Electrokinetic Enhancement Of Phenanthrene Bio Degradation In Creosote-polluted Clay Soil

Environmental Pollution 142 (2006) 326e332 www.elsevier.com/locate/envpol

Electrokinetic enhancement of phenanthrene biodegradation in creosote-polluted clay soil Jose´-Luis Niqui-Arroyo, Marisa Bueno-Montes, Rosa Posada-Baquero, Jose´-Julio Ortega-Calvo* Instituto de Recursos Naturales y Agrobiologı´a, C.S.I.C., Apartado 1052, E-41080-Seville, Spain Received 5 May 2005; received in revised form 29 September 2005; accepted 2 October 2005

Electrokinetic bioremediation is a potentially effective technology to treat PAH-polluted, clay-rich soils. Abstract Given the difficulties caused by low-permeable soils in bioremediation, a new electrokinetic technology is proposed, based on laboratory results with phenanthrene, to afford bioremediation of polycyclic aromatic hydrocarbons (PAH) in clay soils. Microbial activity in a clay soil historically polluted with creosote was promoted using a specially designed electrokinetic cell with a permanent anode-to-cathode flow and controlled pH. The rates of phenanthrene losses during treatment were tenfold higher in soil treated with an electric field than in the control cells without current or microbial activity. Results from experiments with Tenax-assisted desorption and mineralization of 14C-labeled phenanthrene indicated that phenanthrene biodegradation was limited by mass-transfer of the chemical. We suggest that the enhancement effect of the applied electric field on phenanthrene biodegradation resulted from mobilization of the PAH and nutrients dissolved in the soil fluids. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Polycyclic aromatic hydrocarbons; Bioremediation; Electrokinetics; Electro-osmosis; Desorption

1. Introduction Creosote is a complex mixture of organic chemicals, mainly polycyclic aromatic hydrocarbons (PAH). Used worldwide as a wood preservative, its accidental spillage and improper use and handling at processing sites has led to the contamination of underlying soils and groundwater. Bioremediation of creosote-polluted sites is considered a realistic alternative to other remediation methods, as it has the advantages of relatively low cost and reasonable execution periods (Mueller et al., 1989). A major factor limiting the success of bioremediation of PAH is the presence in soil of a high proportion of clay-sized particles. From the experience already gained in bioremediation technology, a high clay content in a contaminated soil places serious doubts on the final success of bioremediation alone

* Corresponding author. Tel.: þ34 95 462 4711; fax: þ34 95 462 4002. E-mail address: [email protected] (J.-J. Ortega-Calvo). 0269-7491/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2005.10.007

as a treatment strategy. Clay-rich soils may present a limited bioavailability of PAH, because of their high surface area available for sorption (Lahlou and Ortega-Calvo, 1999), and a difficulty for bacterial transport through the soil, what limits the access to the source of hydrophobic substrates (Lahlou et al., 2000). The oxygen and nutrient supply to the degrading populations may also be limiting, due to slow diffusion and low hydraulic conductivity. Associated operations such as handling, excavation, and nutrient amendments may be physically hampered because of the high consistency of clay-rich soils, which are also prone to bacterial clogging (Kaufman, 1994). Successful attempts at bioremediation in creosote-polluted sites have been documented only with sandy soils (Breedveld and Karlsen, 2000; Breedveld and Sparrevik, 2000; Carriere and Mesania, 1995; Eriksson et al., 2000). Electroremediation, consisting of the controlled application of low-power DC electric fields to polluted soils, is especially indicated for clay soils. This technology, which relies on three processesdelectromigration (movement of charges),

