w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6

Avai lab le at www.sc iencedi rect .com

journa l homepage : www.e lsev i er . com/ loca te /wat res

Enhanced removal of 1,2-dichloroethane by anodophilicmicrobial consortia

Hai Pham, Nico Boon, Massimo Marzorati, Willy Verstraete*

Laboratory of Microbial Ecology and Technology (LabMET), Ghent University, Coupure Links 653, B 9000 Gent, Belgium

a r t i c l e i n f o

Article history:

Received 16 January 2009

Received in revised form

2 April 2009

Accepted 2 April 2009

Published online 17 April 2009

Keywords:

Microbial fuel cell

Anodophilic bacteria

1,2-Dichloroethane

* Corresponding author. Tel.: þ32 (0)9 264 59E-mail addresses: willy.verstraete@ugentURL: http://labmet.ugent.be

0043-1354/$ – see front matter ª 2009 Elsevidoi:10.1016/j.watres.2009.04.004

a b s t r a c t

1,2-Dichloroethane (1,2-DCA) is a well-known recalcitrant groundwater contaminant. New

environment-friendly approaches for the removal of 1,2-DCA that does not bring about

volatilization of the compound are required. In this study, different anodophilic consortia

enriched in microbial fuel cells (MFCs) operated under airtight conditions were shown to

effectively degrade 1,2-DCA (up to 102 mg per liter reactor volume per day), while

concomitantly generating a current. An anodophilic consortium previously enriched with

acetate as the electron donor changed its composition at the rate of 48% per week and

increased its richness (Rr) 3-fold, upon adapting to 1,2-DCA as the new electron donor.

After being stable, during 1 month of operation, it removed up to 95% of the 1,2-DCA

amount in the medium in the first 2 weeks, while converting 43 � 4% of electrons available

from the removal to electricity. A natural consortium from a 1,2-DCA contaminated site

changed its composition at the rate of 9% per week and increased its Rr 2-fold, upon

adapting to the MFC anode conditions with 1,2-DCA as the electron donor. After being

stable, during 1 month of operation, it removed up to 85% of the 1,2-DCA amount in the

medium in the first 2 weeks and the coulombic efficiency was 25 � 4%. The operation of the

MFCs under closed circuit conditions resulted in higher 1,2-DCA removal rates than the

operation under open circuit conditions, indicating that bioelectrochemical activities

enhanced the removal of 1,2-DCA in the MFC anode. The production of ethylene glycol,

acetate and carbon dioxide indicated that the anodophilic bacteria oxidatively metabolized

1,2-DCA, probably by means of a hydrolysis-based pathway. The results show that MFCs

can be potentially used as a practically convenient technology for the biological removal of

1,2-DCA.

ª 2009 Elsevier Ltd. All rights reserved.

1. Introduction Dinglasan-Panlilio et al., 2006). Poor disposals practices resul-

1,2-Dichloroethane (1,2-DCA), or ethylene dichloride, is the

most abundant chlorinated industrial product (Bhatt et al.,

2007; De Wildeman et al., 2003). It is also an intermediate in

some other industrial processes (such as the synthesis of

fluorocarbon and 1,1,1-trichloroethane) (Bhatt et al., 2007;

76; fax: þ32 (0)9 264 6248.be, haithe.pham@ugent.

er Ltd. All rights reserved

ted in a wide spreading of this compound in groundwater

(Janssen et al., 1984, 1994). This is of major concern since the

compound may cause damages to kidney, liver and nerve

systems in humans and is a suspected carcinogen (Hughes

et al., 1994). Moreover, the possibility of human exposure to it

through groundwater is considerable (Hughes et al., 1994).

.be (W. Verstraete).

.

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6 2937

Approaches to degrade 1,2-DCA have been investigated

intensively. Together with physical and chemical methods,

biodegradation of 1,2-DCA using microorganisms deserves

attention as it offers environment-friendly and low cost reme-

diation (De Wildeman and Verstraete, 2003; Marzorati et al.,

2006). 1,2-DCA can be microbially degraded both under aerobic

and anaerobic conditions (Dinglasan-Panlilio et al., 2006). A

number of pure cultures capable of aerobically degrading 1,2-

DCA have been isolated; most of them belonging to the bacterial

genera of Pseudomonas, Xanthobacter and Ancylobacter (Hage and

Hartmans, 1999; Janssen et al., 1985; Stucki et al., 1983;

Vandenwijngaard et al., 1992). These bacteria use 1,2-DCA as the

electron donor and the degradation occurs either through

a hydrolysis dehalogenation pathway or through an oxidation

pathway both leading to the formation of intermediates such as

2-chloroethanol and monochloroacetate and finally to the

formation of CO2 (Dinglasan-Panlilio et al., 2006; Nobre and

Nobre, 2004). Anaerobic degradation of 1,2-DCA is mainly based

on reductive dechlorination, in which the compound plays

the role of an electron acceptor (Bhatt et al., 2007). Bacteria

capable of doing this are Dehalococoides ethenogenes (He et al.,

2003; Maymo-Gatell et al., 1999) and some methanogens such as

Methanobacterium thermoautotrophicum, Methanosarcina barkeri

and Methylosinus trichosporium that can cometabolically dechlo-

rinate 1,2-DCA to ethene (Egli et al., 1987; Oldenhuis et al., 1989).

Notably, a pure culture, Desulfitobacterium dichloroeliminans DCA

1, was reported to be capable of complete reductive dehaloge-

nationof1,2-DCAtoethene(DeWildemanetal., 2003;Maeset al.,

2006; Marzorati et al., 2007).Onlyrecently, anaerobic oxidation of

1,2-DCA with nitrate as the electron acceptor has been reported

for the first time (Dinglasan-Panlilio et al., 2006).

One challenge to the current 1,2-DCA degradation tech-

nologies is that the compound is volatile (Gordon et al., 2002)

and thus can escape into the air from groundwater, causing

a secondary pollution. Therefore, a technology that can cope

with this challenge is needed. As both aerobic and anaerobic

oxidation of 1,2-DCA is feasible, the bioanode of a microbial

fuel cell could be used to degrade 1,2-DCA under gastight

conditions. In bioanodes, microorganisms catalyze the

oxidation of substrates, converting part of the chemical

energy available in substrates into electrical energy (Allen and

Bennetto, 1993; Clauwaert et al., 2008). Reductive dehaloge-

nation of chlorinated compounds using MFCs has been

reported (Aulenta et al., 2007, 2008; Strycharz et al., 2008).

