Impact Of Electrokinetic Remediation On Microbial Communities Within Pcp Contaminated Soil

ARTICLE IN PRESS +

MODEL

Environmental Pollution xx (2006) 1e8 www.elsevier.com/locate/envpol

Impact of electrokinetic remediation on microbial communities within PCP contaminated soil G. Lear a,b,1, M.J. Harbottle a,b, G. Sills b, C.J. Knowles a,c, K.T. Semple d, I.P. Thompson a,* a

NERC-CEH-Oxford, Virology and Environmental Microbiology, Mansfield Road, Oxford, OX1 3SR, UK b Department of Engineering Science, University of Oxford, Parks Road, Oxford, OX1 3PJ, UK c Department of Earth Sciences, University of Oxford, Banbury Road, Oxford, OX2 6PN, UK d Department of Environmental Science, Faculty of Science and Technology, Lancaster University, Lancaster, LA1 4YQ, UK Received 21 August 2005; received in revised form 4 June 2006; accepted 14 June 2006

Electrokinetics negatively impacted soil. Abstract Electrokinetic techniques have been used to stimulate the removal of organic pollutants within soil, by directing contaminant migration to where remediation may be more easily achieved. The effect of this and other physical remediation techniques on the health of soil microbial communities has been poorly studied and indeed, largely ignored. This study reports the impact on soil microbial communities during the application of an electric field within ex situ laboratory soil microcosms contaminated with pentachlorophenol (PCP; 100 mg kg1 oven dry soil). Electrokinetics reduced counts of culturable bacteria and fungi, soil microbial respiration and carbon substrate utilisation, especially close to the acidic anode where PCP accumulated (36 d), perhaps exacerbated by the greater toxicity of PCP at lower soil pH. There is little doubt that a better awareness of the interactions between soil electrokinetic processes and microbial communities is key to improving the efficacy and sustainability of this remediation strategy. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Electrokinetic; Soil remediation; Soil health; Carbon substrate utilisation; Pentachlorophenol

1. Introduction Electrokinetics provides a physical method for the extraction of both metals and organic chemicals from contaminated sites, stimulating a directed movement of pollutants in response to the presence of an electric current (Maini et al., 2000). The applied current produces hydrogen ions (Hþ) at the anode and hydroxyl ions (OH) at the cathode, with a resulting pH gradient (Acar and Alshawabkeh, 1993). * Corresponding author. Tel.: þ44 1865 281630; fax: þ44 1865 281696. E-mail address: [email protected] (I.P. Thompson). 1 Present address: Williamson Centre for Molecular Environmental Science, School of Earth, Atmospheric and Environmental Sciences, University of Manchester, Manchester, M13 9PL, UK.

During this process, ions, ion complexes and particles are caused to migrate (via electromigration) to electrodes of opposite charge as a consequence of their orientation within the electric field, such that positively charged particles are stimulated to migrate to the negatively charged cathode, and vice versa. Dipolar interactions between the negatively charged surfaces of clay particles also encourage the transport of water and the solutes within it, towards the cathode, via electroosmosis (Acar et al., 1995). Electromobile contaminants are therefore stimulated to migrate to positions where they are amenable to removal, for example, via the collection and treatment of contaminated electrolyte solutions or precipitates formed close to the electrodes. Such processes now support an active remediation industry (e.g. Geokinetics BV, Rotterdam, ND), incorporating a range

0269-7491/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2006.06.037 Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS + 2