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

electro-osmosis (water), and electrophoresis (charged particles)dhas already been used to remove heavy metals and organic pollutants from soil (Virkutyte et al., 2002). Its mobilizing potential can also be coupled to biodegradation processes, as has been shown in laboratory studies where (i) bacterial strains have been transported electrokinetically through diesel-contaminated soil (Lee and Lee, 2001) and model aquifer material (DeFlaun and Condee, 1997; Wick et al., 2004), (ii) the organic pollutant itself was transported towards soil zones harboring microbial populations able to degrade the pollutant, such as 2,4-dichlorophenoxyacetic acid (Jackman et al., 2001) or p-nitrophenol (Ho et al., 1995), and (iii) a co-metabolic substrate was injected into soil to promote TCE biodegradation (Rabbi et al., 2000). To our knowledge, there are no studies about the effect of this technology on the biodegradation of PAH, possibly due to the limited transport that these chemicals often exhibit in clay soil during electrokinetic treatment (Saichek and Reddy, 2003). This research focuses on the effect of an electric field on the biodegradation of PAH present in a clay-rich, creosote-polluted soil. We employed a historically polluted soil containing a high load of native PAH, of which phenanthrene was followed as a target compound. Compound disappearance was measured in electrokinetic cells designed to promote microbial activity, and the kinetics of phenanthrene desorption and biodegradation were determined in solid phase and soil slurries. Our main objectives were (i) to determine if low-voltage DC currents promote phenanthrene biodegradation in clay-rich soil, and, if so, (ii) to determine the possible mechanism(s) involved. 2. Materials and methods 2.1. Soil

327

in the soil inside cylindrical glass filtercandles (Robuglasfilter-Gera¨te GmbH, Hattert, Germany), which served as electrode reservoirs. The amount of dry soil packed in the cell was approximately 700 g. The filtercandles had porous walls (160e250 mm pore size), allowing the exchange of electrode solution into and out of the reservoirs. The separation distance between the electrodes was 16 cm. To promote maximum microbial activity in the soil by a permanent anode-to-cathode flow and controlled pH, these reservoirs were kept filled with a buffer solution, which was recirculated independently. The reservoirs were connected to a peristaltic pump, and 1 M phosphate buffer (K2HPO4/KH2PO4) adjusted to pH 8.00 (anode) or 5.80 (cathode) was recirculated at a constant flow rate of 12 mL min
1. This buffer concentration was chosen not only because it allowed the efficient control of pH in the electrode reservoirs, but also to minimize possible changes induced in the pore fluid properties due to the P consumption associated to microbial assimilation, which is typical of bioremediation of hydrocarbons (Huesemann, 1994). In addition, the soil was packed in the cell in layers after saturating with inorganic salts solution, which contained KH2PO4 (0.9 g L
1), K2HPO4 (0.1 g L
1), NH4NO3 (0.1 g L
1), MgSO4 $ 7H2O (0.1 g L
1), CaCl2 (0.080 g L
1), FeCl3 $ 6H2O (0.01 g L
1), and 1 mL L
1 of a microelements stock to obtain the final concentrations of 0.0014 g L
1 for Na2MoO4 $ 2H2O and 0.002 g L
1 for each of the following: Na2B4O7 $ 10H2O, ZnSO4 $ H2O, MnSO4 $ H2O and CuSO4 $ 5H2O. This solution presented a pH value of 5.8. The porosity of the soil packed in this way ranged from 0.158 to 0.188 (determined gravimetrically). During packing, the soil was inoculated, along the line between the electrodes, with the bacterium Novosphingobium sp. LH128, which was capable of using phenanthrene as a sole carbon and energy source for growth. The bacterium was cultured and prepared for the experiments as previously described (Garcia-Junco et al., 2003), and was added at a cell density of 2.6  107 cells g
1. The test cell was treated for a total of 337 h (14 days), during which the voltage applied was 0.5e0.6 V cm
1 (77 h) and 0.2e0.3 V cm
1 (260 h), in alternate periods. The DC power supply used was a Freak HY3005D-3 model unit. A control cell was maintained under exactly the same conditions (including saturation with water, inoculation, and recirculation of electrode fluids), but without an electric field. An abiotic control was also run with soil autoclaved three times, which received no inoculum, and was treated for 135 h at 0.7e0.8 V cm
1. After treatment, soil cores (25 g dry soil) were taken from the cell along the anode-cathode axis and analyzed in duplicate for residual phenanthrene content. Statistical comparisons were performed with analysis of variance and Scheffe´ post hoc test at p ¼ 0.10 and p ¼ 0.05.

