lable at ScienceDirect

Journal of Environmental Management 153 (2015) 40e47

Contents lists avai

Journal of Environmental Management

journal homepage: www.elsevier .com/locate/ jenvman

Combination of surfactant enhanced soil washing and electro-Fentonprocess for the treatment of soils contaminated by petroleumhydrocarbons

David Huguenot*, Emmanuel Mousset, Eric D. van Hullebusch, Mehmet A. OturanUniversit�e Paris-Est, Laboratoire G�eomat�eriaux et Environnement (EA 4508), UPEM, 77454, Marne-la-Vall�ee, France

a r t i c l e i n f o

Article history:Received 1 July 2014Received in revised form21 January 2015Accepted 23 January 2015Available online

Keywords:Soil pollutionSoil washingColumn experimentTween® 80Electro-Fenton

* Corresponding author.E-mail address: david.huguenot@u-pem.fr (D. Hug

http://dx.doi.org/10.1016/j.jenvman.2015.01.0370301-4797/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

In order to improve the efficiency of soil washing treatment of hydrocarbon contaminated soils, aninnovative combination of this soil treatment technique with an electrochemical advanced oxidationprocess (i.e. electro-Fenton (EF)) has been proposed. An ex situ soil column washing experiment wasperformed on a genuinely diesel-contaminated soil. The washing solution was enriched with surfactantTween® 80 at different concentrations, higher than the critical micellar concentration (CMC). The impactof soil washing was evaluated on the hydrocarbons concentration in the leachates collected at thebottom of the soil columns. These eluates were then studied for their degradation potential by EFtreatment. Results showed that a concentration of 5% of Tween® 80 was required to enhance hydro-carbons extraction from the soil. Even with this Tween® 80 concentration, the efficiency of the treatmentremained very low (only 1% after 24 h of washing). Electrochemical treatments performed thereafterwith EF on the collected eluates revealed that the quasi-complete mineralization (>99.5%) of the hy-drocarbons was achieved within 32 h according to a linear kinetic trend. Toxicity was higher than in theinitial solution and reached 95% of inhibition of Vibrio fischeri bacteria measured by Microtox® method,demonstrating the presence of remaining toxic compounds even after the complete degradation. Finally,the biodegradability (BOD5/COD ratio) reached a maximum of 20% after 20 h of EF treatment, which isnot enough to implement a combined treatment with a biological treatment process.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Soil pollution by hydrocarbons is a worldwide environmentalconcern. For instance, in France, these pollutants are present inabout 32% of the polluted sites referenced in the BASOL database(BASOL, 2015). Among this class of pollutants, 23% of the sites arecontaminated by total petroleum hydrocarbons (TPH). Thesecompounds are often associated with fuel, oils, jet fuel, etc. Theirimpact on human health and environment is well known becauseof their hydrophobic characteristics (Lapinskiene et al., 2006)permitting them to reach and accumulate in the food chain. Theyare indeed likely to cause toxic effects to human and environmentalreceptors (Rowland et al., 2001). Furthermore this toxicity is closelyrelated to their structures; the light fractions being less toxic thanheavier ones (van Gestel et al., 2001). Their physical and chemical

uenot).

properties such as solubility and Kow favour their accumulation inorganic matter and human bodies via food chain. The fate of thesemolecules in the environment is often leaching in soils togroundwater, dispersion, and mostly sorption and biodegradation(Cozzarelli et al., 2014). Due to their harmful effects, TPH have beenregulated in France where a concentration of 500 mg kg�1 of drymatter is the legal threshold, above which soil is considered as awaste and has to be treated (JORF, 2010).

Biological, physical and chemical in situ treatment methods areoften used but these techniques are very often time-consuming(Tang, 2005) and require high engineering costs (Majone et al.,2015). Besides, before implementing an in situ soil treatmenttechnique at full-scale, laboratory tests should be performed inorder to adapt the technique to the field conditions (Gomes et al.,2013). As a consequence, ex situ techniques such as soil washingare getting more and more interest despite that soil excavation isnecessary (Lee et al., 2005; Khalladi et al., 2009). Since 2007, Frenchpolicy on polluted sites is strongly oriented towards the use andoperation of in situ treatment methods. In situ treatment of TPH

Table 1Hydrocarbons concentrations in mg kg�1 DW (dry weight). S1, S2 and S3 stand forthe three different soils. Fractions C10 to C40 stand for the number of carbon atomsin the molecule. Total means the sum of the 4 different fractions.