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

of novel applications for the removal of both organic and heavy metal contaminants in soil (e.g. LasagneÔ, ElektroKleanÔ, Electro-SorbÔ; Virkutyte et al., 2002). An important requirement before new methods for the remediation of contaminated land are accepted is to demonstrate that they are effective, and that they have few negative impacts on soil microbial communities and therefore soil health. Although little is currently known about the effect of electrokinetics on exposed soil microbial communities, recent studies have suggested that no serious impact on microbial health occurs when it is applied to pristine, non-contaminated soils (Lear et al., 2004). However, the influence of electrokinetics on soil health in the presence of a pollutant is not thought to have been previously investigated. It therefore remains possible that using electrokinetics to remove pollutants from contaminated land may affect soil microbial communities, possibly aggravating the toxic influence of contaminants and impairing the ability of the microbial community to intrinsically remediate any further incidences of pollution. Pentachlorophenol (PCP) has been widely used as a wood preservative and biocide since commercial production began in the 1930s, with worldwide annual production reaching 90 000 tonnes in 1983 (IRPTC, 1983). Pentachlorophenol is regarded toxicologically as a priority pollutant (Wild et al., 1993), being a suspected human carcinogen, hepatoxic agent and teratogen (McAllister et al., 1996). In mammals, PCP undergoes oxidative dechlorination in the liver to form tetrachlorohydroquinone (TCHQ) which may damage the liver, kidneys and immune system (EPA, 1986). Concerns about the safety of PCP have meant its use is now banned in many countries within the Northern Economic Divide. However, of all of the simple chlorinated phenols, PCP is thought to be the most resistant to microbial degradation (McAllister et al., 1996) and therefore its historic presence in the environment is a matter of ongoing concern (Muir and Eduljee, 1999), emphasising the need for cost-effective remediation strategies to treat PCP contaminated land. The electrokinetically induced movement of pentachlorophenol has previously been demonstrated to occur within an unsaturated soil (Harbottle et al., 2001). PCP has an acid dissociation constant (pKa) of 4.7 in water, whilst the natural pH of the soil in this study was pH 7.8. Therefore, the initially ionic PCP moves via electromigration towards the anode, but is later affected by the reverse flow of electroosmosis when entering the zone of low pH (i.e. <4.7) that develops close to the anode. The biocidal properties of PCP have been shown to impact soil microbial communities (Scheunert et al., 1995; Martins et al., 1997), especially under acidic conditions as the more toxic non-ionic pentachlorophenate is formed, so impacting negatively on degradation rates (Salminen and Haimi, 1998). Thus, the aim of this study was to develop a better understanding of the interactions between the soil, PCP contaminant, and indigenous microbial communities, during the application of electrokinetic technology. The experimental regime used in the present study was the progression of an earlier investigation (Lear et al., 2004), which examined the effect on a pristine soil, with no history of site contamination.

2. Methods To determine the effect of an electric field on microbial counts and activity within soil contaminated with PCP (100 mg kg1 oven dry soil) and inoculated with the PCP mineralising bacterium Sphingobium sp. UG30 (Leung et al., 1997), two treatments were constructed, in triplicate. Three electrokinetic cartridges were exposed to an applied current (3.14 A m2), using Isotech IPS601A benchtop power supplies (Southport, UK), whilst three more were established as negative controls, with no current applied. Both treatments were inoculated with 1  108 cells g1 oven dry soil of Sphingobium sp. UG30. This experimental regime was essentially as reported earlier (Lear et al., 2004) in a parallel study, which provided complete details of the electrokinetic technique and sampling procedure, also outlining electrokinetically induced changes in temperature, pH and conductivity within the soil cartridges. However, in the previous study, no PCP or degradative strain was added.

2.1. Soil electrokinetic cartridges Electrokinetic experiments were undertaken within a separable acrylic cartridge system (Fig. 1; inner cartridge size 13.0  5.4  5.9 cm), as described by Harbottle et al. (2001), but the present study used smaller cartridge dimensions. Cartridges were designed so that soil could be treated and stored within the detached central chamber for any length of time, and easily slotted into the remainder of the electrokinetic system, as required. Porous Daramic (Daramic Inc., Se´lestat, De) membranes at either end separated the soil within the central cartridge from the ionic solutions which contained the anode and cathode. The membranes prevented the passage of soil into the neighbouring chambers, whilst allowing a flow of water between them. Clamps and rubber seals bonded the three compartments to a base, reducing leakage, whilst ensuring chambers remained amenable to dismantle and reassemble. Graphite carbon electrodes (5.0  5.0  0.8 cm) were used to provide an electrical current, housed within the exterior electrode chambers. Holes (10 mm diameter) were bored into the graphite electrodes to enable a flow of water within each compartment. Chambers were placed within sealed plastic containers (opened only for sampling) in order to collect PCP-derived 14CO2 associated activity, facilitating its subsequent analysis. Electroosmotic flow was determined by measuring liquid loss from a Mariotte bottle, used to keep a constant head of fluid within the anode chamber. The soil used within this study was taken from the Oxford University Field Station, Oxford, and is an Evesham series heavy clay, having 53% clay, 25% sand and 22% silt (Lilley et al., 1996). Prior to use, all soil was air-dried and sieved (<2 mm), re-moistened to a water content of 18e21% (v:w), and stored at 4  C. Soil (0.5 kg) was added to each of six cartridges (statically compacted at a pressure of 50 kPa in 100 g layers, with the final height matching the level of the outflow ports in the electrode chambers). PCP (200 mg ml1) was added to each cartridge in 500 ml of de-gassed, deionised, water, applied to the top of the chamber. The employment of radiolabelled contaminants provides a superior method of tracking the fate of pollutants within soil. Therefore the PCP amendment comprised a component of [UL-14C]PCP (Sigma, UK), providing a final activity of 125 kBq g1 PCP within the soil. The leachate was collected and recycled three times and the cartridges allowed to equilibrate for 24 h to attain a saturation ratio of approximately 70%. This procedure has been perfected to achieve a concentration of w100 mg PCP kg1 oven dry soil, and a water content of 40e44% (v:w). Cartridges were stored in sealed containers for 5 d in the dark at 4  C, to facilitate a period of soil recovery. In order to determine the effect of electrokinetics on the ability of microorganisms to degrade PCP, a PCP mineralising bacterial strain, Sphingobium sp. UG30 was utilised (Leung et al., 1997). Sphingobium sp. UG30 was cultured in a minimal salts medium, containing L-glutamic acid (K2HPO4, 0.65 g l1; KH2PO4, 0.19 g l1; MgSO4$7H2O, 0.1 g l1; NaNO3, 0.5 g l1; 1 1  L-glutamic acid, 4.0 g l ; FeSO4, 0.003 g l ) and incubated (28 C, 140 rpm) in 250 ml Erlenmeyer flasks. Culture density was routinely monitored at OD600 nm, and where required, converted to bacterial numbers, as described by Wall and Stratton (1994). Soil chambers were inoculated using a point augmentation procedure. Briefly, to each chamber, aliquots (100 ml) of a concentrated bacterial suspension (containing 1.7  1010 cells ml1),