2.3. Electro-osmotic flow

The soil used in this study was a clay soil, classified as a calcaric fluvisol, provided by EMGRISA (Madrid, Spain) from a wood-treating facility in Andu´jar (Jae´n, southern Spain), with a record of pollution by creosote exceeding 100 years. The site geology consists of a superficial granular fill of up to 1 m in thickness, a horizon of silts and clays (2e3 m), a water-bearing horizon of sandy gravel, and compact marls. Groundwater is located at a depth of approximately 4 m. After sampling different locations, which served as preliminary tests, a homogenous sample (50 L) of the silty clay layer from a heavily contaminated point was prepared by air drying for 2 weeks, thorough mixing, and sieving (2 mm mesh). The resulting soil sample was characterized according to standard methods of soil analysis (Klute, 1986; Page et al., 1982), and had the following characteristics: 6.6% moisture; pH 7.92; 23.4% CaCO3; 3.26% organic matter; 0.106% organic nitrogen (Kjeldahl); 0.9 mg kg
1 available phosphorus; particle size distribution: 1.1% coarse-grained sand, 2.4% fine-grained sand, 37.0% silt, and 60.0% clay; 2777 mg kg
1 total petroleum hydrocarbons; 4501 mg kg
1 total PAH (sum of 16 EPA PAH). Phenanthrene content was 1319.5  78.5 mg kg
1 dry soil. The number of indigenous microorganisms able to grow with phenanthrene, estimated as colony-forming units on solid medium containing the chemical as the sole source of carbon and energy (Vila et al., 2001), was 2  104 cells g dry soil
1.

Percolated columns were used in a similar way to that one previously described for the study of bacterial transport through clay-rich porous media (Lahlou et al., 2000; Ortega-Calvo et al., 1999). The soil was wet-packed in glass columns with an inside diameter of 0.9 cm (cross-sectional area ¼ 0.125 cm2), and a length of 10 cm. A portion of soil (1 cm) next to the cathode was spiked during packing with 27,000 dpm (266.3 ng) of [14C]phenanthrene (8.3 mCi mmol
1, radiochemical purity >98%, Sigma Chemical Co., St. Louis, MO) completely dissolved in 0.5 ml of inorganic salts solution. Hollow, stainless steel cylindrical electrodes were connected to each side of the columns. A reservoir with inorganic salts solution was connected to the anode side to keep the soil saturation conditions. The formation of an hydraulic gradient was prevented by height adjustment of the reservoir. A constant electric field of 1 V cm
1 was applied for 5 h, and the column effluent passing through the cathode was collected and weighted. Then, it was mixed with 5 mL of liquid scintillation cocktail (Ready Safe, Beckman Instruments, Fullerton, CA, USA), and radioactivity was measured with a liquid scintillation counter (Beckman Instruments Inc., Fullerton, Calif.; model LS5000TD).

2.2. Electrokinetic treatment

2.4. Desorption

The cell designed, shown in Fig. 1, consisted of a polyethylene body protected with an inner layer of glass, to prevent phenanthrene loss due to sorption. Two cylindrical electrodes, made of stainless steel, were inserted

Phenanthrene desorption kinetics were determined in duplicate with the Tenax solid-phase extraction method (Cornelissen et al., 1998). Briefly, 1 g dry soil, 70 mL milli-Q water, 0.35 mL formaldehyde (40%), and 1.5 g Tenax

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

328

Peristaltic pump

A

+ Stainless steel anode

Stainless steel cathode

Water flow

Tygon tubes Filtercandle

Contaminated clay soil Glass body cell

Phosphate buffer pH 8.00

Magnetic stirrer

Phosphate buffer pH 5.80

B

Fig. 1. Design of electroremediation cell used to treat creosote-polluted clay soil. (A) Schematic diagram. (B) Detail.

TA beads were placed in a separation funnel. The funnel was continuously shaken at 23  2  C on an orbital shaker operating at 170 rpm. After certain time intervals, the Tenax was separated from the soil suspension, and replaced by fresh Tenax. The sorbent was extracted by shaking with 50 mL of hexane for 48 h. The extract was evaporated to near dryness, redissolved in acetonitrile, and filtered. Analysis of phenanthrene was performed by HPLC as described later. The total mass of phenanthrene desorbed (extracted by Tenax) plus the amount still present in the soil at the end of the experiments was 97.3  1.7% of the initial mass determined by whole-soil extractions.