However, in those studies, the process was mainly based on

the activity of microorganisms at the cathode. In addition, no

MFC studies have addressed 1,2-DCA removal. In this article,

we reported for the first time the use of a microbial fuel cell for

an accelerated removal of 1,2-DCA, based on the anode

oxidation of this compound.

Fig. 1 – Descriptive plot of the MFC reactor (A) and

simplified scheme (B) of the setup to treat 1,2-DCA used in

this study. R: external resistor; CEM: cation exchange

membrane. Not shown in the figure is the cathode loop

recirculating the catholyte (hexacyanoferrate). The unit of

the dimensions of the anode working space indicated in (A)

is cm.

2. Experimental section

2.1. MFC reactors

Each microbial fuel cell reactor used in this study was con-

structed with Perspex frames and contained an anode

compartment and a cathode compartment (Fig. 1A). A cation-

specific membrane (Ultrex CMI7000, Membranes International

Inc., US) separated the two compartments. Each compartment

included an endplate (10 � 10 � 1.5 cm3) and a square-holed

subframe (8 � 8 � 0.9 cm3), with 0.9 cm being the distance

from the endplate to the membrane (Fig. 1A). There were two

side holes on each frame for the influent and effluent. The

working space of each compartment had a 7 � 7 � 0.9 cm3

dimension. The anodic electrode was a graphite plate

(5 � 4 � 0.3 cm3) (Le Carbone, Belgium) in contact with

a graphite rod (0.5 cm diameter) (Le Carbone, Belgium) that

penetrated the anode endplate through a hole (0.5 cm diam-

eter). Hence, the net anodic capacity (NAC) was 38 mL (cm3)

or 38 � 10�6 m3 (¼the anode working space � the inserting

electrode volume ¼ 7 � 7 � 0.9 cm3 � 5 � 4 � 0.3 cm3). The

cathode working space was filled with graphite granules

(diameters between 1.5 and 5 mm, Le Carbone, Belgium) in

contact with a graphite rod (0.5 cm diameter) (Le Carbone,

Belgium) that penetrated the cathode endplate through a hole

(0.5 cm diameter). The catholyte was an aqueous solution of

50 mM K3Fe(CN)6 and 100 mM KH2PO4 buffer (Merck, Belgium)

adjusted to pH 7 with 1 M NaOH. It was recirculated through

the cathode matrix and replenished before its decoloration

(Aelterman et al., 2006). The anode and cathode graphite rods

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 62938

were connected to an external resistor of 20 ohm and to a data

acquisition system (Fig. 1A). After assembling, the whole

reactor was sealed using superglue (SUPERCOL, BricoBi,

Belgium) to make it airtight.

2.2. Operation

Two types of MFCs were investigated for their capability of

removing 1,2-DCA. The first type (AceMFCs) included two

MFCs that already had acquired well-performing anodic

microbial consortia because they were previously enriched

with acetate, fed continuously, as the substrate. With an

acetate loading rate equal to 19 g COD L�1 NAC d�1, the

AceMFCs had generated a stable current of 6.8 � 0.2 mA for

2 months. This corresponds to the power output of

24.3 � 0.1 W m�3 NAC or 462 � 2 mW m�2 anodic electrode

surface, which are comparable with other working MFC

systems (Clauwaert et al., 2008). When the experiments

started, the acetate feeding was stopped and the entire

anodic content of each of the AceMFCs was replaced by a 1,2-

DCA containing medium before its anode compartment was

connected to a medium circulating system. The 1,2-DCA

containing medium was a modified M9 medium (with 6 g L�1

Na2HPO4.2H2O, 3 g L�1 KH2PO4.2H2O, 0.1 g L�1 NH4Cl, 0.05 g

L�1 NaCl, 0.24 g L�1 MgSO4.7H2O and trace elements as

previously described (Rabaey et al., 2005)) supplemented with

2.5 mL L�1 of a solution of 0.4 M 1,2-DCA in isopropanol.

(Isopropanol was used to lower the volatilization of 1,2-DCA).

Thus the final concentration of 1,2-DCA in the medium was

1 mM (or 99 mg L�1). Five hundreds mLs of the 1,2-DCA con-

taining medium was circulated between an airtight reservoir

that had a headspace volume of 100 mL and the anode

compartment of each MFC at the flow rate of ca. 0.5 mL min�1

by a pump and through an airtight tubing system (Fig. 1B). To

investigate the removal of 1,2-DCA in relation to bio-

electrochemistry, one AceMFC was operated under closed

(electrical) circuit conditions and the other under open

(electrical) circuit conditions while other conditions were the

same for both.

The second type of MFCs (InoMFCs) included two newly

started reactors, each of which was inoculated with 10 mL of

a 1,2-DCA degrading sample from a natural site contaminated

with 1,2-DCA (Maes et al., 2006) in the anode compartment

when the experiments started. When the experiments star-

ted, the InoMFCs were operated with the 1,2-DCA containing

medium in a mode similar to that applied to the AceMFCs, as

described above. Also, one InoMFC was operated under closed

circuit conditions and the other under open circuit conditions

while other conditions were the same for both.

To investigate whether the current production was due to

the degradation of 1,2-DCA or not, a medium with similar

composition, including 2.5 mL L�1 of isopropanol, but not

containing 1,2-DCA, was used to feed the reactors at times of

interest.

A system that circulated the 1,2-DCA containing medium

between a reservoir and an autoclaved and uninoculated MFC

type reactor was set up and monitored as an additional

control to investigate if 1,2-DCA could escape from the

system, adsorb to the reactor materials or could be abiotically

degraded.

2.3. Bacterial community analysis and isolation

In the first month of operation with 1,2-DCA as the

electron donor, after the first, the second and the forth

week, bacterial cells on 1 cm2 of the electrode surface of

each of the MFCs were sampled and suspended in 2 mL of

100 mM phosphate buffer saline (PBS). For the extraction

of total DNA of the original acetate-fed consortium, the

corresponding bacterial suspension as such was used.

For the extraction of total DNA of an anodic community at a

certain experiment time, the mixture of 2.5 mL of

the medium in the reservoir and 1/10 of the corresponding

bacterial suspension from 1 cm2 electrode surface

was used.(The liquid medium volume to electrode surface

ratio was 500 mL/20 cm2 ¼ 2.5 mL/0.1 cm2). The latter was

done because the system was operated in a recirculative

mode, thus the planktonic cells growing in the medium

should also be taken into account. The original 1,2-

DCA degrading inoculum was used as such for DNA

extraction.