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS +

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

3

13.0cm Porous Daramic Membrane

Anode Chamber

Acrylic Chamber

5.9cm

5.4cm

Soil Sampled on day 36 Sampled on day 26 Sampled on day 14 Sampled on day 7 Sampled on day 0

Cathode Chamber

Fig. 1. Schematic of the sampling regime within the soil electrokinetic chamber. Nine soil cores were taken across the length of each chamber in increments of 1.44 cm (first core 0.72 cm from anode chamber), 0, 7, 14, 26 and 36 d after the current was applied. Inner chamber dimensions 13.0  5.4  5.9 cm. Graphite electrode dimensions: 5.0  5.0  0.8 cm. harvested and washed in phosphate buffer (K2HPO4, 0.65 g l1; KH2PO4, 0.19 g l1) during the mid-exponential growth phase, were evenly distributed at depths of 5 mm, 30 mm and 55 mm (utilising 36, 24 and 36 evenly spaced injection points, respectively). This procedure allowed an average cell density of w1  108 cells g1 oven dry soil of Sphingobium sp. UG30 to be achieved for the entire chamber. Many soils, including the one used in this study, lack a significant biomass of PCP degrading microorganisms (Mueller et al., 1991; Seech et al., 1991). Consequently, the specific interaction of the contaminant with an individual, inoculated degradative strain may be studied in detail. Soil samples were removed from the chambers immediately prior to the application of electrokinetics (0 d) and then after a further 7, 14, 26 and 36 d. This was achieved by dividing the chambers into five lanes, such that a fifth of the soil cartridge was sampled during each sampling time, as indicated in Fig. 1. Metal tubing (14 mm inner diameter, Oxford Instruments Superconductivity, Oxford, UK) was employed to extract soil cores (w5 g) from nine locations, evenly spaced along the transect between the anode and cathode chambers. Soil pH, temperature, conductivity and voltage were determined as previously described (Harbottle et al., 2001; Lear et al., 2004), with similar results. During the entire study, the pH of the catholyte was maintained at pH 7 with the addition of sulphuric acid which had the effect of maintaining electroosmotic flow throughout the experiment.

Groningen, NL) for 5 min at a wavelength of 104e165 nm. Water was chosen as an initial extractant to provide a measure of the amount of PCP which is easily desorbed, or present within the pore water, whilst acetonitrile extracted the majority of the remaining, more recalcitrant PCP. Non-extractable [14C]PCP-associated activity was determined using a combustion method as similarly undertaken by Reid et al. (2000). Soil samples (1 g) were packed into paper combustion cones, prior to analysis using a sample oxidiser (Model 307). Carbosorb-E and Permofluor-E were used to trap CO2, and as a scintillant, respectively. Combustaid (100 ml) was added to each sample prior to the 3 min combustion process. The Packard sample oxidiser and respective liquid agents were from Canberra Packard, UK. This combination of procedures provided a mass balance of the total [14C]PCP-associated activity within the soil. The respirometric method of Reid et al. (2001) was used to assess the extent of PCP degradation within the soil. The breakdown of [14C]PCP releases 14 CO2, which may be trapped within a NaOH solution, where measurements of 14 CO2 evolution can be directly correlated to the loss of PCP. Each sealed experimental treatment housed a glass scintillation vial containing 3 ml of 1 M NaOH solution to collect evolved CO2. Traps were regularly removed and solutions replaced. Liquid scintillation fluid (16 ml, Ultima Gold, Perkinelmer, USA) was added to each vial and the activity of the trapped 14CO2 was determined by liquid scintillation counting following a 24 h rest period.