Hydrocarbons concentrations (mg kg�1 DW)

C10eC12 C12eC16 C16eC21 C21eC40 Total

S1 700 2400 2300 660 6100S2 320 1600 1600 480 3900S3 650 2300 2500 1000 6100

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e47 41

contaminated soils could be achieved by soil flushing (Lee et al.,2005). However, this remediation approach is still being devel-oped by companies as it does not require soils excavation, allowingthis technique still to be more cost-effective than ex situ processes.Surfactants are chemical compounds frequently used for theextraction of hydrocarbons from soil. Their amphiphilic propertiesare useful to promote the mobilization of hydrophobic compoundssorbed onto soil particles. Non-ionic surfactants like Tween® 80 areparticularly of interest when dealing with hydrocarbons. Since itsfirst use more than 20 years ago (Laha and Luthy, 1992), Tween® 80has been widely used in soil remediation to mainly clean up hy-drocarbons (Wong et al., 2004; Mousset et al., 2013a). Compared toother surfactants, its chemical characteristics together with its lowcost, and low toxicity (Varma et al., 1985) on soil microorganisms(Bautista et al., 2009) compared to other surfactants, are of greatinterest for soil remediation companies.

Although this washing technique is generally efficient to cleansoil, the major concern remains in the treatment of the leachatescontaining both TPH and surfactants. Such solutions contain asignificant amount of high COD (Chemical Oxygen Demand), whichrequire an advanced oxidation treatment to be degraded. Electro-chemical Advanced Oxidation Processes (EAOPs) have shownpromising results to treat many poorly biodegradable organiccompounds in solutions (Brillas et al., 2009; Panizza and Cerisola,2009) even for highly loaded solutions like reverse osmosis con-centrates and landfill leachates (Zhou et al., 2012). These tech-niques, considered as environmental friendly, promote the in situelectro-generation of hydroxyl radical (�OH) that is a very power-ful oxidizing agent (E� (�OH/H2O) ¼ 2.80 V/SHE (StandardHydrogen Electrode). Among these EAOPs, electro-Fenton (EF)technique has shown good performances towards various organicpollutants (Dirany et al., 2012; El-Ghenymy et al., 2013; Loaiza-Ambuludi et al., 2013; Oturan et al., 2012) and more especiallywith synthetic soil washing solutions (Mousset et al., 2014a). EFallows the production of �OH through the Fenton reaction (Eq. (1)):

Fe2þ þ H2O2 þ Hþ / Fe3þ þ H2O þ �OH (1)

Compared to chemical Fenton process, the EF process allowsminimizing the use of reagent since H2O2 is insitu produced and acatalytic amount of soluble iron is enough because it is continu-ously electro-regenerated at the cathode via the reactions 2 and 3(Sires et al., 2007):

O2 þ 2Hþ þ 2ee / H2O2 (2)

Fe3þ þ ee / Fe2þ (3)

Thanks to these enhancements, higher degradation rate, highmineralization degree of organic pollutants and no sludge pro-duction are observed (Rodrigo et al., 2014).

This study aims at implementing an innovative combination ofsoil column washing with Tween® 80 and EF treatment of thecollected leachates. To the best of our knowledge, this is the firsttime that such a combination is applied on a diesel-contaminatedsoil. Some studies dealing with electrocoagulation (L�opez-Vizcaíno et al., 2012) and Fenton oxidation (Rosas et al., 2013)have been performed but never used soil columnwashing togetherwith EF. Iglesias et al. (2014) performed an EF treatment on marinesediment contaminated with phenanthrene considering the solu-tion and the slurry after the sediment washing. Another studyperformed column soil washing in the presence of Tween® 80 andphenanthrene combined with an electrochemical treatment(G�omez et al., 2010). However, this electrochemical treatment is notconsidered as an EAOP, since it is not based on the oxidation with

�OH. In this paper, the washing of TPH-contaminated soil withTween® 80 has been studied using a soil column experimental setup. The experimental configuration is close to what could be usedon the site where the contaminated soil was sampled. The collectedleachates containing TPH were then studied for their degradationfeatures with EF as an advanced electrochemical technique. Finally,the biodegradability of the treated leachates was also monitored.

2. Materials and methods

2.1. Surfactant

Polyoxyethylene (20) sorbitan monooleate, also known asTween® 80, was purchased from SigmaeAldrich (France) and wasused for the soil washing experiments. This non-ionic surfactanthas a molecular mass of 1310 g mol�1 and a critical micellar con-centration (CMC) of 0.012 mM (0.0016% w/v) at 25 �C. Distilledwater was used to prepare solutions containing different concen-trations of Tween® 80.

2.2. Soil

Soil samples were collected at the depth of 100 cm in an urbansite where a genuine contamination by TPH was discovered. Thesamples were collected on July 12, 2012 and were stored in sealedbuckets protected from sunlight until the column experiments. Thesoil collected on that site was characterized as a sandy loam soilwith more than 60% sand, 25% loam and 15% clay. The main soilcharacteristics are as follows: pH(H2O) 8.4; organic matter content,44.6 g kg�1 DW; cationic exchange capacity, 15.7 cmol kg�1 DW.Three different samples were used for the experiments and werenamed soil S1, soil S2 and soil S3. TPH concentrations are shown inTable 1.

2.3. Materials

Polyvinyl chloride (PVC) columns of 30 cm height and 20 cmdiameter were used. These PVC columns were put on a polymethylmethacrylate (PMMA) (Plexiglas®) slab of 0.8 cm width. This slabwas pierced by 2 mm diameter holes in order to evacuate theleachates. Eluates were then collected in a high-density poly-ethylene (HDPE) funnel.