Total DNA of the aforementioned sample mixtures were

extracted using standard methods (Boon et al., 2000). 16S rRNA

gene fragments were amplified with the primers PRBA338fGC

and P518 (Muyzer et al., 1993) and analyzed using denaturing

gradient gel electrophoresis (DGGE) with a denaturing

gradient ranging from 45 to 60% (Boon et al., 2002). Based on

the DGGE patterns, bacterial communities of the MFCs were

analyzed in terms of dynamics (rate of composition change),

range-weighted richness and functional organization (Pareto–

Lorenz curves) using the methods proposed previously (Mar-

zorati et al., 2008). For more details about these methods, see

Supporting Material.

2.4. Analysis

1,2-DCA in the medium was measured by headspace gas

chromatography with a flame ionization detector (Chrompack

9002) as previously described (De Wildeman et al., 2003). The

headspace gas was sampled using a needle by injecting it

through the gastight rubber stopper of the medium container.

The same procedure was applied to detect and measure

ethene, ethane, vinyl chloride, 2-chloroethanol, ethylene

glycol, which are the potential intermediates and products

produced during the removal of 1,2-DCA. The concentration of

ethylene glycol was calculated using calibration curves

established from the measurement of standards made from

the pure compounds (purchased from Sigma–Aldrich). CO2

and CH4 produced in the headspace of a medium reservoir

was measured with an Intersmat IGC 120MB gas chromato-

graph connected to a Hewlett–Packard 3390A. Chloride (Cl�)

was determined using an ion chromatography (Compact IC

761 with conductivity detector, Metrohm, Switzerland) with

a metrosep A supp 5 column and a metrosep A 4/5 guard

column. Acetate was measured by gas chromatography

(Rabaey et al., 2003). Electrical parameters were measured and

calculated as previously described (Logan et al., 2006). All

analyses or measurements, unless otherwise explained, were

repeated three times. Data reported were taken from three

repetitions of each experiment (n ¼ 3).

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6 2939

3. Results

3.1. Current generation corresponding to 1,2-DCAremoval by the MFCs

In the first month of operation with 1,2-DCA as the main

substrate, the current generation by the closed circuit MFCs

started to increase after the first week (Fig. 2A), together with

the decrease of 1,2-DCA concentration in the medium (data

not shown). The current production reached a maximum level

in the third week (Fig. 2A). The maximum level of the current

generated by the AceMFC (containing a consortium previously

enriched with acetate) was 0.17 � 0.02 mA while that by the

InoMFC (inoculated with a sample from a natural site

contaminated with 1,2-DCA) were 0.10 � 0.03 mA (Fig. 2A).

The coulombic efficiency of the former was also higher than

that of the latter: 43 � 4% vs. 25 � 4%. After reaching the

Fig. 2 – (A): Typical patterns of current generation by the

MFCs during the first month of operation with 1,2-DCA as

the substrate (i.e. the adaptation period).(B): The changes of

the current generated by the AceMFC (after the anodic

consortium was stable) in response to the absence of 1,2-

DCA in the feeding medium. Dashed arrows indicate the

time points when the medium containing no 1,2-DCA was

fed to the systems. Solid arrows indicate the time points

when the medium containing 1,2-DCA was again fed to the

systems.

maximum level, if the AceMFC was operated with the medium

that had a similar composition, including 2.5 mL L�1 of iso-

propanol, but not containing 1,2-DCA, the current signifi-

cantly decreased to the background level of about 0.05 mA

(Fig. 2B). The InoMFC showed similar responses (data not

shown). This indicated that the current generation was due to

the presence of 1,2-DCA in the medium and that isopropanol

in the medium was not used by the anode bacteria as the main

electron donor.

3.2. Changes in bacterial communities of the MFCs

The most significant shifts of the bacterial communities in the

anodes of the MFCs occurred already after 1 week of operation

with 1,2-DCA as the substrate, as shown by the DGGE patterns

(Fig. 3), the changes of range-weighted richness (Rr) and

functional organization (Fo) values (Table 1, Fig. S1). During

the following weeks, smaller shifts continued to occur for

most of the communities, except for that of the open circuit

InoMFC. After 2 weeks, the changes slowed down for the

AceMFCs. The dynamic values (rates of change) of the AceMFC

communities were 5–6-fold higher than those of the InoMFC

ones: 48.3 vs. 8.9%/week (closed circuit conditions) and 47.3 vs.

9.3%/week (open circuit conditions).

Interestingly, when stability was achieved after the first

4 weeks, the community of the AceMFC operated under closed

circuits conditions was not significantly different from that

operated under open circuit conditions (Fig. 3). The commu-

nities indeed had changed at approximately equal rates (Table

1). However, in terms of richness and functional organization,

it could be observed that after the first week, the community

Fig. 3 – DGGE patterns of the 16S rRNA gene fragments of

the bacterial communities from the AceMFCs and the

InoMFCs at different time points during the first month of

operation with the 1,2-DCA containing medium (i.e. the

adaptation period). Meanings of sample abbreviations: A:

AceMFC; I: InoMFC; C: closed circuit; O: open circuit; to: at

the starting time of the experiment (1,2-DCA feeding

started); 1w: after 1 week; 2w: after 2 weeks; 4w: after

4 weeks.(Each sample was taken and analyzed in triplicate

and similar patterns were observed. Thus only one

triplicate of each sample was selected for this gel analysis).

Table 1 – Analyses of bacterial communities of the MFCs in terms of the rate of change, range-weighted richness andfunctional organization (after Marzorati et al. (2008)) during the first month of operation with 1,2-DCA containing medium(i.e. the adaptation period).

Community Rate of change (Dy) (%/week) Range-weighted richness (Rr) Functionality organization (Fo)(Gini index value*)

To 1w 2w 4w To 1w 2w 4w

AceMFC (closed ) 48.3 15.5 45.9 27.8 31.6 0.65 0.39 0.52 0.49

AceMFC (open) 47.3 15.5 38.5 21.6 25.3 0.65 0.49 0.52 0.51

InoMFC (closed ) 8.9 6.1 13.2 10.9 15.7 0.67 0.51 0.48 0.39

InoMFC (open) 9.3 6.1 6.0 2.4 1.6 0.67 0.63 0.68 0.65

*: the Gini index value reflects the degree of evenness of species distribution in a microbial community and is a value representing the area

between a given Pareto–Lorenz curve and the line of the perfect evenness. A perfectly even community corresponds to a Gini ¼ 0 (Lee, 1997;

Marzorati et al., 2008). Note: To: at the beginning of the experiments (1,2-DCA feeding started); 1w: after 1 week; 2w: after 2 weeks; 4w: after

4 weeks; closed: operated under closed circuit conditions; open: operated under open circuit conditions.