2.3. Analysis of soil microbial numbers and activity 2.2. Determination of the loss and extractability of [UL-14C]PCP-associated activity in soil [14C]PCP-associated activity was assessed by sequential extraction, first agitating 0.5 g soil with 10 ml water and then 10 ml acetonitrile, on a rotary shaker (6 rpm, 24 h), centrifuging the sample (5000 rpm, 15 min) and removing the supernatant between stages. Supernatant (6 ml) was combined with 12 ml UltimaGold XR scintillation fluid (PerkinElmer, USA) and activity counted using a Wallac Liquid Scintillation Counter (1217 Rackbeta;

Culturable counts of soil bacteria and fungi, soil microbial respiration and carbon substrate utilisation patterns were examined using methods previously outlined (Lear et al., 2004). Counts of culturable soil bacteria and fungi were examined across the cartridge transect by incubating soil dilutions on Plate Count Agar (Oxoid Ltd., Hampshire, UK; containing 100 mg l1 cyclohexamide, as an antifungal agent) and potato dextrose agar (Oxoid Ltd.; containing 320 mg l1 AureomycinÒ as an antibacterial agent), respectively. Soil micro¨ hlinger (1996) bial respiration was monitored using a modified method of O

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS +

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

which quantifies CO2 production by recording the pH change of a NaOH solution, determined by colorimetric analyses. The ability of the soil microbial community to utilise a range of different carbon sources was assessed by the colour development of a tetrazolium dye within BIOLOG EcoplatesÔ (Hayward, CA, USA), allowing both spatial and temporal differences in the soil microbial communities to be determined.

2.4. General statistical analysis Bacterial and fungal counts were calculated per gram of oven dry soil and log-transformed [log10(x þ 2)] to improve the homogeneity of the variance of the data. Data were analysed using the statistical package GENSTAT (GENSTAT V release 6.1). Unless otherwise stated, analyses of variance (ANOVAs) were constructed using a general two-way design where position between electrodes was a fixed effect and sampling time was a repeated measures factor. Standard methods in GENSTAT were used to test the homogeneity of variance and normality of the data to ensure it remained within the required parameters for the statistical test. t-tests were paired and two-tailed. BIOLOGÔ data from each plate were transformed to facilitate the confounding effects on rates of colour development caused by differing inoculation densities, using the following formula, (S  C )/B  109, where S is the sample well OD620 nm, C is the control well OD620 nm and B is bacterial CFU g1 oven dry soil.

By the end of the study (36 d), a pH profile developed across the length of the electrokinetic chambers from pH 3.8 close to the anode (2.2 cm) to pH 7.9 close to the cathode (2.2 cm), as similarly detailed within Lear et al. (2004). The distribution of PCP within the soil of the electrokinetic chambers concentrated towards the anode after 36 d of electrokinetic exposure; the result of electromigration. [14C]PCPassociated activity increased by 13% close to the anode (1.3 cm), with no associated increase within the control cartridges (Fig. 2). Over the duration of study, a decrease was observed of water-extractable PCP (Fig. 3), more so within the electrokinetic treatment than the control (17  5% (SE) and 46  4% (SE) of original soil 14C activity after 36 d, respectively). However, on the final sampling day, no significant difference was detected between total [14C]PCP Recovered from Treatments as a Percentage of Original PCPAssociated Activity

(i)

100 80 60 40 20 0

0

6

12

18

24

30

120

36

(ii)

100 80 60 40 20 0

0

6

12

18

24

30

36

Time (Days)

3. Results

14C

120

Percentage14C-PCP-Associated Activity Recovered (Relative to Day 0)

4

140 120 100 80 60 40 20 0

0

2

4

6

8

10

12

Distance from Anode (cm) Fig. 2. Total PCP associated activity recovered from soil after 36 d of applied current in each treatment (3.14 mA m2) from soil cartridges inoculated with 1  108 cells Sphingobium UG30 sp. g1 oven dry soil; data are 14C-activity recovered as a percentage of original PCP-associated activity inoculated (:) without current applied; (6) with current applied; Error bars are 1  SE; three replicates.

Fig. 3. Percentage [14C]PCP-associated activity recovered over 36 d (relative to activity measured on 0 d) from soil inoculated with w1  108 cells Sphingobium sp. UG30 g1 dry soil (i) without current applied (ii) with current applied (3.14 mA m2); data are [14C]PCP-associated activity recovered via [14C]PCP trapping ( ); acetonitrile extraction ( ); water extraction ( ). Error bars are 1  SE (three replicates).

