Microbial Community Composition Is Unaffected by Anode PotentialXiuping Zhu,† Matthew D. Yates,† Marta C. Hatzell,† Hari Ananda Rao,‡ Pascal E. Saikaly,‡

and Bruce E. Logan*,†

†Department of Civil and Environmental Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802,United States‡Water Desalination and Reuse Center, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, SaudiArabia

*S Supporting Information

ABSTRACT: There is great controversy on how different set anodepotentials affect the performance of a bioelectrochemical system (BES).It is often reported that more positive potentials improve acclimationand performance of exoelectrogenic biofilms, and alter microbialcommunity structure, while in other studies relatively more negativepotentials were needed to achieve higher current densities. To addressthis issue, the biomass, electroactivity, and community structure ofanodic biofilms were examined over a wide range of set anodepotentials (−0.25, −0.09, 0.21, 0.51, and 0.81 V vs a standard hydrogenelectrode, SHE) in single-chamber microbial electrolysis cells.Maximum currents produced using a wastewater inoculum increasedwith anode potentials in the range of −0.25 to 0.21 V, but decreased at0.51 and 0.81 V. The maximum currents were positively correlated withincreasing biofilm biomass. Pyrosequencing indicated biofilm commun-ities were all similar and dominated by bacteria most similar to Geobacter sulfurreducens. Differences in anode performance withvarious set potentials suggest that the exoelectrogenic communities self-regulate their exocellular electron transfer pathways toadapt to different anode potentials.

■ INTRODUCTIONVarious bioelectrochemical systems (BESs), such as microbialfuel cells (MFCs) and microbial electrolysis cells (MECs), havebeen developed to extract energy from wastewaters in differentforms including electricity and hydrogen.1−4 Microorganismson the anode oxidize organics to CO2, releasing protons intosolution and electrons to the circuit. In MFCs, oxygen isreduced at the cathode to water, consuming protons andelectrons (e.g., O2 + 4H+ + 4e− ⇌ 2H2O, E° = 1.229 V vsSHE) and producing electrical power.1,5 By application ofadditional voltage of 0.3−1 V to that produced by the anode,hydrogen can be generated in MECs at the cathode (e.g., H+ +2e− ⇌ H2, E° = 0 V vs SHE).2,6 This voltage can be producedby an external power source, or through insertion of a reverseelectrodialysis stack between the electrodes.7,8

The anode potential can influence the performance of BESs.9

Theoretically, microorganisms can gain more energy for growthat higher anode potentials if metabolic pathways are available tocapture this energy, as shown by9−11

Δ °′ = − − °′G nF E E( )anode substrate (1)

where ΔG°′ (J/mol) is the Gibbs free energy at standardbiological conditions (T = 25 °C, pH = 7), n is the number ofelectrons transferred, F is the Faradays constant (96,485 C/mole−), and Eanode (V) and Esubstrate°′ (V) are the anode potential and

the standard biological redox potential of the substrate. Despitethe availability of greater energy with higher anode potentials, areview of the literature on the effects of set anode potentials onBESs performance revealed no clear effect of anode potentialon performance.9 Several studies have shown that improvedperformance (higher current) was obtained at more positivepotentials, in the potential ranges of −0.09−0.41,12 0.15−0.83V,13 or 0.30−0.80 V14 (vs SHE). In other studies, currentincreased with anode potential in a lower (more negative)potential range, but it did not further increase when potentialswere increased to 010 or 0.21 V.15 In some cases, currentdecreased at potentials higher than 0 or 0.41 V.17 Two studieshave shown the opposite trend in current and set potential, asthe highest current densities were produced at the lowestanodes potentials of −0.15 (compared to −0.09, 0.02, and 0.37V)18 and 0.54 V (relative to 0.74 and 0.94 V).19

The lack of agreement in studies with set potentials could bedue to a combination of factors, but the most important factoris the predominance of specific microorganisms in the anodecommunity. High power densities are usually associated with

Received: October 21, 2013Revised: December 10, 2013Accepted: December 23, 2013Published: December 23, 2013