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 62940

of the closed circuit AceMFC changed more profoundly than

that of the open circuit AceMFC, becoming richer with more

even species distribution (Table 1, Fig. S1A). Specifically, the Rr

of the former increased from 15.5 to 45.9 while that of the

latter increased to 38.5; the Fo value of the former decreased

from 0.65 to 0.39 (i.e. the species within the community

became more evenly-distributed), while that of the latter only

decreased to 0.49. In these communities, dominant species of

the previous acetate-fed community became less common

while several previously less prominent species became

dominant (Fig. 3, Fig. S1A).

The anode bacterial community of the closed circuit

InoMFC also changed after the first week, becoming different

from that of the original inoculum and from those of the

AceMFCs (Fig. 3). However, no species tended to be dominant

after 4 weeks (Figs. 3 and S1B). The anode bacterial commu-

nity of the open circuit InoMFC remained unchanged and died

away. This can be observed from the decrease in intensity of

the DGGE bands (Fig. 3) and of the Rr values (Table 1).

3.3. 1,2-DCA removal by the MFCs

Hereafter, unless otherwise indicated, results are reported for

a batch of 1 month operation of the reactors after the micro-

bial consortia were stable (i.e. after the first month, which was

the adaptation period). Three replicates in each case repre-

sented 3 months of operation, during which the 1,2-DCA

containing medium was renewed after each month. The

AceMFC operated under closed circuit conditions already

removed more than 55% of 1,2-DCA after 1 week of operation

(Fig. 4B). Its highest removal rate of 102 mg 1,2-DCA L�1 NAC

d�1 was thus achieved in the first week. After 2 weeks, almost

95% of 1,2-DCA was removed by this closed circuit AceMFC.

After 4 weeks, the removal was 98%. The average 1,2-DCA

removal rate of this MFC in 1 month was 45.6 � 0.5 mg L�1

NAC d�1. The AceMFC operated under open circuit conditions

(after 1 month of enrichment) only removed about 25, 53 and

73% of 1,2-DCA after 1, 2 and 4 weeks of operation, respec-

tively. The average 1,2-DCA removal rate of the open circuit

AceMFC in 1 month was 34.0 � 1.5 mg L�1 NAC d�1.

Overall, the closed circuit InoMFC had a poorer perfor-

mance, compared to the closed circuit AceMFC. It removed 32,

85 and 95% of 1,2-DCA after 1, 2 and 4 weeks of operation,

respectively (Fig. 4B). The highest removal rate of 98 mg 1,2-

DCA L�1 NAC d�1 was achieved in the second week. The

average 1,2-DCA removal rate of this closed circuit InoMFC in

1 month was 44.1 � 0.2 mg L�1 NAC d�1. The open circuit

InoMFC almost could not remove 1,2-DCA (Fig. 4B).

The 1,2-DCA concentrations in the autoclaved and unin-

oculated MFC (control) did not change significantly (Fig. 4B),

indicating that 1,2-DCA did not escape from the system or

adsorb onto the materials and was not abiotically degraded.

It should be noticed that the production of current by the

MFCs corresponded to the 1,2-DCA removing capability of

the MFCs (Figs. 4A,B). Usually after 4 weeks of operation, the

currents decreased, probably due to the shortage of the main

electron donor (1,2-DCA).

3.4. Changes of the pH, chloride and CO2 concentrations

The amount of chloride produced was proportional to the

amount of 1,2-DCA removed (Fig. 4C, Table 2). After 4 weeks of

operation, the concentration of chloride produced in the

medium of the AceMFC reached 2 mM, which is the theoret-

ical maximum concentration of chloride that can be produced

from 1 mM 1,2-DCA. Chloride was produced at the highest

rates and levels in the closed circuit AceMFC (Fig. 4C).

CO2 was produced in the systems together with the

removal of 1,2-DCA (Table 2). Interestingly, the CO2 concen-

trations were the highest after 1 week of operation. CO2 was

produced the most in the closed circuit AceMFC (up to

0.22 mmol L�1 medium) and the second most in the closed

circuit InoMFC. Much less CO2 was produced in the reactors

operated under open circuit condition and no CO2 was

detected in the open circuit InoMFC. CO2 was not produced

when the MFCs were fed with the medium containing no 1,2-

DCA. Also, in the control (abiotic MFC), no CO2 was detected

(Table 2). These observations indicate that the production of

CO2 was a biological process related to the removal of 1,2-DCA.

The pH of the medium decreased in the closed circuit

AceMFC and closed circuit InoMFC (Table 2) although the

medium was buffered. For the AceMFC, the pH decrease was

the most pronounced, from 7.1 � 0.1 to 6.6 � 0.1 after 1 week

of operation, and to 6.5 � 0.1 after 2 weeks. The pH changes in

the InoMFC were lower but had a similar trend. Finally, the pH

changes were also less prominent in the AceMFC operated

Fig. 4 – The changes of average current (A), 1,2-DCA concentration in the medium (B) and chloride concentration in the

medium (C) during a month of operation of the MFCs with their stable anodic consortia. Control: the autoclaved and

uninoculated MFC operated under closed circuit conditions; closed circuit: operated under closed circuit conditions; open

circuit: operated under open circuit conditions.(Error bars were calculated based on three replicates representing 3 months

of operation; the 1,2-DCA containing medium was renewed after each month).

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6 2941

under open circuit conditions and null in the InoMFC operated

under open circuit conditions.

