Combining Soil Vapor Extraction and Electrokinetics for the Removal of Hexachlorocyclohexanes from Soil

Abstract

This paper focuses on the evaluation of the mobility of four hexachlorocyclohexane (HCH) isomers by soil vapor extraction (SVE) coupled with direct electrokinetic (EK) treatment without adding flushing fluids. SVE was found to be very efficient and remove nearly 70 % of the four HCH in the 15‐days of the tests. The application of electrokinetics produced the transport of HCH to the cathode by different electrochemical processes, which were satisfactorily modelled with a 1‐D transport equation. The increase in the electric field led to an increase in the transport of pollutants, although 15 days was found to be a very short time for an efficient transportation of the pollutants to the nearness of the cathode. Loss of water content in the vicinity of the cathode warns about the necessity of using electrokinetic flushing technologies instead of simple direct electrokinetics. Thus, results point out that direct electrokinetic treatment without adding flushing fluids produced low current intensities and ohmic heating that contributes negatively to the performance of the SVE process. No relevant differences were found among the removal of the four isomers, neither in SVE nor in EK processes.

Soil vapor extraction (SVE) and electrokinetic treatment (EK) were used to evaluate gaseous flux generation and mobility of hexachlorocyclohexane isomers. SVE removed almost 70 % of the isomers in 15 days. The EK treatment without the addition of washing fluids produced low current intensities and ohmic heating that contributed negatively to the performance of the SVE process.

Introduction

Development of efficient technologies for the remediation of soils polluted with hazardous species is becoming a priority for which a great effort is being made in the recent years. ^[1] A particular case of interest is the remediation of soils polluted with synthetic pesticides because of the great impact that this chemical may have on the environment. ^[2] In this context, hexachlorocyclohexane (HCH) is one of the organochlorine pesticides that has been widely used in recent decades, where widespread pollution has been reported in all spheres of the environment, such as in subsurface water, soil and air. Additionally, a large amount of waste was dumped around the world, especially in the places where it was produced. ^[3] Factors such as volatility, solubility and the long‐range atmospheric transport capacity of the HCH isomers further facilitate their entry into the most varied ecosystems. Thus, the correct management and remediation of areas contaminated by HCH represents an urgent demand. ^[4] Recent studies have demonstrated the suitability of electrochemically‐assisted technologies to remove this pollution, either using electrokinetic or soil washing processes, whose efficiency can be largely improved if properly combined with other non‐electrochemical treatments.[ ^1b , ⁵ ]

In electrokinetic treatment, the application of electric fields between anodes and cathodes placed into the polluted soil produced the transport of pollutants by different mechanisms, including dragging by the electro‐osmotic flux, electrophoresis (when flushing fluids containing surfactants are used) and electromigration (when the pollutants contains ionic groups). ^[6] Recently, several authors have mathematically modeled the electrokinetic removal of organic and inorganic species in polluted clay‐soils (kaolin), considering the chemical equilibria of species in the soil solution, the electrochemical reactions, and the transport due to the diffusion, electromigration and electroosmosis. ^[7]

However, it has been also reported that in going to large scale applications, the EK transport is not the most important process and ohmic heating may become the primary removal mechanisms, in which the organics are volatilized and then, they should be adequately managed. ^[8]

This opens a new category of treatment technologies: the combination of soil vapor extraction (SVE) with electrokinetics or what it can be called electrokinetic assisted soil vapor extraction EK‐SVE. In SVE injection of pressurized air in the subsurface of soil by injection point is used to drag volatile species to extraction point from which this pollution is channeled to a suitable treatment of the gas, including electrochemically assisted such as electro‐scrubbing. ^[9]

With this background, this work evaluates the combination of electrokinetic soil flushing (EKSF) with soil vapor extraction (SVE) for the removal of four hexachlorocyclohexane (HCH) isomers contained in a real matrix without adding flushing fluid. To shed light on the mechanisms involved in EK‐SVE process, the HCH mobility has been also modelized by a simple 1‐D model. To do this, it is simulated the pollution of soil with a real sample of loam taken from a highly polluted industrial site in a mixture or chlorinated organics, among which, we focus the attention in the more concentrated, which where four isomers of hexachlorocyclohexane: γ‐HCH; ϵ‐HCH; α‐HCH and δ‐HCH (samples also contained the β isomer but in negligible concentrations, as well as traces of other chlorinated hydrocarbons). The soil matrix prepared for testing was covered with a capillary barrier and two specially designed devices were manufactured to place electrodes and to inject and extract air from the soil, preventing its spreading to atmosphere. Electric gradients of 0.0, 1.0 and 3.0 V cm⁻¹ were applied and the evolution of the concentration of these species in the gas and soil was monitored for fifteen days, trying to shed light on the mechanisms of the processes occurring in soil.

