup occurs. Based on these observations, new empirical dissolution models are proposed in the presence and the absence of SDS surfactant and SNP in which ai, Strap, and surfactant concentrations are introduced as new parameters. The evaluated mass transfer rate coefficients using the proposed models show a significant improvement compared to available empirical models. The finding of this study might be attractive for application in field-scale simulations of surfactant-enhanced aquifer remediation (SEAR) and NP-assisted methods.

Pore scale study of amphiphilic fluids flow using the Lattice Boltzmann model

We study ternary amphiphilic fluid flow at pore scale using a “bottom-up” Lattice Boltzmann (LB) model. The model introduces a dipole to capture the amphiphilic structure of surfactant molecules, and characterizes microscopic fluid interactions at the kinetic level. We first verify the established model by three examples, i.e., interfacial tension test, phase separation, and droplet dynamics in shear flow. The numerical results show that surfactants reduce the interfacial tension (IFT), and are able to emulsify the amphiphilic fluids, and the contours of the droplets are quantitatively in agreement with the results of level set method. Then we study the droplet dynamics in a confined environment. The surfactants make the droplet easy to deform and break up in shear flow, but inhibit coalescence due to the Marangoni stress. Moreover, there are four typical patterns for droplet collisions, i.e. coalescence, breakup after coalescence, brushing past, and brushing back. Furthermore, we improve the model by considering the wettability reversal effect caused by surfactant adsorption onto walls, and find that surfactants can decrease the critical Bond number for oil detaching from walls by 90%. Finally, we simulate the surfactant-enhanced aquifer remediation (SEAR) or enhanced oil recovery (EOR) process in a 2D porous medium. Surfactants improve the oil recovery in all wetting conditions, but the surfactant loss due to adsorption onto walls diminishes the effect of IFT reduction.