3.5. The production of potential intermediates andproducts

2-Chloroethanol, a typical intermediate in aerobic oxidation of

1,2-DCA (Nobre and Nobre, 2004), was not detected in the

circulated medium in any the experimental case. Ethene,

ethane and vinyl chloride, which are potential products and

intermediates of reductive dechlorination of 1,2-DCA (Nobre

and Nobre, 2004), were also not detected. Meanwhile, ethylene

glycol, an intermediate in anaerobic hydrolysis-based degra-

dation of 1,2-DCA (Nobre and Nobre, 2004), was found to

accumulate in the medium of the AceMFCs, both under closed

circuit and open circuit conditions (Table 2). Ethylene glycol

was less prevalent in the medium of the closed circuit AceMFC

after 1 and 2 weeks but reached 0.25 � 0.05 mM after 4 weeks,

when the currents significantly decreased. In the medium of

the open circuit AceMFC, ethylene glycol was already present

at a concentration of 0.2 � 0.06 mM after 2 weeks. Acetate was

also detected in the medium of the AceMFCs, both under

closed circuit and open circuit conditions, e.g. at 0.15 � 0.02

and 0.08 � 0.03 mM, respectively, after the first week (Table 2).

Table 2 – The values of the parameters monitored at different times during a month of operating the MFCs with the stable anodic consortia.

Reactors Time [1,2-DCA](mM)

[Cl� produced] (mM) CO2 produced (mmol L�1 medium) pH [EthylGly](mM)

[Acetate](mM)

CH4 produced(mmol L�1 medium)

Aver. I(mA)

CE (%)

Theo Real Theo Real

AceMFC (closed ) To 1.00 – – – – 7.1 � 0.1 0.015 � 0.005 – – –

1w 0.44 � 0.07 1.12 1.20 � 0.33 1.11 0.22 � 0.02 6.6 � 0.1 0.08 � 0.005 0.15 � 0.02 0.02 � 0.007 0.22 � 0.04

2w 0.05 � 0.004 1.90 1.77 � 0.27 1.90 0.14 � 0.02 6.5 � 0.1 0.12 � 0.02 0.22 � 0.03 0.03 � 0.01 0.17 � 0.02

4w 0.03 � 0.005 1.95 2.03 � 0.06 1.95 0.10 � 0.02 6.8 � 0.1 0.25 � 0.05 0.05 � 0.01 0.06 � 0.01 0.10 � 0.02 43 � 4

AceMFC (open) To 1.00 – – – – 7.1 � 0.1 0.02 � 0.004 – – n.a.

1w 0.74 � 0.13 0.52 0.42 � 0.07 0.52 0.08 � 0.01 6.8 � 0.1 0.05 � 0.005 0.08 � 0.03 0.11 � 0.02 n.a.

2w 0.47 � 0.05 1.06 1.05 � 0.32 1.06 0.07 � 0.01 6.9 � 0.1 0.25 � 0.06 0.13 � 0.01 0.23 � 0.04 n.a.

4w 0.27 � 0.03 1.46 1.15 � 0.34 1.46 0.02 � 0.01 6.9 � 0.1 0.41 � 0.07 0.02 � 0.01 0.31 � 0.05 n.a.

InoMFC (closed ) To 1.00 – – – – 7.1 � 0.1 0.015 � 0.005 – – –

1w 0.68 � 0.05 0.64 0.51 � 0.14 0.64 0.18 � 0.02 6.8 � 0.1 0.01 � 0.002 0.18 � 0.01 0.01 � 0.001 0.13 � 0.03

2w 0.15 � 0.007 1.70 1.58 � 0.18 1.70 0.10 � 0.02 6.8 � 0.1 0.01 � 0.001 0.03 � 0.01 0.17 � 0.07 0.12 � 0.02

4w 0.05 � 0.004 1.90 1.70 � 0.14 1.90 0.02 � 0.01 6.9 � 0.1 0.01 � 0.001 0.02 � 0.01 0.20 � 0.03 0.07 � 0.02 25 � 4

InoMFC (open) To 1.00 – – – – 7.1 � 0.1 0.01 � 0.001 – – n.a.

1w 0.99 � 0.03 0.02 w0 0.02 w0 7.0 � 0.2 0.01 � 0.002 w0 0.01 � 0.001 n.a.

2w 0.97 � 0.03 0.06 w0 0.06 w0 7.2 � 0.1 0.01 � 0.003 w0 0.01 � 0.001 n.a.

4w 0.84 � 0.04 0.32 w0 0.32 w0 7.1 � 0.1 0.01 � 0.002 w0 0.10 � 0.05 n.a.

Control To 1.00 – – – – 7.1 � 0.1 w0 – – w0

1w 0.99 � 0.03 w0 w0 w0 w0 7.1 � 0.2 w0 w0 w0 w0

2w 1.02 � 0.03 w0 w0 w0 w0 7.1 � 0.1 w0 w0 w0 w0

4w 0.98 � 0.04 w0 w0 w0 w0 7.1 � 0.1 w0 w0 w0 w0

n.a.: not applicable. Abbreviations: EthylGyl: ethylene glycol; Aver. I: average current; CE: coulombic efficiency (calculated for the whole period of operation: 4 weeks). Other notes: Real: real values;

Theo: theoretical values, calculated from the concentrations of 1,2-DCA removed; To: at the beginning of the operation; 1w: after 1 week; 2w: after 2 weeks; 4w: after 4 weeks; closed: operated under

closed circuit conditions; open: operated under open circuit conditions; Control: the autoclaved and uninoculated MFC operated under closed circuit conditions. The theoretical values and coulombic

efficiencies were calculated based on the assumption that 1,2-DCA is oxidized completely to CO2 (1 mole of 1,2-DCA produces 10 moles of electrons). The concentration of produced chloride was

calculated as the concentration of chloride in the medium at the time of interest subtracted by the chloride concentration in the original medium (110 mg L�1).

wa

te

rr

es

ea

rc

h4

3(2

00

9)

29

36

–2

94

62

94

2

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6 2943

The concentration of acetate was low in both cases after

4 weeks of operation.

In the cases of the InoMFCs, ethylene glycol was only

detected at the basic level (equal to that of the initial medium).

Acetate was detected only in the medium of the closed circuit

InoMFC, but at lower concentrations (Table 2). The acetate

concentration also tended to decrease during a course of a test

run (lowest after 4 weeks, when the currents significantly

decreased).

4. Discussion

4.1. Closed circuit operation vs. open circuit operation

The results showed that the closed circuit MFCs performed

better than their open circuit counterparts. For instance, the

closed circuit operation led to a 1.34-fold higher 1,2-DCA

removal rate in the AceMFCs. This indicated that the anode

activities accelerated the removal of 1,2-DCA. Being operated

under open circuit conditions means that the processes

occurring in the anodes are not electrochemical but plainly

aerobic oxidation or anaerobic fermentation/oxidation, as the

medium was anoxic (gastight). Possibly, due to the fact that

1,2-DCA is a recalcitrant compound (Bhatt et al., 2007), the

spontaneous metabolism of the bacteria on 1,2-DCA occurs at

a low rate. When an active electrode (anode) based with-

drawal of electrons was introduced, more pathways might be

activated and the metabolic rate increased.