2.4. Washing experiment

Columns were filled with 15 kg of contaminated soil (S1, S2 andS3) without any preparation. Experiments were performed at roomtemperature (approx. 20 �C). Washing solutions were stored in aplastic tank during the experiment. The required volumes wereinjected by a 4-channels peristaltic pump IPC-N (Ismatec,Switzerland). Tygon® tubes with an internal diameter of 0.76 mmwere used.

Washing solutionwas preparedwith Tween® 80 dissolved in tapwater. Tween® 80 solutionwas prepared at different concentrations

Fig. 1. Measured concentrations (log mg L�1) of total fraction of Total Petroleum Hydrocarbons in collected eluates during time and considering steps 2 and 3. Tween® 80 con-centrations are expressed as percentages. n.d.: not determined. bql: below quantification limit (20 mg L�1 for total TPH).

Table 2Calculated cumulative amounts (mg) of total and each fraction of Total PetroleumHydrocarbons in collected eluates during time and considering steps 2 and 3. Cu-mulative data are expressed per extraction step. Tween® 80 concentrations areexpressed as percentages.

Cumulative amounts (mg)

C10eC12 C12eC16 C16eC21 C21eC40 Total

Tw0.01% Step 2 17 28 26 38 110Step 3 22 29 26 403 480

Tw0.1% Step 2 20 137 293 39 362Step 3 4 51 99 71 224

Tw1% Step 2 524 3481 3712 1601 9283Step 3 636 3825 3630 1477 9747

Tw5% Step 2 36,169 192,905 184,868 219,697 630,960Step 3 17,440 100,313 94,821 95,129 305,303

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e4742

(Tw0.01%, Tw0.1%, Tw1% and Tw5%). Each concentration wasapplied on separated columns. Tw0.01% and Tw0.1% solution wereused on soil S1, Tw1% solution was used on soil S2 and Tw5% so-lution was used on soil S3.

The elution procedure was implemented as follow:

- Step 1: 24 h of a tap water washing in order to reach pore watersaturation. Flow rate was set at 3 mL min�1.

- Step 2: 24 h of a Tween® 80 solution at the rate of 1 mL min�1

- Step 3: 24 h of a tap water washing, considering as a rinse, at therate of 3 mL min�1.

After elution throughout the column, eluates were collected inglass flasks at different times after the beginning of each step:

- T0þ 3 h stands for the volumes collected for 3 h between theinitial time (T0) of each step and 3 h after;

- T0þ16 h stands for the volumes collected for 13 h between 3 hand 16 h after the initial time;

- T0þ 24 h stands for the volumes collected for 8 h between 16 hand 24 h after the initial time.

Collection flasks were weighed to measure the volume and thenstored in glass flasks at 4 �C prior to analysis. Only eluates from step2 and step 3 were analyzed for TPH concentration.

2.5. Total petroleum hydrocarbons analysis

TPH analyses in eluates were performed with a GC-FID by acertified lab (AlControl, the Netherlands) according to an internallydeveloped analytical method based on the standard procedure ISO11046 with C10 and C40 compounds as internal standards. Prior toanalysis, liquid samples were stored in amber flasks of 100 mL with0.5 mL of concentrated H2SO4. The limit of quantification was of5 mg L�1 for each fraction and of 20 mg L�1 for the TPH concentration.

2.6. Electrochemical treatment of eluates

2.6.1. ChemicalsSodium sulfate, 2-(P-toluidino)naphthalene-6-sulfonic acid so-

dium (TNS) and Tween® 80 (polyoxyethylene (20) sorbitan mon-ooleate) were purchased from Aldrich. Potassium dihydrogenphosphate (KH2PO4) and Mohr's salt were supplied by Acros atanalytical grade. N-Allylthiourea (98%) was provided by Alfa Aesar.

Sodium phosphate dibasic (Na2HPO4), ammonium chloride(NH4Cl), heptahydrated magnesium sulfate (MgSO4$7H2O), dehy-drated calcium chloride (CaCl2$2H2O), D(þ)-Glucose$H2O andphenanthroline 1,10 were purchased from Merck at analyticalgrade. Analytical reagents like dipotassium hydrogenophosphate(K2HPO4), hexahydrated ferric chloride (FeCl3$6H2O) and NaOHwere provided by VWR. Potassium chloride (KCl) (>99.0%, Fluka)was also used. In all experiments, ultrapure water from a MilliporeSimplicity 185 (resistivity > 18 MU cm) system was used.