2.5. Biodegradation Biodegradation was measured in solid phase and in slurries. Solid-phase experiments were performed in closed biometer flasks (Bellco glass, NJ) with 25 g of dry soil adjusted to 80% of field capacity with sterile, distilled water. A portion of the water used to adjust humidity was an aqueous solution of approximately 100,000 dpm of [9-14C]phenanthrene (8.3 mCi mmol
1, radiochemical purity >98%, Sigma Chemical Co., St. Louis, MO). This guaranteed the homogenization of the labeled compound with native phenanthrene. The flasks were closed with Teflon-lined stoppers, and incubated at 23  2  C. Production of 14CO2 was measured as radioactivity appearing in the alkali trap

of the biometer flasks. The trap contained 1 mL of 0.5 M NaOH. Periodically, the solution was removed from the trap and replaced with fresh alkali. The NaOH solution was mixed with 5 mL of liquid scintillation cocktail, and the mixture kept in darkness for about 8 h for dissipation of chemiluminescence. Radioactivity was measured with a liquid scintillation counter. Residual contents of native phenanthrene were determined in separate flasks that were incubated under the same conditions but contained no 14C-labeled compound. After certain time intervals, duplicate flasks were sacrificed and kept frozen at
80  C until analysis for phenanthrene content by HPLC. For experiments with soil slurries, 15 g of soil was placed in 250-mL Erlenmeyer flasks, 1 mL of distilled water containing 100,000 dpm of radiolabeled phenanthrene was added to the soil, and the mixture was homogenized. A sterile, inorganic salts solution (pH 5.8), described above, was added to complete a final volume of 100 mL. The slurries were then inoculated with Novosphingobium sp. LH128, which was cultured and prepared for mineralization experiments as previously described (Garcia-Junco et al., 2003). Each flask received an inoculum of approximately 107 cells g
1. The flasks were then closed with Teflon-lined stoppers, from which an 8-mL vial containing 1 mL of 0.5 M NaOH was suspended to trap 14CO2. The flasks were incubated at 23  2  C on an orbital shaker operating at 100 rpm. Measurements of mineralization of the radiolabeled phenanthrene and residual concentration of the native compound were carried out as above.

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

2.6. Data analysis Desorption and mineralization/biodegradation data were analyzed in two ways. On the one hand, mineralization and biodegradation rates were calculated and compared as the slope of the regression lines drawn with the points belonging to the phase of maximum mineralization or disappearance in experiments with radiolabeled and native phenanthrene, respectively (Ortega-Calvo et al., 1995). On the other hand, data of desorption and biodegradation of native phenanthrene could be empirically described by the following twocompartment kinetic model: ÿ
St =So ¼ Frap exp
Krap t þ Fslow expð
Kslow tÞ ð1Þ where St and So (g) are the soil-sorbed amounts of phenanthrene at time t (h) and at the start of the experiment, respectively (Cornelissen et al., 1998), Frap and Fslow are the rapidly and slowly desorbing (or biodegrading) fractions, Krap and Kslow (h
1) are the rate constants of rapid and slow desorption/biodegradation. Frap, Fslow, Krap and Kslow were obtained by minimizing the cumulative squared residuals between experimental and calculated values of ln(St/So). The software used for the minimization was Microsoft Excel 97 (Solver option). It should be noted that the use of this empirical, two-compartment model with desorption and biodegradation data does not necessarily imply a direct dependence or influence of one of these two processes on the other.

2.7. Phenanthrene analysis Soil samples were defrosted, dried at 40  C, ground with a mortar and pestle, and extracted in a Soxhlet with 100 mL dichloromethane for 8 h. Then, the organic solvent was evaporated to near dryness, and the residue was dissolved in 15 mL acetonitrile and filtered through Whatman grade 1 filter paper. Phenanthrene analysis was carried out using a Waters HPLC system (2690 separations module and 474 scanning fluorescence detector. Column: Nova-Pak C18 Waters, 3.9  150 mm; flow: 1 mL min
1; mobile phase: 60% acetonitrile 40% water). Analysis of phenanthrene content of a reference material (diesel particulate matter 2975, National Institute of Standards and Technology, Gaithersburg, USA) was in good agreement with the certified value (the measured value being 10.8  2.8 mg kg
1, and the certified value 17.0  2.8 mg kg
1). A control was also run for possible losses during drying. The soil sample was wetted with distilled water up to 80% field capacity, and processed as described above. The concentration of phenanthrene was 1111.5  89.1 mg kg
1. In experiments with soil slurries, the solids were collected by centrifugation at 7000 rpm for 10 min, and treated as described above.