-associated activity extracted from experimental or control treatments (t ¼ 0.68, df ¼ 14, p ¼ 0.504). No 14C activity was detected from the soil following sequential extraction using the combustion method, suggesting a complete removal of PCP. Culturable numbers of both bacteria and fungi remained constant within the control treatments over the experimental period (i.e. 6.74  0.04SE, 3.34  0.04SE Log10 (xþ2) CFU g1 oven dry soil of bacteria and fungi, respectively; Fig. 4). However, bacterial counts decreased (4.3%, Log10(xþ2) units) within the electrokinetic cell as compared to the control treatment (t ¼ 2.53, df ¼ 58, p ¼ 0.014; all samples combined). The application of electrokinetics reduced culturable counts of both bacteria and fungi close to the anode (0.7 cm) by 17%, and 30%, respectively (0 d as compared to all other time-points combined). Bacterial counts increased within the inoculated soils, reflecting the addition of Sphingobium cells (as compared to 6.50.03 Log10 (xþ2) CFU g1 oven dry soil (Lear et al., 2004), all samples combined (t ¼ 3.81, df ¼ 48, p ¼ 4.00  104)). The application of electric current significantly (p ¼ 0.037) reduced soil microbial respiration below that of the control by 21 d (i.e. 3.320.33SE and 3.93  0.42SE mg CO2 g1 oven dry soil, respectively; Fig. 5). Furthermore, respiration was lowest close to the anode end (0.72 cm) of the electrokinetic chambers; only 2.55  0.57SE mg CO2 g1 oven dry soil. Colour development profiles of the BIOLOG EcoPlatesÔ showed the preferential order of carbon substrate utilisation

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS +

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

5

Fig. 4. Colony forming units of soil fungi (left) and bacteria (right) (i) without electrokinetics (ii) with electrokinetics, sampled along the cartridge transect from anode to cathode for each sampling time-point during electrokinetic treatment initially contaminated with 100 mg PCP kg1 oven dry soil. Data are log-transformed log10(x þ 2) CFU g1 oven dry soil. Sampling timepoints: () 0 d; (:) 7 d; (A) 14 d, (-) 26 d; (d) 36 d; pooled standard error ¼ 0.06 and 0.26 for each treatment respectively; an ANOVA for the combination is shown on the graph (L, sampling position; S, sampling timepoint; P, probability.) Error bars are 1  SE; three replicates.

for the control treatment to be amino acids > amines z carboxylic acids z carbohydrates z phenols > polymers (as determined by the greatest absorbance (OD620 nm) after 10 d incubation (data not shown)). Within soil exposed to the electric field, the order of carbon substrate utilisation was amines > polymers > amino acids > carboxylic acids > phenols > carbohydrates (Fig. 6). The extent of carbon substrate utilisation was significantly less with an electric field applied (x¼ 7.4 and 40.9, with and without an applied current, respectively; t ¼ 11.7, df ¼ 1112, p ¼ 6.54  1030). Both the rate

mg CO2 g-1 Oven Dry Soil

6

p 0.632

5

0.037

4 3 2 1 0

2

5

8

11

Distance from Anode (cm) Fig. 5. Soil respiration rate in each treatment (red, control (no applied current); blue, electrokinetic), sampled along the cartridge transect from anode to cathode on 21 d. Data are mg CO2 g1 oven dry soil; Pooled standard error ¼ 0.99 and 0.45 for each treatment respectively; an ANOVA for the sampling position is shown on the graph. Bars are 1  SE; three replicates.

and extent of carbon substrate utilisation was inhibited adjacent to the anode (0.72 cm), with no visible utilisation of carbon source occurring by 36 d.

4. Discussion The toxic effect of PCP is thought to be due to its ability to uncouple oxidative phosphorylation (Steiert et al., 1988) and alter the composition of bacterial membranes (Dercova´ et al., 2004). The impact of even low concentrations (50e100 mg kg1 dry soil) of PCP have been described as severe (Chaudri et al., 2000), reducing soil microbial counts (Chaudri et al., 1996; Martins et al., 1997; Salminen and Sulkava, 1997), biomass, microbial respiration rates (Zelles et al., 1986) and substrate induced respiration (Scheunert et al., 1995). In the present study, PCP reduced fungal counts as compared to a non-contaminated soil (4.9  0.1 log10(x þ 2) fungal CFU g1 dry soil; Lear et al., 2004). Bacterial counts were higher, reflecting the addition of a Sphingobium inoculum, but declined towards the anode of the electrokinetic treatment over the duration of the study. However, within the control treatment (where no electrokinetics was applied), the application of PCP stimulated the range of carbon sources utilised by the soil microbial count as compared to a PCP free, non-electrokinetic treatment (Lear et al., 2004). This is thought to be a stress-induced response as, following the loss of PCP susceptible soil fauna, the availability of resources may have provided energy and nutrients available to the