4.2. Acetate feeding enriched anodophilic consortia vs.1,2-DCA feeding enriched anodophilic consortia

The consortia previously enriched with acetate functioned

better than the ones enriched with 1,2-DCA from a 1,2-DCA

removing inoculum, in terms of both current generation and

1,2-DCA removal. Even under open circuit conditions, the

former removed 75% of 1,2-DCA after 4 weeks while the latter

almost could not remove 1,2-DCA. The reason for this might be

the higher flexibility of the former, as they were richer in

species (Table 1). It is known that not only stability but also

flexibility are essential for a stable performance and swift

adaptation of a microbial community (Loreau et al., 2001;

Torsvik and Ovreas, 2002). The results showed that the

communities of the AceMFCs changed faster than those of

the InoMFCs in response to the feeding with 1,2-DCA (Table 1).

The higher metabolic versatility probably allows the AceMFC

communities to more easily shift to new compositions and

adapt to new conditions. The InoMFC communities might be

rather specific for their original bio-niche (as the species

evenness of the inoculum was relatively low – Fig. S1) and

appeared less adaptable to new conditions. The site from

which the inoculum of the InoMFCs was taken was treated

with the strain Desulfitobacterium dichloroeliminans DCA 1 and

subjected to reductive dechlorination (Maes et al., 2006). This is

probably another reason why the InoMFC communities were

less adaptable to the oxidative conditions in the MFC anode

than the AceMFC communities. However, the profound

changes of the community of the closed circuit InoMFC showed

that the MFC anode could stimulate this consortium to shift its

composition, survive and be able to metabolize 1,2-DCA.

4.3. Possible metabolism of 1,2-DCA by the anodophilicbacteria

As shown in Table 2, the measured concentration values of

chloride produced in the medium of the MFCs corresponded to

the theoretical values. This indicates complete dechlorination

of 1,2-DCA. The decreases of pH and the production of CO2

strongly suggested that the degradation of 1,2-DCA was an

oxidative process. The fact that potential products of reduc-

tive dechlorination of 1,2-DCA such as ethene, ethane and

vinyl chloride were not detected while acetate was produced

in the medium of the MFCs also corroborated this. Indeed, the

possibility for oxidative dechlorination in the anode is

eminent as the redox potential of the cathode oxidant, which

is hexacyanoferrate, is high (Eh� ¼ þ 436 mV vs. SHE as

mentioned above). The current generation also indicated that

an oxidative process must have occurred in the anodes of the

closed circuit MFCs.

This raises the question about the pathway employed by

the anodophilic bacteria for the oxidative dechlorination of

1,2-DCA. In this study, ethylene glycol and acetate, and not

chloroethanol, were detected in the medium of the MFCs,

showing that the anode bacteria might not employ an oxida-

tive pathway similar to that in aerobic oxidation of 1,2-DCA.

Indeed, a pathway that involves ethylene glycol and acetate

was previously reported for the degradation of 1,2-DCA under

anaerobic conditions (Nobre and Nobre, 2004; Stucki and

Thuer, 1995). In this process, 1,2-DCA is first hydrolyzed to

ethylene glycol, which is subsequently converted to acetate

and CO2 by acetogenic bacteria. This pathway fits the case of

the anodophilic bacteria, which are used to utilizing acetate as

their only substrate. Interestingly, the conversion of 1,2-DCA

to glycol was described as an abiotic process that occurs very

slowly (half life: 5.8 years) (Nobre and Nobre, 2004). In this

study, ethylene glycol kept accumulating in the medium of

the AceMFCs during a month of operation (Table 2). Therefore,

in our MFCs, probably this process was biotically accelerated

by the anodophilic bacteria. Another possibility is that the

bacteria kept consuming ethylene glycol to produce acetate

and consuming acetate to produce electrons; which might

cause the equilibrium of the abiotic hydrolytic reaction to

keep moving towards the production of ethylene glycol.

The fact that the current generation of the AceMFC

decreased in response to the decrease of acetate concentra-

tion (after 4 weeks) also strongly corroborates the hypothesis

that the oxidation of acetate (produced from ethylene glycol)

was a cause of the current generation. Moreover, ethylene

glycol concentrations were low in the medium of the closed

circuit AceMFC during the first and the second weeks (in an

1 month test run), when the currents were high, indicating

that this compound was also used for current generation. The

accumulation of ethylene glycol indicates that there might be

an ethylene glycol concentration limit for the conversion by

the anodophilic bacteria. Also, this was probably the reason

why the coulombic efficiency could not be higher.

Based on the discussion above, it can be concluded that the

hydrolysis pathway is the most possible explanation for the

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 62944

degradation of 1,2-DCA by the anodophilic bacteria in the

AceMFCs. It can be summarized as follows:

Step 1: Cl–CH2–CH2–Cl (1,2-DCA) þ 2H2O / CH2OH–CH2OH

(ethylene glycol) þ 2HCl

Step 2: 2CH2OH–CH2OH (ethylene glycol) þ 2H2O�������!Acetogenesis

CH3COOH (acetate) þ 2CO2 þ 12Hþ þ 12e�

Step 3: CH3COOH (acetate) þ 2H2O / 2CO2 þ 8Hþ þ 8e�

Overall, the electron balances at the beginning and at the

end of an 1 month test run of the closed circuit AceMFC

appeared to be relatively equal (Table S1). This suggests that

the above proposed pathway is probably the sole process in

the anode of the AceMFC under closed circuit conditions.

It can be noted that the concentration of CO2 produced was

decreased and not well correlated with 1,2-DCA removal and

that the total carbon balances of the substrate (1,2-DCA) and

of the analyzed products (ethylene glycol, acetate, CO2) were

not equal (Table S2). Our hypothesis is that part of the carbon

in 1,2-DCA was possibly used for building biomass. This

aspect requires further examination.