2.6.2. Electro-Fenton treatmentsEF experiments were performed in a 0.40 L undivided glass

electrochemical reactor at current controlled conditions with apower supply HAMEG 7042-5 (Germany) set at 1000 mA, since itwas assumed as the best compromise in a former study (Moussetet al., 2013a). The studied solution derives from the soil washingexperiment performed with the highest concentration of Tween®

80 (5%) at time Tþ 24 h (COD ¼ 22,000 mg O2 L�1;TOC¼ 5100mg C L�1). The cathodewas a 150 cm2 carbon-felt piece(Carbone-Lorraine, France) and the anode was a Boron-DopedDiamond (BDD) plate (5 � 4 cm), which is centred in the cell andsurrounded by cathode covering the inner wall of the cell. Anelectrolyte (Na2SO4 at 0.15 mM) was added to the medium, as theinitial conductivity of solutions was too low (1.39mS cm�1). The pHof initial solution was not adjusted to the optimal pH of 3 as usual(Mousset et al., 2013a). Further details regarding temperature, O2bubbling and stirring can be found in a previous study (Moussetet al., 2014a).

Fig. 2. Role of each washing steps regarding the elution of hydrocarbons. Percentages are expressed considering the total amount leached. Step 2: tap water only, 3 mL min�1. Step3: Tween® 80 (at the appropriate concentration), 1 mL min�1.

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e47 43

2.6.3. BioassaysThe biodegradability is given by the ratio between BOD5 and

COD. BOD5 is determined by respirometric method (OECD 301 F,1993) with the OxiTop® control system (WTW). The method fol-lows the procedure detailed by Mousset et al. (2014b). COD mea-surements are accomplished by colorimetric method (Moussetet al., 2014b).

Toxicity assays were performed with a Berthold Autolumat PlusLB 953 equipment by using Microtox® standard method (ISO11348-3) based on 15 min of exposure of marine bacteria Vibriofischeri (LUMIStock LCK-487; Hach Lange) (Mousset et al., 2014b).

2.6.4. Analytical determinationsThe TOC decay was determined by catalytic oxidation using a

Shimadzu VCSH TOC analyser.The pH was monitored during all the treatment with a com-

bined glass electrode (WTW, Germany) calibrated with standardbuffers of pH 4.0 (±0.1) and 7.0 (±0.1).

Fig. 3. Total Organic Carbon (TOC) evolution during EF mineralization of column soil washinBDD anode.

The total iron (ferrous and ferric) concentration dissolved insolution was determined by colorimetry (510 nm) with 1,10-Phenanthroline as indicator (Rodier et al., 2009). After calculation,the total concentration of iron was 0.06 (±0.01) mM, which wasenough to ensure the EF process to proceed, since only a catalyticamount is needed (Eq. (3)) (around 0.1 mM) (Brillas et al., 2009;Mousset et al., 2014a).

3. Results and discussion

3.1. Effect of soil washing with Tween® 80 on hydrocarbon ineluates

Similar amounts of eluates were collected from the differentTween® 80 treatments (data not shown). Additionally, preliminaryTPH analyses were performed on eluates from step 1 (i.e. washingonly with water). Those analyses revealed an average total hydro-carbons concentration of 87 ± 10 mg L�1 in the total volume eluted

g solution. [Fe]total dissolved ¼ 0.06 mM, I ¼ 1000 mA, [Na2SO4] ¼ 150 mM, V ¼ 400 mL,

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e4744

at the end of step 1. This result shows that hydrocarbon desorptionwith water is possible, even without the use of surfactants; thiscould be ascribed to the presence of easily accessible hydrocarbonsthat could be mobilised by water.

Hydrocarbons concentrations in soil eluates collected duringstep 2 and step 3 are displayed in Fig. 1. No occurrence of prefer-ential flux was noticed since there was no correlation betweendischarged volumes and hydrocarbons concentrations. Datashowed an effect of the washing for all studied Tween® 80 con-centrations, especially for Tw1% and Tw5% treatments. These con-centrations of Tween® 80 appear to be much more efficientcompared to Tw0.01% and Tw0.1%. Maximum concentrations ofhydrocarbon of 25 mg L�1 and of 920 mg L�1 were reached forTw1% and for Tw5%, respectively. These were lower than what wasobtained in a previous study on an urban contaminated soil wherethe maximum concentration of diesel exported by a solution of 2%of sorbitan monooleate was 2500 mg L�1 (Lee et al., 2005). How-ever, our result with 5% of Tween® 80 is much higher than the onereported by Khalladi et al. (2009) with 8 mM Sodium DodecylSulfate (SDS) in a diesel spiked soil (TPH soil concentration was 10times higher than in the present study). The increase in TPHdesorption seen in this study with increasing Tween® 80 isconsistent with Mousset et al. (2013a).

Regarding organoleptic characteristics, it is important to noticethat some eluates of the Tw5% treatment, with TPH concentrationshigher than 54 mg L�1, showed iridescence features at the surfaceand the presence of a strong fuel smell in these washing solutions.

3.2. Role of Tween® 80 in hydrocarbons extraction and impact onthe different fractions

The calculated cumulative amounts of extracted hydrocarbonsper step are shown in Table 2. Considering these results, the gap

Fig. 4. Reaction sequence of alkane compounds oxidation by �OH.

between Tw5% and the three others treatments is very important.Total extracted amounts of TPH with 5% of Tween® 80 reached936mgwhichwas 49-, 1600- and 1586-times higher than for Tw1%(19 mg), Tw0.1% (0.59 mg) and Tw0.01% (0.59 mg), respectively. ForTw5% the maximum removal rate represents only 1% of the totalamount of hydrocarbons in the soil column. These performanceswere lower than reported by Mousset et al. (2013a) with a PAH-contaminated soil where 49% of phenanthrene was extracted af-ter only onewashing step with a solution containing 1% Tween® 80.