329

temperature in the control cell without current). Voltage measurements along the anode-cathode axis at points located in the same sections of the cells showed a linear relationship to the distance from the electrodes, and confirmed the monodimensional distribution of voltage within the cell. Fig. 2 shows the results obtained in a representative experiment. Phenanthrene losses after treatment were significantly higher in the electroremediated soil than in the control cells without current or microbial activity. The enhancement effect of the electric current on biodegradation was observed in the middle of the cell and close to the cathode, where a significant reduction ( p ¼ 0.10) of phenanthrene concentration was observed after treatment, whereas little or no compound was lost without current. Assuming that biodegradation occurred linearly during that period, the rate of compound disappearance in this region was 1.320  0.450 mg kg
1 h
1 with current, tenfold higher than the rate without current (0.140  0.002 mg kg
1 h
1). No significant compound dissipation was detected in the abiotic control. There was a slight compound depletion observed close to the anode, but it was not statistically significant ( p ¼ 0.10 and p ¼ 0.05). This shows that the electric current did not cause any chemical degradation of phenanthrene. We considered two hypotheses that could explain the observed electrokinetic stimulation: (i) Direct effect of an increased soil temperature observed during electrokinetic treatment; (ii) a positive influence of the electro-osmotic flow through soil caused by the current. 3.2. Effect of temperature The possible contribution to the observed electrokinetic stimulation by heating of the soil was examined in a experiment in 1600

3. Results 1400

An electroremediation cell was designed to test the effect of electric fields on phenanthrene biodegradation in soil. Preliminary tests showed that for a successful electrokinetic bioremediation it was crucial to avoid the generation of extreme pH values in the electrode reservoirs (acidic pH at the anode and basic pH at the cathode) due to electrolytic reactions: Migration of protons and/or hydroxyl ions into soil would have prevented biodegradation, due to unfavorable ecological conditions for phenanthrene-degrading bacteria and cessation of electro-osmotic flow. This was achieved by continuous renewal of electrode reservoir solutions, maintaining their pH at a constant value. Other relevant operating aspects were (1) the progressive increase in current intensity observed during the treatment, which accompanied the application of a constant potential (0.8 V cm
1), and was possibly caused by a higher conductivity of the soil, and (2) an increased temperature of the treated soil (35  C, consistently 5e9  C above the

[phen] (mg/Kg)

3.1. Electrokinetic bioremediation 1200

ABIOTIC

- FIELD

1000

+ FIELD

800

600 2

4

6

8

10

12

14

Distance to anode (cm) Fig. 2. Effect of electric field on phenanthrene biodegradation in creosotepolluted clay soil. Profiles of phenanthrene concentrations in soil treated in electrokinetic cells under current application (þFIELD), as compared with a control without current (
FIELD), and sterilized soil (ABIOTIC). Initial phenanthrene content is indicated by the dotted line. Error bars correspond to one standard deviation of duplicate measurements.

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

330

which [14C]phenanthrene mineralization was compared in duplicate samples of soil maintained at 25 and 35  C. The soil (25 g) was inoculated with 8  107 cells of Novosphingobium sp. LH128. Mineralization of the labeled compound was determined as described in Section 2 for solid-phase experiments. The results showed that the increase in temperature observed during electrokinetic treatment of the soil had little influence on biodegradation (Fig. 3). At 25  C, the maximum rate of mineralization was 1.03  0.03 mg kg
1 h
1, while at 35  C the rate increased only to 1.11  0.11 mg kg
1 h
1. 3.3. Electro-osmotic mobilization of phenanthrene The electro-osmotic flow of water through soil is a welldocumented process (see Section 1). Indeed, during electrokinetic treatments we could observe macroscopically the movement of water in the anode-to-cathode direction. However, the design of the cell did not allow measurements of the mobilized volume of water. Therefore, to test whether the applied electric field induced the electro-osmotic flow of water through this soil, and the mobilization of the phenanthrene dissolved in the soil fluid, a different experimental setup was used. The system was a column-type cell in which the electric field was applied between two hollow, cylindrical electrodes. The portion of the soil close to the cathode was spiked with radiolabeled phenanthrene. In this way, we could estimate an electroosmotic flow toward the cathode of approximately 0.03 ml/h. The amount of [14C]phenanthrene eluted after 5 h was 1.1 ng. 3.4. Phenanthrene desorption and biodegradation A series of desorption and biodegradation experiments was performed in the absence of an electric field to understand possible effects on bioavailability of phenanthrene by the observed electro-osmotic flow. The kinetics of desorption, determined by Tenax extraction for 41 days in sterilized soil suspensions, showed the existence of two first-order kinetic fractions 16