Results and Discussion

Figure 1a shows the changes observed during fifteen days in the electric current and in the total electric charge passed in the three mockups evaluated in this work, in which the electric fields of 0.0, 1.0 and 3.0 V cm⁻¹ were applied. In each of these mockups, pressurized air was injected in a special cap‐device placed surrounding the anode and extracted with another special cap‐device placed surrounding the cathode. Both electrodes were buried directly in soil and no electrolyte wells were used. The bentonite barrier placed on the top of the mockup prevents uncontrolled release of gases to atmosphere. Because of the depth of the soil in which the anode and cathode are placed, it is expected that vapors flow mainly in the sand layer below the bentonite. During the tests, no water was added to the system. In Figure 1b, it is shown the average temperature and measurements of tensiometers placed in the zone close to the anode and the cathode.

Figure 1.

Changes in operation parameters during the SVE‐EK tests. a) Changes in the electric current (markers) and total electric charge passed through the mockup (lines): ▪ 0.0 V cm⁻¹ – dashed line; ▴ 1.0 V cm⁻¹ – point line; • 3.0 V cm⁻¹ – continuous line. b) Average values of the soil water suction and temperature. ○ Tensiometer between pollution source and cathode; ▪ tensiometer between pollution source and anode; × temperature (secondary y‐axis).

As shown in Figure 1a, in the two tests in which an electric field is applied, electric current is produced between the anode and the cathode. Values obtained are very low and decreases over time, down to a constant value, observation that matches with what it was observed in many other works in which electrokinetic technology is used and that it is typically explained in terms of the decrease in the water content of soil or alternatively in a worse distribution of water which affects to the distribution of the current lines. The electric current produced is not proportional to the electric gradient but conversely, it is lower for the higher electric gradient, what initially was unexpected. Similar results were obtained in a previous work of de Melo Henrique et al., ^[10] which investigated the remediation of soil spiked with lindane using air stripping associated and electrokinetic soil flushing applied different electric fields (0.75 V cm⁻¹ and 1.50 V cm⁻¹) during 14 days. In this previous work, it was observed that the accumulated charge was higher for the electric field of 0.75 V cm⁻¹. Thus, an explanation of this phenomenon can be derived from the values reported by the tensiometer (data shown in Figure 1b). As can be observed, the tensiometer placed next to the cathode shows important suctions values which correspond to important loses of water content in the soil in this region. These decreases are directly proportional on the magnitude of the electric gradient and this more irregular distribution of water content supports the lower extension of the electrokinetic processes. Regarding temperature, values increase with temperature, but they are very limited and differences between gradient magnitudes are almost negligible with an increase of only 0.52 °C/(V cm⁻¹). From the practical point of view this effect should have a very limited influence on the extension of volatilization of HCH.

Figure 2 shows the profiles of pH and conductivity in the soil after the fifteen days of treatment. As can be observed in, the largest changes in the pH were obtained in the test made at the highest current, which can be explained in terms of the higher production of protons and hydroxyls radicals in the nearness of the anode and the cathode. Important to consider that no electrolyte wells are used in this work, so this means that these changes are produced in the water surrounding the electrodes directly. Also, important differences were observed in the conductivity, although in this case these changes are caused by the transport of ions from the pollution source to the surroundings. In both cases, a very high dispersion in data is observed. Regarding possible reactions of hydro dehalogenation of HCHs, pHs are far from the alkaline values required and hence, they are not expected.

Figure 2.

Average values of pH (a) and conductivity (b) of the soil portions. ⧫ 0.0 V cm⁻¹ – dashed line; ▴ 1.0 V cm⁻¹ – point line; • 3.0 V cm⁻¹ – continuous line.