From the previous results, it could be concluded that a solutionwith 5% of Tween® 80 is more efficient than with 1% Tween® 80.Furthermore, from a cost balance assessment, it is preferable to usea solution that is 5-time more concentrated and is finally 49-timemore efficient. However, its efficiency per se has to be carefullyconsidered since 99% of hydrocarbons remains in the soil.

Table 2 also shows the effect of the treatment on the differenthydrocarbons fractions. It is interesting to notice that for Tw5%, theheaviest fraction (C21eC40), which is representative for heavy fueland lubricant oils. This was not observed for the three othertreatments. As a consequence, a correlation between Tween® 80concentration and the number of carbon atoms in hydrocarboncould be drawn: the amount of hydrocarbons from the fraction C21to C40 leached is greater when Tween® 80 concentration is high.One of the possible explanations may be due to the formation ofmicelles in presence of very high Tween® 80 concentration. Thesemicelles would be able to gather and then be responsible for thesorption of the high carbon content chains. Considering thesefractions, it is therefore important to notice here that Tween® 80can not be confused as a petroleum hydrocarbon in the eluatescollected in this study during the GC-FID analysis. Given its highcarbon atoms amount, Tween® 80 is rather difficult to degradebiologically. Besides, it has already been shown that when mole-cules are heavy they are less biodegradable (Leahy and Colwell,1990; Singh et al., 2012). In this study, the steps 2 and 3 are notlong enough to allow the appearance of biodegradation pathwaysof Tween® 80.

To decipher the respective role of each step, the amount of hy-drocarbons leached after steps 2 and 3 are presented in the Fig. 2.For treatments with Tw0.1% and Tw5% the step 2 led to a higherelution of hydrocarbons whereas for Tw1%, there was no cleardifference between the two steps. These results indicate that step 3,consisting of an elution only with water, played a role in the re-covery of previously desorbed hydrocarbons by Tween® 80 duringthe step 2.

3.3. Mineralization of soil washing solution by EF

The EF treatment succeeds to achieve quasi-complete mineral-ization (>99.5% TOC removal) after 32 h of treatment, knowing thatthe initial load of solutions was about 22,000 mg O2 L�1 (COD). Interms of energy consumption calculated according to Eq. (4) (Brillaset al., 2009), it leads to 0.1 kWh per g TOC consumed after quasi-complete mineralization.

Energy consumption�kWh ðg TOCÞ�1

�¼ EcellIt

ðDTOCÞtVS(4)

where Ecell is the average cell voltage (V), I is the applied current (A),t is the electrolysis time (h), VS is the solution volume (L) and(DTOC)t is the TOC decay (mg C L�1) at time t in h.

These solutions are mostly composed of Tween® 80 compound,which is in competition with hydrocarbons during the oxidationwith hydroxyl radicals generated by the process. A linear trend(R2 ¼ 0.995) is obtained until 24 h of treatment. Between 24 h and32 h of treatment the curve trend becomes exponential (Fig. 3). This

Fig. 5. Biodegradability assessment (BOD5/COD) during EF mineralization of column soil washing solution. [Fe]total dissolved ¼ 0.06 mM, I ¼ 1000 mA, [Na2SO4] ¼ 150 mM,V ¼ 400 mL, BDD anode. (a) COD decay, BOD5 evolution and biodegradability, (b) BOD of each sample taken after 0, 4, 8, 12, 16, 20 and 24 h of EF treatment.

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e47 45

behaviour has been already observed in anodic oxidation (Panizzaet al., 2001a, 2001b) and in EF (Mousset et al., 2013b). It meansthat the electrolysis is under current control until 24 h of treatment(applied current density (jappl) is below limiting current density(jlim)). It generally happens when the organic matter initial load ofthe solution is high and jappl is low (50 mA cm�2 here). After 24 h oftreatment, the oxidation at BDD anode surface is under masstransfer condition, and a first order reaction can occur.

A TOC decrease of 400 mg C L�1 was observed in 2 h of EFtreatment compared to a decrease of 100 mg C L�1 of soil washingsolutions containing Tween® 80 and p-cresol and treated withchemical Fenton process in another study (Rosas et al., 2013). Itconfirms that EF has better performance than the Fenton treatment(Canizares et al., 2009).

Regarding the pH evolution, the initial value of 7.4 decreasedquickly until an average value of 2 after only 2 h of treatment. Thisindicates the formation of carboxylic acids that are formed during

the oxidation of organic compounds (€Ozcan et al., 2013) throughthe suggested reaction sequence of alkane compounds oxidation by�OH represented in Fig. 4.