(Fig. 4). The size of the rapidly desorbing fraction calculated by the model accounted for 96.1  0.1% of the initial amount of phenanthrene present in the soil. The rate constant for the rapid fraction (Krap) was 0.38  0.00 h
1, and several orders of magnitude larger than the corresponding Kslow (2.70  0.03  10
3 h
1). The dissipation of the native compound observed in biodegradation experiments, performed in solid- and slurry-phase conditions, also followed first-order kinetics (Fig. 5). The final extents of compound disappearance were very close to the size of the rapidly desorbing fraction determined in Tenax experiments. Furthermore, Fig. 5 shows good agreement between phenanthrene concentrations determined experimentally in soil during biodegradation and those predicted by the first-order model of Eq. (1). The calculated values for Krap were 4.05  0.21  10
3 and 45.95  5.80  10
3 h
1 in solid- and slurry-phase biodegradation experiments, respectively, while the corresponding Kslow were 1.74  2.46  10
5 and 1.35  0.21  10
3 h
1. The calculated values for Frap in solid- and slurry-phase experiments were, respectively, 96.10  2.82% and 95.40  0.59%. 4. Discussion The enhancement effect of an applied electric field on phenanthrene biodegradation was very likely caused by the mobilization of the soil fluids associated to electro-osmosis. This is supported by (i) the results from mineralization experiments, showing that the enhancement of biodegradation by the electric field was not due to an increased temperature, (ii) the observed mobilization of water and associated [14C]phenanthrene due to electro-osmotic flow, and (iii) the occurrence of a significant amount (more than 95%) of phenanthrene as a fast-desorbing fraction, as revealed by Tenax-assisted desorption experiments. The electrokinetic mobilization of phenanthrene in soils has been observed in several studies that used solubilityenhancing agents (surfactants, co-solvents, and cyclodextrins) in the flushing solutions. These studies have shown that, in spite of the limited solubility of the compound (1.1 mg L
1 (Schwarzenbach et al., 2003)), which restricts its transport

14

0 -1

10

-2

Ln St / S0

% 14C mineralized

12

8 6

-3 -4

4

-5

2

-6 0

0 0

2

4

6

8

10

12

14

16

18

200

400

600

800

1000

TIME (h)

Time (days) Fig. 3. Mineralization of phenanthrene in inoculated, creosote-polluted clay soil maintained at 25  C (circles) and 35  C (triangles).

Fig. 4. Kinetics of phenanthrene desorption from creosote-polluted clay soil, as measured by Tenax extraction. The dashed line represents the curve fitting the two-phase desorption model in Eq. (1).

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

331

60

1800

A 1600 50

mg/Kg soil

1200

40

1000 30 800 600

20

% 14C mineralized

1400

400 10 200 0

0 500 1000 1500 2000 2500 3000 3500 4000 4500

0

TIME(h) 50

1800

B 1600

mg/Kg soil

1200 30 1000 800 20

% 14C mineralized

40

1400

600 400

10

200 0 0

25

50

75

0 100 125 150 175 200 225 250

TIME(h) Fig. 5. Mineralization of [14C]phenanthrene (circles) and evolution of concentration of native phenanthrene (triangles) by indigenous microbial population in creosote-polluted clay soil (A) and inoculated soil slurries (B). The dashed line represents the curve fitting the two-phase desorption/biodegradation model in Eq. (1). Error bars correspond to one standard deviation of measurements in duplicate flasks.