Figure 3 shows the changes in the concentration of γ‐HCH after fifteen days in 3‐D maps corresponding to the characterization described in the Experimental Section. The initial concentration of the source zone was 848 mg kg⁻¹ and, as observed, it distributes in a different way in the three mockups studied in this work (Figure 3). However, it can be seen that the mobility in the transverse directions (y and z axis) was almost negligible as compared with the longitudinal mobility in the x axis (consider the log scale), what means that the systems can be considered as a 1‐D to do further evaluations based on a maximum‐gradient modelling approach. ^[8b]

Figure 3.

3‐D scatter maps of γ‐HCH after fifteen days. Colors: Gray: 1–10 mg kg⁻¹; Blue: 10–100 mg kg⁻¹; Green: 100–1000 mg kg⁻¹; Yellow: 1000–10000 mg kg⁻¹.

Considering this simplification, the transport equation for each of the four HCH monitored may take the form shown in Equation (1), where C_i (x, t) is the concentration of the species i (mg species i kg⁻¹ dry soil) in a position x and time t, u_eff (m d⁻¹) is related to the dragging of the species i by different induced flows (including not only the hydraulic but also the electroosmotic, electromigration and electrophoretic fluxes) and D_eff (m² d⁻¹) is the effective diffusion/dispersion coefficient of this species under the experimental conditions, which may also be influenced by the application of electric fields because the simplifications carried out may include in this diffusive transport contribution from other transport processes not associated to dragging.

$${\displaystyle \frac{\partial C_i(x,t)}{\partial t}=D_{eff}\frac{{\partial }^2{C}_i(x,t)}{\partial x^2}-u_{eff}\frac{\partial C_i(x,t)}{\partial x}}$$ (1)

One solution that satisfies this equation is given by the gaussian curve shown in Equation (2), where m stands for the total amount of species i in the soil mockup (mg), S is the cross section area (m² ) and ρ _d is the dry density of soil (Kg dry soil m⁻³). ^[11]

$${\displaystyle C_i\left(x,t\right)=\frac{m}{{\rho }_{d}S\sqrt{4\pi D_{eff}t}}{e}^{-\frac{{\left(x-u_{eff}t\right)}^2}{4D_{eff}t}}}$$ (2)

Experimental data obtained after the 15 days of treatment were fitted to this equation and Figure 4 shows the influence of the applied electric gradient in the effective diffusion/dispersion coefficient. In fact, the fitting of the experimental data for the four isomers leads to very similar values of both parameters, which gives feedback about the robustness of the simplifications made. The electric field applied favors the dragging of the pollution in the direction anode to cathode (following the electro‐osmotic flow). As well the effective diffusion/dispersion coefficient increases almost linearly with the electric field (0.168 (cm² d⁻¹)/(V cm⁻¹)) from the value of the pure SVE system (0.2491 cm² d⁻¹), which means that electrokinetic processes are being developed in the system and have a very positive effect on the mobility of the HCH.

Figure 4.

Effect of the electric field applied on the model parameters: a) effective diffusion and b) effective velocity. □ α‐HCH ◊ ϵ‐HCH ▴ δ‐HCH • γ‐HCH.

However, this positive effect on the mobility in the anode‐cathode direction has negative implications in the in situ volatilization, as shown in Figure 5, where it is shown the amounts of the four HCH isomers volatilized after fifteen days, in which it can be clearly seem an almost linear decrease with the electric field applied. The onset of this figure shows the normalized volatilization rate, in which the data are normalized against the initial quantity of HCH in the soil and the treatment time. All points define a descending linear trend which indicates a decrease in the removal of 0.0086 mg HCH volatilized/(mg HCH contained ⋅ d)/(V cm⁻¹) from the 0.0467 mg HCH volatilized/(mg HCH contained ⋅ d)/(V cm⁻¹) obtained in the pure SVE process.

Figure 5.

Amounts of the four HCH isomers volatilized after fifteen days. □ α‐HCH ◊ ϵ‐HCH ▴ δ‐HCH • γ‐HCH.