A similar pathway is probably applied for the degradation

of 1,2-DCA in the anode of the open circuit AceMFC, because

the same intermediates were detected. Indeed, this is

corroborated by the insignificant difference in composition

between the stable microbial community of the closed circuit

AceMFC and that of the open circuit AceMFC. However, the

data indicated that the conversion rate under open circuit

conditions might be significantly lower than under closed

circuit conditions. In addition, instead of the generation of

electrons and protons, anaerobic digestion products might be

produced. Indeed, methane was detected from the open

circuit AceMFC, more than from the closed circuit AceMFC

(Table 2). The electron and carbon balance data for the open

circuit AceMFC (with methane also considered) (Tables S1 and

S2), showing high percentages of recovery (>90%), corroborate

this hypothesis.

For the closed circuit InoMFC, acetate and CO2 were also

detected in its medium, indicating that also in this case, the

degradation of 1,2-DCA might be an oxidative process.

However, due to the fact that no other intermediates could be

detected and that the electron and carbon balances at the

beginning and at the end of each operational batch were

significantly unequal (Tables S1 and S2), the possible mecha-

nisms for the degradation remain unclear. As the InoMFCs

performed more poorly than the AceMFCs, we decided not to

further investigate these mechanisms.

4.4. The potential use of MFCs for the removal of1,2-DCA

In this study, the maximum 1,2-DCA removal rate of 102 mg

L�1 reactor volume d�1 and the average removal rate of

45.6 mg L�1 reactor volume d�1 were achieved with the

AceMFC operated under closed circuit condition. These values

are comparable to those reported for a 1,2-DCA oxidizing

mixed culture using nitrate as the electron acceptor (Table S3).

Although this value is still lower than those reported for pure

cultures (Table S3), the removal efficiency reached >99% after

1 month for the AceMFC. One reason for the low removal rates

of the MFCs might be the limited mass transfer due to the

restricted contact surface and the low hydraulic mixing of the

medium in the system. Another reason might be the low

intrinsic rate of the hydrolytic conversion, as mentioned

above, which the anodophilic bacteria employ for the

metabolism of 1,2-DCA. Yet, as discussed above, the use of the

MFC anode significantly improved this rate. Pure cultures,

though reported to have high rates of 1,2-DCA removal (De

Wildeman et al., 2003; Hage and Hartmans, 1999; Janssen

et al., 1985), require specific environmental conditions.

Moreover, contamination is a significant problem decreasing

the performance of the reactors operating with pure cultures.

Anaerobic treatment (Dinglasan-Panlilio et al., 2006; Gupta

and Mali, 2008; Maes et al., 2006) for 1,2-DCA removal repre-

sent other options but these still require special care and

operation to maintain anaerobic conditions. Meanwhile, the

operation of MFCs is relatively simple, no special care is

required, and they can be used to recover part of the energy

derived from the oxidation of 1,2-DCA, transforming it into

electricity. Assuming that 1,2-DCA could be oxidized

completely to CO2, the coulombic efficiency of the AceMFCs

was about 43% (Table 2), which is indeed competitive with

respect to using MFCs for removing pollutants. In addition, in

this study, it was shown that using a gastight MFC system, 1,2-

DCA volatilization problems could be resolved. The fact that

the AceMFCs functioned better than the InoMFCs suggested

that there is no need of a special inoculum to enrich a 1,2-DCA

removing anodic consortium, which is convenient for prac-

tical application.

The remaining challenge for the use of MFCs might be how

to improve the removal rate of the anode consortia to the

levels comparable with those reported for practice (at gram

per liter reactor per day levels). Definitely, the mass transfer of

1,2-DCA to the anodophilic degraders will need optimizing.

Further studies of the 1,2-DCA metabolizing pathway(s)

employed by these bacteria in order to improve the removal

rate are also warranted.

5. Conclusion

In this study, we have shown that:

– 1,2-DCA, a well-known recalcitrant chlorinated ground-

water pollutant, was efficiently degraded by anodophilic

bacteria enriched in microbial fuel cells, at the rate of up to

102 mg per liter reactor volume per day. Concomitantly,

energy released from this degradation could be partially

recovered (up to 43%) as electricity.

– The flexibility and the anodic pre-acclimation of a microbial

community is probably more important than its substrate

pre-acclimation in determining its 1,2-DCA removing

performance in a MFC anode.

– A hydrolysis-based oxidative pathway is possibly employed

by anodophilic bacteria for the degradation of 1,2-DCA.

– The metabolic flexibility of anodophilic consortia, the high

1,2-DCA removal efficiency and the simple operation of

MFCs offer a new potential application of MFCs in the

domain of bioremediation.

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 6 2945

Acknowledgement

This research was supported by a grant from the Flanders

Research Foundation (FWO project G.0172.05) and an EU

Neptune project (Contract No 036845, SUSTDEV-2005-3.II.3.2).

The authors would like to deeply thank Peter Clauwaert for

critically reading the manuscript and giving useful comments.

Appendix A.Supplemental material

Supplementary information for this manuscript can be

downloaded at doi:10.1016/j.watres.2009.04.004.

r e f e r e n c e s

Aelterman, P., et al., 2006. Continuous electricity generationat high voltages and currents using stacked microbialfuel cells. Environmental Science & Technology 40 (10),3388–3394.

Allen, R.M., Bennetto, H.P., 1993. Microbial fuel cells – electricityproduction from carbohydrates. Applied Biochemistry andBiotechnology 39, 27–40.

Aulenta, F., et al., 2007. Electron transfer from a solid-stateelectrode assisted by methyl viologen sustains efficientmicrobial reductive dechlorination of TCE. EnvironmentalScience & Technology 41 (7), 2554–2559.

Aulenta, F., et al., 2008. Trichloroethene dechlorination and H-2evolution are alternative biological pathways of electriccharge utilization by a dechlorinating culture ina bioelectrochemical system. Environmental Science &Technology 42 (16), 6185–6190.

Bhatt, P., et al., 2007. Biodegradation of chlorinated compounds –a review. Critical Reviews in Environmental Science andTechnology 37 (2), 165–198.

Boon, N., et al., 2000. Bioaugmentation of activated sludge by anindigenous 3-chloroaniline degrading Comamonas testosteronistrain, I2gfp. Applied and Environmental Microbiology 66,2906–2913.

Boon, N., et al., 2002. Evaluation of nested PCR-DGGE (denaturinggradient gel electrophoresis) with group-specific 16S rRNAprimers for the analysis of bacterial communities fromdifferent wastewater treatment plants. Fems MicrobiologyEcology 39 (2), 101–112.