through the soil, phenanthrene can be mobilized even in the absence of these agents. For example, Saichek and Reddy (2003) observed the nearly complete removal of 500 mg kg
1 phenanthrene in the region adjacent to the anode after the electrokinetic treatment of kaolin soil for 60 days with only deionized water. In the rest of the soil profile, however, the phenanthrene concentration remained unaltered. A significant removal of compound (up to 35%) was also reported in another study where kaolinite samples were initially amended with 1.9 mg kg
1 phenanthrene and treated electrokinetically for 6 days with a NaCl purging solution (Ko et al., 2000). In these studies, performed with column-type reactors, the uneven distribution of phenanthrene through the soil after electrokinetic treatment evidences that the electro-osmotic flow may

generally not be uniform. Some sections present variable flow rates and pore pressures, probably as a result of the complex nature of the factors that control the electro-osmotic flow, including zeta potential and electrical gradient. This may apply to our study, and explain the differences in phenanthrene biodegradation along the anode-cathode axis. Although we observed mobilization of [14C]phenanthrene induced by electro-osmotic flow, the electrokinetic treatment in the absence of biological activity (Fig. 2) did not cause significant changes in the phenanthrene concentration profile. However, the first-order kinetics of biodegradation observed in separate biodegradation experiments are indicative of concentration-dependent biodegradation. It is therefore possible that the electrokinetic treatment caused a mobilization of

332

J.-L. Niqui-Arroyo et al. / Environmental Pollution 142 (2006) 326e332

phenanthrene towards bacterial cells, thus contributing to an increased mass transfer, and therefore biodegradation rate, of the compound within the soil. Indeed, with a typical population of 107 to 108 cells g
1, the colonies of degrading microorganisms in soil are separated by a typical average distance of 100 mm, being surrounded by a desolate environment where the slow resupply of hydrophobic pollutants by dissolution and diffusion from distant sources dictates the biodegradation rate (Bosma et al., 1997). According to this mechanism, a significant electroosmotic mobilization of phenanthrene at the macroscale level would have not been needed for the observed stimulation. The high clay content of the soil probably limited not only bioavailability, but also nutrient diffusion. Indeed, minimal sample handling and adjustment of humidity yielded immediate, although slow, biodegradation of phenanthrene in solidphase experiments. The application of the electric current may have led to a number of simultaneous processes and changes within the soil which may have influenced biodegradation, including ionic movements (affecting cations such a Ca2þ, as well as anions such as PO3þ 4 , present in the electrode buffers) and changes in moisture content and/or dissolved oxygen, producing the latter in electrolytic reactions (Virkutyte et al., 2002). These factors limit the natural attenuation of PAH in creosote-polluted soils (Mueller et al., 1989). 5. Conclusions Our results strongly suggest that electrokinetic bioremediation may be a reasonable alternative to treat in situ clay-rich soils polluted with creosote and other PAH-containing materials. The optimization of the process with other PAH than phenanthrene and for a cost-effective application of this technology in situ will be the subject of future investigations. Acknowledgements We thank EMGRISA for the provision of the soil sample. Support for this research was provided by the European Union (contract QLRT-1999-00326), Spanish CICYT (BIO20001857-CE and REN2001-3523), and Spanish Ministry of Environment (057/2004/3). References Bosma, T.N.P., Middeldorp, P.J.M., Schraa, G., Zehnder, A.J.B., 1997. Mass transfer limitation of biotransformation: quantifying bioavailability. Environmental Science and Technology 31, 248e252. Breedveld, G.D., Karlsen, D.A., 2000. Estimating the availability of polycyclic aromatic hydrocarbons for bioremediation of creosote contaminated soils. Applied Microbiology and Biotechnology 54, 255e261. Breedveld, G.D., Sparrevik, M., 2000. Nutrient-limited biodegradation of PAH in various soil strata at creosote contaminated site. Biodegradation 11, 391e399. Carriere, P.P., Mesania, F.A., 1995. Enhanced biodegradation of creosotecontaminated soil. Waste Management 15, 579e583. Cornelissen, G., Rigterink, H., Ferdinandy, M.M.A., Van Noort, P.C.M., 1998. Rapidly desorbing fractions of PAHs in contaminated sediments as a predictor of the extent of bioremediation. Environmental Science and Technology 32, 966e970.