The dehydrogenation of saturated compounds by �OH has beenalready reported (Brillas et al., 2009; Mousset et al., 2014b) and canoccur also with Tween® 80 that is partly composed of an alkanechain. However, the surfactant contains also two double bonds thatwill be firstly hydroxylated by �OH, since these radicals are knownto react quickly with unsaturated bonds (Brillas et al., 2009)through addition reactions leading to hydroxylation of the organiccompounds.

A value of pH 2 is good and better than the initial value tooperate EF, since the optimal conditions of EF are at about pH 3,however the process can occurs at 2 < pH < 4 without significantloss of efficiency. Indeed, a pH value higher than 3.5 favours Fe3þ

precipitation as Fe(OH)3 and a pH value below 2 supports the for-mation of complex between Fe2þ and H2O2 (Brillas et al., 2009). It

Fig. 6. Biodegradability enhancement factor evolution during EF mineralization of column soil washing solution. [Fe]total dissolved ¼ 0.06 mM, I ¼ 1000 mA, [Na2SO4] ¼ 150 mM,V ¼ 400 mL, BDD anode.

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e4746

means that no pH adjustment (no reagent addition) is needed atthe beginning of the treatment. This has been observed in a pre-vious study reporting soil washing of PAHs-contaminated soilscombined with EF (Mousset et al., 2013a).

3.4. Toxicity and biodegradability assays

Evolution of the toxicity was monitored by Microtox® methodbased on the inhibition of the luminescence of marine bacteria V.fischeri. Toxicity level reached rapidly 90% inhibition of V. fischeriluminescence after 2 h of treatment and then stabilized about 95%inhibition until the completemineralization (data not shown). Evenafter the quasi-complete treatment of the solutions, some toxicityfeatures are still present, probably because of the presence ofmineral toxic compounds, extracted during the flushing experi-ment, which remained in the solution during the EF (Mousset et al.,2013a). Heavy metals such Cr, Cu, Ni, Pb and Zn were analysed byICP-OES in order to check their initial concentrations in washingsolution but remained not detectable (data not shown).

Chemical and biological oxygen demands are presented inFig. 5a. The COD is decreasing linearly with the time until 24 h of EFtreatment, as observed with the TOC removal. The obtained resultsare quite satisfactory as biodegradability increases during 4 h of EFtreatment. However, after 20 h of treatment, it reaches a maximumof 20%, which is too low to start a biological treatment. Indeed, thethreshold value to combine EF and biological treatment is assumedto be acceptable when higher than 33% (Rodier et al., 2009). Asalready stated before, this could be the consequence of the pres-ence of inhibitory substances. This is corroborated by the presenceof latent phases in the first hour of BOD measurement as shown inFig. 5b. This is particularly observed for the samples analysed after8 h and 24 h of EF treatment.

Fig. 6 depicts the biodegradability enhancement (Ebiodeg) factorcalculated according to Eq. 14 (Mousset et al. (2014a)):

Ebiodeg ¼ 100� ð1� Ri=RÞ (5)

where R and Ri are the BOD5/COD ratio and BOD5/COD initial ratio,respectively.

Although biodegradability is low, a great improvement could be

noticed after 8 h of treatment; enhancement factor being between96% and 99% (Fig. 6). Before 8 h of treatment, biodegradability iseven much lower than the initial one. This could be due to thepresence on numerous chemical by-products produced by theoxidation reaction (Oturan et al., 2008; Dirany et al., 2012). It alsoconfirms the increase of toxicity at this period of time.

4. Conclusion

The combination of soil washing and eluates treatment wasachieved considering a quasi-complete mineralization of hydro-carbons present in the solution after washing. The use of Tween®

80 allowed the extraction of diesel hydrocarbons and particularlythe heaviest fractions. Concentrations in eluates were higher than900 mg L�1 with a washing solution containing 5% of Tween® 80.Nevertheless, the global efficiency of the washing was very lowwith only 1% of extracted hydrocarbons. The complete minerali-zation of the hydrocarbons in solution was not related to thetoxicity of the solution, which increased throughout the degrada-tion experiment. Meanwhile the biodegradability of the solution,which increased during time reaching a maximum of 20% revealingfurther potential biodegradation features.

This study showed that the treatment of soil washing solutionsby EF process is a technology that needs to be improved. The lowefficiency of the washing process could be partly explained by thelow availability of pollutants because of the ageing of the pollution.

Further work should focus on the possible implementation of athermal treatment of the soil column in order to enhance TPHextraction yield.

References

BASOL, 2015. French Database Website on Polluted Sites and Soils (Or PotentiallyPolluted) in French. http://basol.developpement-durable.gouv.fr. RetrievedJanuary 9, 2015.

Bautista, L.F., Sanz, R., Molina, M.C., Gonz�alez, N., S�anchez, D., 2009. Effect ofdifferent non-ionic surfactants on the biodegradation of PAHs by diverse aer-obic bacteria. Int. Biodeterior. Biodegrad. 63, 913e922.

Brillas, E., Sir�es, I., Oturan, M.A., 2009. Electro-Fenton process and related electro-chemical technologies based on Fenton's reaction chemistry. Chem. Rev. 109,6570e6631.