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS + 6

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

Fig. 6. Colour development profiles (OD620) of BIOLOG EcoplatesÔ incubated over 10 d for soil taken (i) 12.7, (ii) 6.5, and (iii) 0.7 cm from the anode, respectively, at the final timepoint (36 d). Treatment was 100 mg PCP kg1 oven dry soil, exposed to electrokinetics (3.13 mA cm1) Data were transformed OD620 values grouped as an average for each of the six carbon substrate types in the BIOLOGÔ EcoPlate: >, polymers; B, amino acids; d, carbohydrates; -, carboxylic acids; 6, phenols; , amines. Error bars are 1  SE; three replicates.

saprophytic microbial community. As a consequence of this selection, a shift in the composition of the microbial community may have resulted, to one able to utilise a greater amount of carbon per cell than the initial community. Hence, carbon substrate utilisation increased, but without any associated detectable increase in soil respiration, or culturability. Previous studies revealed that the addition of a Sphingobium inoculum did not significantly affect patterns of carbon substrate utilisation (data not shown), and this was perhaps a consequence of its poor survival in PCP contaminated soil. Indeed, loss of cell viability of PCP degraders has been observed in both liquid culture media (Gonzalez and Hu, 1991) and soil (Karlson et al., 1995). This may account for the observation that the application of electrokinetics did not accelerate the loss of PCP in contaminated soil, the very purpose for which such an approach may be applied in soil remedial applications.

Importantly, electrokinetics appeared to increase the proportion of PCP extractable only with acetonitrile, as compared to water. You and Liu (1996) revealed the increased desorption of PCP from soil at higher pH. It is therefore suggested that the acidfront caused by the application of current may have increased the extent of PCP sorption to the soil interface within the electrokinetic treatments, which could inhibit the efficacy of the physical remediation strategy. Bioaugmentation failed to reduce PCP concentrations sufficiently to remove its toxic effect, as observed by a decrease in carbon substrate utilisation and soil microbial respiration. The addition of PCP affected the order of utilisation of carbon sources within the BIOLOGÔ plates. However, it remains difficult to hypothesise why these changes may have occurred as few publications exist regarding the significance of the order of substrate utilisation, in environmental terms. Nevertheless, the greater use of L-glutamic acid observed within this study is a likely consequence of the greater number of PCP degrading organisms within the soil; glutamic acid is the precursor of glutathione, which PCP degrading bacteria have to synthesise in order for degradation to occur (Leung et al., 1997). Both the application of PCP and electrokinetics were previously shown to induce changes in the soil microbial community (Martins et al., 1997; Lear et al., 2004, respectively), the latter reducing bacterial numbers close to the anode. In addition to any direct impact of electrokinetics on the microbial community, the movement of PCP towards the anode end of the electrokinetic cell, as detected in this study (similarly observed in Harbottle et al., 2001) may have caused further disruption to the microbial community. It has been suggested that PCP may be more toxic under acidic conditions, as PCP is in its phenolic, lipophilic form. Moreover, the most significant influence of electrokinetics on the soil is thought to be due to changes in soil pH, as oxidation reactions at the anode generate an acid front, whilst reduction at the cathode produces a base front (Acar and Alshawabkeh, 1993). Although the pH was controlled in the cathode chamber, acidic conditions developed in the soil close to the anode (pH w 2, as similarly observed in Lear et al., 2004), which may have caused stress to exposed soil microorganisms, reducing their tolerance to the pollutant. Hence, the decrease in culturable counts of both bacteria and fungi may have been enhanced by the combination of low pH and soil PCP toxicity that interacted synergistically to make conditions even more harmful. It is known that PCP may retain its biocidal activity, even in its bound form by releasing small amounts of soluble residues (Scheunert et al., 1995). We therefore suggest the prevailing acidic conditions acted to increase the toxicity of PCP within the soil. That counts of soil bacteria and fungi were not more greatly impacted (at least close to the anode) could be the result of the soil pH profile and PCP distribution within the microclimate of the soil, which combined to produce a protective effect. Microorganisms located in the inner compartment of soil microcosms may have been relatively unaffected, as they were not exposed to the same pH profile or PCP toxicity as those at the exterior (Martins et al., 1997), or within the free pore water.