DeFlaun, M.F., Condee, C.W., 1997. Electrokinetic transport of bacteria. Journal of Hazardous Materials 55, 263e277. Eriksson, M., Dalhammar, G., Borg-Karlson, A.K., 2000. Biological degradation of selected hydrocarbons in an old PAH/creosote contaminated soil from gas work site. Applied Microbiology and Biotechnology 53, 619e626. Garcia-Junco, M., Gomez-Lahoz, C., Niqui-Arroyo, J.L., Ortega-Calvo, J.J., 2003. Biodegradation- and biosurfactant-enhanced partitioning of polycyclic aromatic hydrocarbons from nonaqueous-phase liquids. Environmental Science and Technology 37, 2988e2996. Ho, S.V., Sheridan, P.W., Athmer, C.J., Heitkamp, M.A., Brackin, J.M., Weber, D., Brodsky, P.H., 1995. Integrated in situ soil remediation technology: the Lasagna process. Environmental Science and Technology 29, 2528e2534. Huesemann, M.H., 1994. Guidelines for land-treating petroleum hydrocarboncontaminated soils. Journal of Soil Contamination 3, 299e318. Jackman, S.A., Maini, G., Sharman, A.K., Sunderland, G., Knowles, C.J., 2001. Electrokinetic movement and biodegradation of 2,4-dichlophenoxyacetic acid in silt soil. Biotechnology and Bioengineering 74, 40e48. Kaufman, A.K., 1994. Selection of bioremediation for site cleanup: decision factors. In: Flathman, P.E., Jerger, D.E., Exner, J.H. (Eds.), BioremediationdField Experience. CRC Press, Boca Raton, pp. 51e57. Klute, A., 1986. Methods of Soil Analysis. Physical and Mineralogical Methods. Soil Science Society of America, Madison, WI. Ko, S.-O., Schlautman, M.A., Carraway, E.R., 2000. Cyclodextrin-enhanced electrokinetic removal of phenanthrene from a model clay soil. Environmental Science and Technology 34, 1535e1541. Lahlou, M., Ortega-Calvo, J.J., 1999. Bioavailability of labile and desorptionresistant phenanthrene sorbed to montmorillonite clay containing humic fractions. Environmental Toxicology and Chemistry 18, 2729e2735. Lahlou, M., Harms, H., Springael, D., Ortega-Calvo, J.J., 2000. Influence of soil components on the transport of polycyclic aromatic hydrocarbondegrading bacteria through saturated porous media. Environmental Science and Technology 34, 3649e3656. Lee, H.-S., Lee, K., 2001. Bioremediation of diesel-contaminated soil by bacterial cells transported by electrokinetics. Journal of Microbiology and Biotechnology 11, 1038e1045. Mueller, J.G., Chapman, P.J., Pritchard, P.H., 1989. Creosote-contaminated sites, their potential for bioremediation. Environmental Science and Technology 23, 1197e1201. Ortega-Calvo, J.J., Birman, I., Alexander, M., 1995. Effect of varying the rate of partitioning of phenanthrene in nonaqueous-phase liquids on biodegradation in soil slurries. Environmental Science and Technology 29, 2222e2225. Ortega-Calvo, J.J., Fesch, C., Harms, H., 1999. Biodegradation of sorbed 2,4dinitrotoluene in a clay-rich, aggregated porous medium. Environmental Science and Technology 33, 3737e3742. Page, A.L., Miller, R.H., Keeny, D.R., 1982. Methods of Soil Analysis. Chemical and Microbiological Properties. Soil Science Society of America, Madison, WI. Rabbi, M.F., Clark, B., Gale, R.J., Ozsu-Acar, E., Pardue, J., Jackson, A., 2000. In situ TCE bioremediation study using electrokinetic cometabolite injection. Waste Management 20, 279e286. Saichek, R.E., Reddy, K.R., 2003. Effect of pH control at the anode for the electrokinetic removal of phenanthrene from kaolin soil. Chemosphere 51, 273e287. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003. Environmental Organic Chemistry, second ed. John Wiley & Sons, Hoboken. Vila, J., Lopez, Z., Sabate, J., Minguillon, C., Solanas, A.M., Grifoll, M., 2001. Identification of novel metabolite in the degradation of pyrene by Mycobacterium sp. strain AP1: actions of the isolate on two and three ring polycyclic aromatic hydrocarbons. Applied and Environmental Microbiology 67, 5497e5505. Virkutyte, J., Sillanpa¨a¨, M., Latostenmaa, P., 2002. Electrokinetic soil remediationdcritical overview. The Science of the Total Environment 289, 97e121. Wick, L.Y., Mattle, P.A., Wattiau, P., Harms, H., 2004. Electrokinetic transport of PAH-degrading bacteria in model aquifers and soil. Environmental Science and Technology 38, 4596e4602.