Ca~nizares, P., Paz, R., Sa�ez, C., Rodrigo, M.A., 2009. Costs of the electrochemicaloxidation of wastewaters: a comparison with ozonation and Fenton oxidation

D. Huguenot et al. / Journal of Environmental Management 153 (2015) 40e47 47

processes. J. Environ. Manag. 90, 410e420.Cozzarelli, I.M., Mckelvie, J.R., Baehr, A.L., 2014. In: Holland, H.D., Turekian, K.K.

(Eds.), 11.12-Volatile Hydrocarbons and Fuel Oxygenates Treatise onGeochemistry, second ed. Elsevier, Oxford, pp. 439e480.

Dirany, A., Sir�es, I., Oturan, N., Ozcan, A., Oturan, M.A., 2012. Electrochemicaltreatment of the antibiotic sulfachloropyridazine: kinetics, reaction pathways,and toxicity evolution. Environ. Sci. Technol. 46, 4074e4082.

El-Ghenymy, A., Oturan, N., Oturan, M.A., Garrido, J.A., Cabot, P.L., Centellas, F.,Rodríguez, R.M., Brillas, E., 2013. Comparative electro-Fenton and UVAphotoelectro-Fenton degradation of the antibiotic sulfanilamide using a stirredBDD/air-diffusion tank reactor. Chem. Eng. J. 234, 115e123.

Gomes, H.I., Dias-Ferreira, C., Ribeiro, A.B., 2013. Overview of in situ and ex situremediation technologies for PCB-contaminated soils and sediments and ob-stacles for full-scale application. Sci. Total Environ. 445e446, 237e260.

G�omez, J., Alc�antara, M.T., Pazos, M., Sanrom�an, M.A., 2010. Remediation of pollutedsoil by a two-stage treatment system: desorption of phenanthrene in soil andelectrochemical treatment to recover the extraction agent. J. Hazard. Mater. 173,794e798.

Iglesias, O., Sanrom�an, M.A., Pazos, M., 2014. Surfactant-enhanced solubilizationand simultaneous degradation of phenanthrene in Marine sediment by Electro-Fenton treatment. Indian Eng. Chem. Res. 53, 2917e2923.

ISO/TR 11046, 1994. Soil Quality e Determination of Mineral Oil Content e Methodby Infrared Spectrometry and Gas Chromatographic Method.

JORF, 2010. Arret�e du 28 octobre 2010 relatif aux installations de stockage ded�echets inertes. http://www.legifrance.gouv.fr/affichTexte.do?cidTexte¼JORFTEXT000023082021&dateTexte¼&categorieLien¼id.

Khalladi, R., Benhabiles, O., Bentahar, F., Moulai-Mostefa, N., 2009. Surfactantremediation of diesel fuel polluted soil. J. Hazard. Mater. 164, 1179e1184.

Laha, S., Luthy, R.G., 1992. Effects of nonionic surfactants on the solubilization andmineralization of phenanthrene in soil-water systems. Biotechnol. Bioeng. 40,1367e1380.

Lapinskiene, A., Martinkus, P., Rebzdaite, V., 2006. Eco-toxicological studies of dieseland biodiesel fuels in aerated soil. Environ. Pollut. 142, 432e437.

Leahy, J.G., Colwell, R.R., 1990. Microbial degradation of hydrocarbons in the envi-ronment. Microbiol. Rev. 54, 305e315.

Lee, M., Kang, H., Do, W., 2005. Application of nonionic surfactant-enhanced in situflushing to a diesel contaminated site. Water Res. 39, 139e146.

Loaiza-Ambuludi, S., Panizza, M., Oturan, N., €Ozcan, A., Oturan, M.A., 2013. Electro-Fenton degradation of anti-inflammatory drug ibuprofen in hydroorganic me-dium. J. Electroanal. Chem. 702, 31e36.

L�opez-Vizcaíno, R., S�aez, C., Ca~nizares, P., Rodrigo, M.A., 2012. Electrocoagulation ofthe effluents from surfactant-aided soil-remediation processes. Sep. Purif.Technol. 98, 88e93.

Majone, M., Verdini, R., Aulenta, F., Rossetti, S., Tandoi, V., Kalogerakis, N.,Agathos, S., Puig, S., Zanaroli, G., Fava, F., 2015. In situ groundwater and sedi-ment bioremediation: barriers and perspectives at European contaminatedsites. New Biotechnol. 32 (1), 133e146.

Mousset, E., Huguenot, D., van Hullebusch, E.D., Oturan, N., Esposito, G.,Oturan, M.A., 2013a. A new integrated approach to remove PAHs from highlycontaminated soil: soil washing combined to electro-Fenton process andpossible post-biological treatment. In: Lekkas, T.D. (Ed.), Proceeding of the 13th

International Conference in Environmental Science and Technology (Athens,Greece), p. 315, 1106e5516.