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS +

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

Our previous studies demonstrated that microbial respiration may increase close to the anode even when culturable counts of soil microorganisms decrease (Lear et al., 2004). It is possible that this was a stress-induced response. However, when applying electrokinetics to a PCP contaminated soil, microbial respiration was greatly reduced, especially close to the acidic anode (23%), a likely consequence of the significant decline in microbial counts and biomass. PCP contamination reduced the extent to which the polymers were utilised, whilst increasing the utilisation of the simpler amino acids and amines. In conclusion, the results of this study (as Lear et al., 2004) demonstrate that the application of electrokinetics to soil alters both the physico-chemical characteristics of the soil and the exposed microbial community. The direct effect of the applied current on soil bacteria could not be ascertained, this influence being inseparable from other electrochemically induced soil changes such as soil acidification close to the anode. However, the combined effects of PCP and electrokinetic exposure reduced microbial counts, respiration and carbon substrate utilisation potential greatly, probably in part at least, as a consequence of the increased toxicity of PCP at lower soil pH. Further studies are now required to examine the impact of electrokinetics and other physical remediation strategies on soil microbial communities, also assessing the long-term sustainability of such approaches. Whilst the small-scale laboratory-based microcosms used within this study allowed differing treatment regimes to be easily assessed and regulated, we must remain cautious of extrapolating such findings to the field scale. Here non-regular influences including soil heterogeneity and changing weather patterns will further influence results and process design may vary, conforming to the constraints of time, economics and land ownership. In addition to the influence of electrokinetics, mass transfer processes may also have a greater effect on soil pH and PCP distribution. It is hoped that future work will begin to examine the effects of electrokinetics on soil microbial communities at the field-scale, an issue that has so far been widely overlooked. This will help to ensure that the ability of soils to microbially remediate further incidences of pollution are not unnecessarily compromised by otherwise well-intentioned remedial intervention. Acknowledgements This work was funded via an Engineering and Physical Sciences Research Council (EPSRC) CASE studentship with CEH-Oxford. The majority of the experimental apparatus used was expertly constructed by Mr C. Waddup within the Department of Engineering Science, University of Oxford. We thank Paula Clasper and Tatiana Bouchard at the University of Lancaster, UK, for their kind assistance in using the soil oxidation equipment. References Acar, Y.B., Alshawabkeh, A., 1993. Principles of electrokinetic remediation. Environmental Science and Technology 27, 2638e2647.