As a consequence, and considering the low treatment times, the direct application of electrokinetics is negative from the view point the remediation of the soil as it is summarized in Figure 6, where it can be seen that after 15 days the SVE drags nearly 70 % of the HCHs initially contained in soil (very small differences among the four isomers, as it may be expected considering their very similar liquid‐vapor thermodynamic properties) while the amounts extracted are much lower when applying the electric field because of the higher transport of the pollutants which makes more inefficient the transport of the gases to the extraction point. The much lower rate of the electrokinetic process as compared to that of the SVE indicates that longer treatment times will not make the combined treatment more efficient because it can only be suggested an exhaustion of the HCHs from soil. Not adding water to the system has led to a very low electric current which also results in a very low ohmic heating of soil. In a previous work of Miller de Melo Henrique et al., ^[12] it was evaluated the combination of soil vapor extraction with electrokinetic soil flushing for the removal of hexachlorocyclohexane isomers. Results demonstrate that combination can efficiently exhaust the HCH isomers reaching a removal of more than 90 % after 15 days of treatment (20 % more than values attained by SVE). Another important aspect is the possibility of coupling EK with permeable reactive barriers such zerovalent iron or granular‐activated carbon (GAC) beds, in which it was also found the huge relevance of the volatilization processes. ^[13]

Figure 6.

Removal of HCH after 15 days of treatment. □ α‐HCH ◊ ϵ‐HCH ▴ δ‐HCH • γ‐HCH.

The 1.5 °C increase in temperature was irrelevant to enhance the removal of the pollutants and the loss of water content in regions of the soil because of the transport of water to the cathode during the treatment was very negative and prevents a better performance of the electrochemically assisted SVE. Hence, these data confirm that EK transport should be promoted with the addition of flushing fluids that prevent the desiccation in the direction anode‐cathode and higher currents that make the temperature increase with higher ohmic loses.

Conclusions

From this work it can be concluded that SVE can efficiently remove the HCH isomers through volatilization, reaching a removal of 70 % after 15 days of treatment. Additionally, the application of electric fields contributed to a higher mobility of the HCH, mainly in the anode‐cathode direction. Here the experimental set‐up was large enough to demonstrate the higher resistance resulting from an increase in the distance between the electrodes and the absence of an electrolyte. The unsuccessful coupling of electrokinetics with SVE can be ascribed to changes in the resistance because of an increase in the temperature, due to higher ohmic losses. In general, no relevant differences were observed between the performance of the technologies with applied the four HCH isomers

Experimental Section

Chemicals. Hexane (Sigma‐Aldrich) and ethyl acetate (Scharlau) were of analytical grade and used as received. Two types of soil with low hydraulic conductivity were used: real clay sample taken from a highly polluted industrial site in a mixture of chlorinated organics (Figure S1 in Supporting Information, location 1), among which, we focused our attention on the most concentrated, where there are four hexachlorocyclohexane isomers: γ‐ HCH; ϵ‐HCH; α‐HCH and δ‐HCH; and natural soil free from contamination, collected in a quarry in the vicinity of Toledo/Spain (Figure S1, location 2). The main physicochemical properties are described in Table 1. The first was used as the contamination source while the second as the recipient of this contamination to evaluate the influence of the transport processes. This means that this second soil is the important matrix from the viewpoint of pollutant mobility and according to the United States Department of Agriculture (USDA), it was classified as Silty Loam.

Table 1.

Properties of soil.

Properties	Value
Mineralogical composition [%]
Calcite	4.0
Feldspar	15.0
Illite	20.0
Kaolinite	26.0
Smectite	28.0
Quartz	7.0
Particle size distribution [%]
Clay	4.9
Sand	26.9
Silt	68.2
Physicochemical properties
pH	6.49
Electrical conductivity [μS cm−1]	84.2
Dry density [g cm−3]	1.4
Water content [%]	20.0