Mousset, E., Oturan, N., van Hullebusch, E.D., Guibaud, G., Esposito, G., Oturan, M.A.,2013b. Role of anode materials and initial COD on toxicity and biodegradabilityof synthetic soil washing solution containing Tween 80 and phentanthreneduring electro-Fenton treatment. In: Abstract of 16�eme Journ�ees D'Elec-rochimie'2013 (Paris, France), p. 138. http://www.je2013.fr/images/fascicule/book%20Oraux.pdf.

Mousset, E., Oturan, N., van Hullebusch, E.D., Guibaud, G., Esposito, G., Oturan, M.A.,2014a. Influence of solubilizing agents (cyclodextrin or surfactant) on phen-anthrene degradation by electro-Fenton process e study of soil washing recy-cling possibilities and environmental impact. Water Res. 48, 306e316.

Mousset, E., Oturan, N., van Hullebusch, E.D., Guibaud, G., Esposito, G., Oturan, M.A.,2014b. Treatment of synthetic soil washing solutions containing phenanthreneand cyclodextrin by electro-oxidation. Influence of anode materials on toxicityremoval and biodegradability enhancement. Appl. Catal. B Environ. 160e161,666e675.

Oturan, N., Trajkovska, S., Oturan, M.A., Couderchet, M., Aaron, J., 2008. Study of thetoxicity of diuron and its metabolites formed in aqueous medium duringapplication of the electrochemical advanced oxidation process “electro-Fen-ton”. Chemosphere 73, 1550e1556.

Oturan, N., Brillas, E., Oturan, M.A., 2012. Unprecedented total mineralization ofatrazine and cyanuric acid by anodic oxidation and electro-Fenton with aboron-doped diamond anode. Environ. Chem. Lett. 10, 165e170.

€Ozcan, A., Sahin, Y., Oturan, M.A., 2013. Complete removal of the insecticideazinphos-methyl from water by the electro-Fenton methodea kinetic andmechanistic study. Water Res. 47, 1470e1479.

Panizza, M., Cerisola, G., 2009. Direct and mediated anodic oxidation of organicpollutants. Chem. Rev. 109, 6541e6569.

Panizza, M., Michaud, P.A., Cerisola, G., Comninellis, C., 2001a. Anodic oxidation of2-naphthol at boron-doped diamond electrodes. J. Electroanal. Chem. 507,206e214.

Panizza, M., Michaud, P.A., Cerisola, G., Comninellis, C., 2001b. Electrochemicaltreatment of wastewaters containing organic pollutants on boron-doped dia-mond electrodes : prediction of specific energy consumption and requiredelectrode area. Electrochem. Commun. 3, 336e339.

Rodier, J., Legube, B., Merlet, N., 2009. Analyse de l'eau (Water analysis), ninth ed.Dunod, Paris (in French).

Rodrigo, M.A., Oturan, N., Oturan, M.A., 2014. Electrochemically assisted remedia-tion of pesticides in soils and water: a review. Chem. Rev. 114 (17), 8720e8745.

Rosas, J.M., Vicente, F., Santos, A., Romero, A., 2013. Soil remediation using soilwashing followed by Fenton oxidation. Chem. Eng. J. 220, 125e132.

Rowland, S., Donkin, P., Smith, E., Wraige, E., 2001. Aromatic hydrocarbon “humps”in the marine environment: unrecognized toxins? Environ. Sci. Technol. 35,2640e2644.

Singh, S.N., Kumari, B., Mishra, S., 2012. Microbial degradation of alkanes. In:Singh, S.N. (Ed.), Microbial Degradation of Xenobiotics, Environmental Scienceand Engineering. Springer-Verlag, Berlin Heidelberg.

Sires, I., Garrido, J., Rodriguez, R., Brillas, E., Oturan, N., Oturan, M.A., 2007. Catalyticbehavior of the Fe3þ/Fe2þ system in the electro-Fenton degradation of theantimicrobial chlorophene. Appl. Catal. B Environ. 72, 382e394.

Tang, W.Z., 2005. Physicochemical Treatment of Hazardous Wastes. CRC Press,London.

van Gestel, C.A.M., van der Waarde, J.J., Derksen, J.G.M., van der Hoek, E.E.,Veul, M.F.X.W., Bouwens, S., Rusch, B., Kronenburg, R., Stokman, G.N.M., 2001.The use of acute and chronic bioassays to determine the ecological risk andbioremediation efficiency of oil-polluted soils. Environ. Toxicol. Chem. 20,1438e1449.

Varma, R.K., Kaushal, R., Junnarkar, A.Y., Thomas, G.P., Naidu, M.U., Singh, P.P.,Tripathi, R.M., Shridhar, D.R., 1985. Polysorbate 80: a pharmacological study.Drug Res. 35, 804e808.

Wong, J.W.C., Fang, M., Zhao, Z.Y., 2004. Effect of surfactants on solubilization anddegradation of phenanthrene under thermophilic conditions. J. Environ. Qual.33, 2015e2025.

Zhou, M., Tan, Q., Wang, Q., Jiao, Y., Oturan, N., Oturan, M.A., 2012. Degradation oforganics in reverse osmosis concentrate by electro-Fenton process. J. Hazard.Mater. 215e216, 287e293.