7

Acar, Y.B., Gale, R.J., Alshawabkeh, A.N., Marks, R.E., Puppala, S., Bricka, M., Parker, R., 1995. Electrokinetic remediation e basics and technology status. Journal of Hazardous Materials 40, 117e137. Chaudri, A.M., McGrath, S.P., Knight, B.P., Johnson, D.L., Jones, K.C., 1996. Toxicity of organic compounds to the indigenous population of Rhizobium leguminosarum biovar Trifolii in soil. Soil Biology and Biochemistry 28, 1483e1487. Chaudri, A.M., Lawlor, K., McGrath, S.P., 2000. Pentachlorophenol utilisation by indigenous soil microorganisms. Soil Biology and Biochemistry 32, 429e432. ˇ ertı´k, M., Mal’ova´, A., Seja´kova´, A., 2004. Effect of chloropheDercova´, K., C nols on the membrane lipids of bacterial cells. International Biodeterioration and Biodegradation 54, 251e254. EPA, 1986. Chemical Fact Sheet: Pentachlorophenol, Environmental Protection Agency (US EPA) gopher://ecosys.drdr.Virginia.EDU/00/library/gen/ toxics/pentachlorophenol. 2002 Gonzalez, J.F., Hu, W.S., 1991. Effect of glutamate on the degradation of pentachlorophenol by Flavobacterium sp. Applied Microbiology and Biotechnology 35, 100e104. Harbottle, M.J., Sills, G.C., Thompson, I.P., Jackman, S.A., 2001. Movement of pentachlorophenol in unsaturated soil by electrokinetics. Proceedings of the Third Symposium and Status Report on Electrokinetic Remediation, Karlsruhe, Germany, April 17, pp. 1e13. IRPTC, 1983. Data Profile on Pentachlorophenol. International Register of Potentially Toxic Chemicals. United Nations Environment Program, Geneva. Karlson, U., Miethling, R., Schu, K., Scholtz Hansen, S., Uotila, J., 1995. Biodegradation of PCP in soil. In: Hinchee, R.E., Wilson, J.T. (Eds.), Bioremediation of Recalcitrant Organics 3(7). Batelle Press, 380 pp. Lear, G., Harbottle, M.J., van der Gast, C.J., Jackman, S.A., Knowles, C.J., Sills, G.C., Thompson, I.P., 2004. The effect of electrokinetics on soil microbial communities. Soil Biology and Biochemistry 36, 1751e1760. Leung, K.T., Cassidy, M.B., Shaw, K.W., Lee, H., Trevors, J.T., LohmeirVogel, E.M., Vogel, H.J., 1997. Pentachlorophenol biodegradation by Pseudomonas spp. UG25 and UG30. World Journal of Microbiology and Biotechnology 13, 305e313. Lilley, A.K., Fry, J.C., Bailey, M.J., Day, J., 1996. Comparison of aerobic heterotrophic taxa-isolated from four root domains of mature sugar beet (Beta vulgaris). FEMS Microbiology Ecology 21, 231e242. Maini, G., Sharman, A.K., Knowles, C.J., Sunderland, G., Jackman, S.A., 2000. Electrokinetic remediation of metals and organics from historically contaminated soil. Journal of Chemical Technology and Biotechnology 75, 657e664. Martins, J.M., Monrozier, L.J., Chalamet, A., Bardin, R., 1997. Microbial response to repeated applications of low concentrations of pentachlorophenol in Alfisol under pasture. Chemosphere 35, 1637e1650. McAllister, K., Lee, H., Trevors, J.T., 1996. Microbial degradation of pentachlorophenol. Biodegradation 7, 1e40. Mueller, J., Middaugh, D.P., Lantz, S.E., Chapman, P.J., 1991. Biodegradation of creosote and pentachlorophenol in contaminated groundwater: chemical and biological assessment. Applied and Environmental Microbiology 57, 1277e1285. Muir, J., Eduljee, G., 1999. PCP in the freshwater and marine environment of the European Union. The Science of the Total Environment 236, 41e56. ¨ hlinger, R., 1996. Soil respiration by titration. In: Schinner, F., O ¨ hlinger, R., O Kandeler, E., Margesin, R. (Eds.), Methods in Soil Biology. SpringerVerlag, London, pp. 95e98. Reid, B.J., Stokes, J.D., Jones, K.C., Semple, K.T., 2000. Nonexhaustive cyclodextrin-based extraction technique for the evaluation of PAH bioavailability. Environmental Science and Technology 34, 3174e3179. Salminen, J.E., Sulkava, P.O., 1997. Decomposer communities in contaminated soil: Is altered community regulation a proper tool in ecological risk assessment of toxicants? Environmental Pollution 97, 45e53. Salminen, J.E., Haimi, J., 1998. Responses of the soil decomposer community and decomposition processes to the combined stresses of pentachlorophenol and acid precipitation. Applied Soil Ecology 9, 475e481. Scheunert, I., Attar, A., Zelles, L., 1995. Ecotoxicological effects of soil-bound pentachlorophenol residues on the microflora of soils. Chemosphere 30, 1995e2009.

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037

ARTICLE IN PRESS + 8

MODEL

G. Lear et al. / Environmental Pollution xx (2006) 1e8

Seech, A.G., Trevors, J.T., Bulman, T.L., 1991. Biodegradation of pentachlorophenol in soil: the response to physical, chemical, and biological treatments. Canadian Journal of Microbiology 37, 440e444. Steiert, J.G., Thoma, W.J., Ugurbil, K., Crawford, R.L., 1988. P-31 Nuclear magnetic-resonance studies of effects of some chlorophenols on Escherichia-coli and a pentachlorophenol-degrading bacterium. Journal of Bacteriology 170, 4954e4957. Virkutyte, J., Sillanpaa, M., Latostenmaa, P., 2002. Electrokinetic soil remediation. The Science of the Total Environment 289, 97e121.

Wall, A., Stratton, G., 1994. Effects of a chromated-copper-arsenate wood preservative on the growth of a pentachlorophenol degrading bacterium. Water, Air and Soil Pollution 82, 723e737. Wild, S., Harrad, S.J., Jones, K.C., 1993. Chlorophenols in digested UK sewage sludges. Water Research 27, 1527e1534. You, C.N., Liu, J.C., 1996. Desorptive behaviour of chlorophenols in contaminated soil. Water Science and Technology 33, 263e270. Zelles, L., Scheumert, I., Korte, F., 1986. Comparison of methods to test chemicals for side-effects on soil-microorganisms. Ecotoxicology and Environmental Safety 12, 53e69.

Please cite this article in press as: G. Lear et al., Impact of electrokinetic remediation on microbial communities within PCP contaminated soil, Environmental Pollution (2006), doi:10.1016/j.envpol.2006.06.037