Experimental design. Three different tests were performed: 1) reference test to verify the natural dispersion of contaminants; 2) electrokinetic remediation tests with electrodes spiked in the soil, subjected to an electric field of 1 V cm⁻¹; and 3) electrokinetic remediation tests with electrodes spiked in the ground, subjected to an electric field of 3 V cm⁻¹. The experimental installation consisted of methacrylate reactors with a total capacity of 78.2 dm³, tensiometers, thermometers, dataloggers for data storage and power supply. To simulate a soil vapor extraction process, blowers were used to vent the atmosphere from the wells at predetermined times. A two‐bottle hexane trap was placed in series with the mockup to capture the volatile species generated during the treatment. The total soil volume place in each mockup was approximately 50 dm³. The soil preparation process in the reactor consisted of: 1) placement of a layer of gravel for mechanical support and drainage; 2) distribution of three soil layers of 4 cm thick that, when placed individually, were compacted to a dry density around 1.4 g cm⁻³ with a water content of 20 %, as described in previous works. ^[14] The layers were scarified to ensure continuity and prevent the development of preferential flow paths during the operation; 3) perforation of the compacted soil to place the electrodes, the contaminated soil in the center of the model (10×10×10 cm) and the instrumentation (tensiometers and thermocouples); 4) sealing (to collect the contaminant generated from the evaporation flows) and instrumentation of the wells for: air inlet and outlet to collect the gaseous flows, in addition to the electrode conductor wire; 5) placement and compaction of a 4 cm thick layer of sand, 6) placement a 4 cm thick layer of bentonite sludge (preventing uncontrolled loss of vapor to the atmosphere), 7) placement and compaction of a 4 cm thick top layer of sand (which acts as a capillary barrier to reduce evaporation of water from the bentonite, thus protecting bentonite from cracking). Between the layers, geotextiles were placed for reinforcement, separation, filtration, and drainage purposes. The electrodes used were graphite rods with dimensions 2.5×2.5×17.0 cm. The interelectrode gap was 30 cm. The contaminated sample was located exactly in the middle of the mockup, equidistant from the electrodes. Electric current was applied by power supplies (MPL‐3505M, Minipa, 400 SM‐8‐AR ELEKTRONIKA DELTA BV) operating in potentiostatic working mode for 15 days. At predetermined times, blowers were used to vent the atmosphere from the wells for a period of 10 min. The air flow was directed to the hexane trap, composed of two bottles in series with hexane, to collect and quantify the contaminants generated by evaporation.

To characterize the distribution of pollution in the direction anode‐cathode, the soil was submitted to a postmortem analysis, being sampled at six points in the longitudinal axis (x) and at three points in the transversal axis (y), totalizing 18 samples. In addition, each sample was divided in vertical into three portions (z), as shown in Figure 7. Different parameters were measured including room temperature, electric current, pH, conductivity, soil water suction, soil temperature, the concentration of HCHs and intermediates.

Figure 7.

Experimental setup and 2D‐sampling map.

Analytical procedures. HCHs and derivatives were identified and quantified via gas chromatography with an electron capture detector (GC‐ECD) (Thermo Fisher Scientific) using a TG‐5MS capillary column (30 m×0.25 mm×0.25 mm) and an electron detector ⁶³Ni micro electron‐capture, a split/splitless injector, and ChromCard software. The flow rate of the He gas was 1.0 mL min⁻¹. The injector temperature was 210 °C.

In the post‐mortem soil samples, a Solid‐Liquid (S‐L) extraction with ethyl acetate in a ratio of 1 g of soil/4 mL of solvent was carried out in 15 mL flasks. Subsequently, the contents were vigorously stirred for 5 min in a vortex mixer, sonicated for 10 min in ultrasound (JP Selecta, Barcelona), and centrifuged at 3500 rpm for 10 min. Then, the supernatant was collected and analyzed on the GC‐ECD. The hexane from the trap was collected at predetermined times and analyzed in the GC‐ECD to monitor the evaporated contaminant. Analysis of pH (pHmeter Crison GLP 22) and conductivity (Conductivity meter Crison Ecmeter Basic 30+) of the soil samples were carried out using the standard method EPA 9045 C, in which 10 g of soil was mixed with 25 mL of deionized water (Millipore system Milli‐Q, 18.2 MΩ cm, 25 °C). It was stirred for 10 min in a vortex mixer and, after sedimentation, the aqueous phase was collected for analysis. Water content and soil temperature were measured continuously (more details of the arrangement of tensiometers and thermometers can be found in Figure S2 and Table S1). Water content measurements were obtained by analyzing the soil water suction, whose values were determined by a set of model T5 tensiometers (UMS GmbH, Munich, Germany) inserted into the soil, connected to a model DL6 datalogger (UMS GmbH, Munich, Germany) that stored the data. The temperature was obtained through ECT model thermocouples (Decagon Devices, Pullman, USA) embedded in the soil. Data collection was also performed by connecting these sensors to dataloggers.

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Henrique J. M. M., Isidro J., Saez C., Lopez-Vizcaíno R., Yustres A., Navarro V., Dos Santos E. V., Rodrigo M. A., ChemistryOpen 2023, 12, e202200022.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.