Clauwaert, P., et al., 2008. Minimizing losses in bio-electrochemical systems: the road to applications. AppliedMicrobiology and Biotechnology 79 (6), 901–913.

De Wildeman, S., Verstraete, W., 2003. The quest for microbialreductive dechlorinationof C-2 toC-4chloroalkanes iswarranted.Applied Microbiology and Biotechnology 61 (2), 94–102.

De Wildeman, S., et al., 2003. Growth-substrate dependentdechlorination of 1,2-dichloroethane by a homoacetogenicbacterium. Biodegradation 14 (4), 241–247.

Dinglasan-Panlilio, M.J., et al., 2006. Microbial oxidation of 1,2-dichloroethane under anoxic conditions with nitrate aselectron acceptor in mixed and pure cultures. FEMSMicrobiology Ecology 56 (3), 355–364.

Egli, C., et al., 1987. Anaerobic dechlorination oftetrachloromethane and 1,2-dichloroethane to degradableproducts by pure cultures of Desulfobacterium sp and

Methanobacterium sp. FEMS Microbiology Letters 43 (3),257–261.

Gordon, S., et al., 2002. A laboratory technique for investigation ofdiffusion and transformation of volatile organic compounds inlow permeability media. Journal of Contaminant Hydrology 57(3–4), 223–240.

Gupta, S.K., Mali, S.C., 2008. Reductive dechlorination of1,2-dichloroethane using anaerobic sequencing batchreactor (ASBR). Water Science and Technology 57 (2), 225–229.

Hage, J.C., Hartmans, S., 1999. Monooxygenase-mediated1,2-dichloroethane degradation by Pseudomonas sp strainDCA1. Applied and Environmental Microbiology 65 (6),2466–2470.

He, J.Z., et al., 2003. Detoxification of vinyl chloride to ethenecoupled to growth of an anaerobic bacterium. Nature 424(6944), 62–65.

Hughes, K., et al., 1994. 1,2-dichloroethane – evaluation of risks tohealth from environmental exposure in Canada.Environmental Carcinogenesis & Ecotoxicology Reviews-Part C of Journal of Environmental Science and Health 12 (2),293–303.

Janssen, D.B., et al., 1984. Biodegradation of 2-chloroethanol and1,2-dichloroethane by pure bacterial cultures. InvertebrateBiotechnology, 169–178.

Janssen, D.B., et al., 1985. Degradation of halogenated aliphaticcompounds by Xanthobacter autotrophicus Gj10. Applied andEnvironmental Microbiology 49 (3), 673–677.

Janssen, D.B., et al., 1994. Genetics and biochemistry of1,2-dichloroethane degradation. Biodegradation 5, 249–257.

Lee, W.C., 1997. Characterizing exposure-disease association inhuman populations using the Lorenz curve and Gini index.Statistics in Medicine 16 (7), 729–739.

Logan, B.E., et al., 2006. Microbial fuel cells: methodology andtechnology. Environmental Science & Technology 40 (17),5181–5192.

Loreau, M., et al., 2001. Ecology – biodiversity and ecosystemfunctioning: current knowledge and future challenges.Science 294 (5543), 804–808.

Maes, A., et al., 2006. Transport and activity of Desulfitobacteriumdichloroeliminans strain DCA1 during bioaugmentation of 1,2-DCA-contaminated groundwater. Environmental Science &Technology 40 (17), 5544–5552.

Marzorati, M., et al., 2006. Response of 1,2-dichloroethane-adapted microbial communities to ex-situ biostimulation ofpolluted groundwater. Biodegradation 17 (2), 41–56.

Marzorati, M., et al., 2007. A novel reductivedehalogenase, identified in a contaminated groundwaterenrichment culture and in Desulfitobacteriumdichloroeliminans strain DCA1 is linked to dehalogenation of1,2-dichloroethane. Applied and EnvironmentalMicrobiology 73 (9), 2990–2999.

Marzorati, M., et al., 2008. How to get more out of molecularfingerprints: practical tools for microbial ecology.Environmental Microbiology 10 (6), 1571–1581.

Maymo-Gatell, X., et al., 1999. Reductive dechlorination ofchlorinated ethenes and 1,2-dichloroethane by‘‘Dehalococcoides ethenogenes’’ 195. Applied and EnvironmentalMicrobiology 65 (7), 3108–3113.

Muyzer, G., et al., 1993. Profiling of complex microbialpopulations using denaturing gradient gel electrophoresisanalysis of polymerase chain reaction-amplified genes codingfor 16S rRNA. Applied and Environmental Microbiology 59,695–700.

Nobre, R.C.M., Nobre, M.M.M., 2004. Natural attenuation ofchlorinated organics in a shallow sand aquifer. Journal ofHazardous Materials 110 (1–3), 129–137.

Oldenhuis, R., et al., 1989. Degradation of chlorinated aliphatic-hydrocarbons by Methylosinus trichosporium Ob3b expressing

w a t e r r e s e a r c h 4 3 ( 2 0 0 9 ) 2 9 3 6 – 2 9 4 62946

soluble methane monooxygenase. Applied and EnvironmentalMicrobiology 55 (11), 2819–2826.

Rabaey, K., et al., 2003. A microbial fuel cell capable of convertingglucose to electricity at high rate and efficiency. BiotechnologyLetters 25 (18), 1531–1535.

Rabaey, K., et al., 2005. Continuous microbial fuel cellsconvert carbohydrates to electricity. Water Science andTechnology 52 (1–2), 515–523.

Strycharz, S.M., et al., 2008. Graphite electrode as a sole electrondonor for reductive dechlorination of tetrachloretheneby Geobacter lovleyi. Applied and Environmental Microbiology74 (19), 5943–5947.

Stucki, G., Thuer, M., 1995. Experiences of a large-scaleapplication of 1,2-dichloroethane degrading microorganismsfor groundwater treatment. Environmental Science &Technology 29 (9), 2339–2345.

Stucki, G., et al., 1983. Bacterial growth on 1,2-dichloroethane.Experientia 39 (11), 1271–1273.

Torsvik, V., Ovreas, L., 2002. Microbial diversity and function insoil: from genes to ecosystems. Current Opinion inMicrobiology 5 (3), 240–245.

Vandenwijngaard, A.J., et al., 1992. Degradation of 1,2-dichloroethaneby Ancylobacter aquaticus and other facultative methylotrophs.Applied and Environmental Microbiology 58 (3), 976–983.