Article

pubs.acs.org/est

© 2013 American Chemical Society 1352 dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−1358

communities dominated by various Geobacter species.20 Severalstudies have shown that current generation by Geobactersulfurreducens was positively correlated with biomass.10,21 It hasalso been shown using cyclic voltammetery (CV) that theexpressed cytochromes, based on predominant peaks in theCVs, are altered with set potentials, suggesting that G.sulfurreducens self-regulates their extracellular electron transportpathways to adapt to different anode potentials.10,14 While theinfluence of anode potential on the current and powergeneration has been well studied in BESs using mixedcultures,12,16−19 there are few comprehensive investigationson how biofilm community structure and biomass on the anodeare influenced by different set potentials. In one of the mostcomprehensive examinations of set potentials in MECs, Torreset al.18 showed that low set potentials of −0.15, −0.09, and 0.02V selectively enriched G. sulfurreducens (92−99% of clonesbased on 16S rDNA clone library) with the highest currentdensity produced at −0.15 V, but a higher set potential of 0.37V resulted in a very diverse microbial community that hadlowest current production. This finding of lower currentgeneration at higher potential (0.02 V) is surprising given thatG. sulfurreducens has been shown to produce higher currentdensities at potentials of 0 V relative to −0.16 V by others.10,15

One unusual aspect of the study by Torres et al. was that asingle reactor was inoculated with four electrodes all set atdifferent potentials in the same anode chamber.18 Placement ofelectrodes next to each other in the same vessel will result in“crosstalk” between these electrodes. Electric field gradientswould exist between these electrodes and influence themigration of charged ions, substrate, and bacteria. Mediatorsproduced on one electrode could be used by biofilms growingon other electrodes. Thus, the effect of set anode potentialmight not have been adequately addressed in their study.To more comprehensively examine the effects of set anode

potential, MECs (in duplicate) were inoculated and operated atdifferent set potentials of −0.25, −0.09, 0.21, 0.51, and 0.81 Vvs SHE using a potentiostat. When current generation becamestable and the reactors were operated for more than onemonth, cyclic voltammetry (CV) was performed to characterizethe electroactivity of the biofilms, the microbial community wasanalyzed by pyrosequencing, and total biomass was measuredto determine the extent of bacterial growth at each set potential.

■ MATERIALS AND METHODSReactor Configuration and Operation. Single-chamber

MECs were constructed using 5 mL clear glass serum bottles.22

Graphite plates 1 cm × 1.5 cm (GM-10, 0.32 cm thick) wereused as anodes. Prior to use, they were polished usingsandpaper (grit type 400), sonicated to remove debris, cleanedby soaking in 1 M HCl overnight, and rinsed three times withMilli-Q water. Stainless steel (SS) mesh (Type 304, mesh size90 × 90) with the same size was used as the cathode. Thedistance between the anode and the cathode was ∼1 cm, andAg/AgCl reference electrodes (+210 mV vs SHE) wereinserted between the two electrodes. Bottles were sealedusing thick butyl rubber stoppers (20 mm diameter) andaluminum crimp caps. The electrode wires for the anodes (Ti)and cathodes (SS) as well as the reference electrode wereinserted through the rubber stopper. All potentials werereported here versus SHE for comparisons to other studies.The reactors (duplicates) were sparged with CO2/N2

(20%:80%) gas, and then inoculated using 1 mL of primaryclarifier effluent collected from the Pennsylvania State

University wastewater treatment plant, and 4 mL of growthmedium sparged and maintained under a CO2/N2 (20%:80%)atmosphere. The growth medium contained (per L): 0.82 g ofsodium acetate, 0.6 g of NaH2PO4, 1.5 g of NH4Cl, 0.1 g ofKCl, 2.5 g of NaHCO3, 10 mL of minerals, and 10 mL ofvitamins (pH 6.8). The MECs were incubated at five differentanode potentials (−0.25, −0.09, 0.21, 0.51, and 0.81 V) setusing a potentiostat (Uniscan PG580RM) in a temperaturecontrolled room (30 °C). Solution in the MECs was replacedwith 5 mL of fresh growth medium every cycle by piercing thebutyl stopper with needles and syringes.

Biomass Measurement. Biomass on the anode wasmeasured using a bicinchoninic acid protein assay kit(Sigma). Protein was extracted from the electrodes with 0.2N NaOH.23 The graphite anode was removed from the MECs,placed in a test tube containing 2 mL of 0.2 N NaOH at 4 °Cfor 1 h, and vortexed every 15 min for 10 s. The extractedsolution was collected, and the electrode was further rinsedwith 2 mL of deionized water. The liquids were mixed together(final concentration of 0.1 N NaOH), frozen at −20 °C, andthawed at 90 °C. This freeze−thaw cycle was conducted threetimes. Then, 0.1 mL of the sample was used for proteinanalysis.

Cyclic Voltammetry. Cyclic voltammetry (CV) wasconducted on the anode of MECs with the cathode as thecounter electrode, and an Ag/AgCl reference electrode, in freshgrowth medium. Scans ranged from −0.39 V to +0.91 V at arate of 1 mV/s with only the fourth stable cycle shown.Nonturnover CVs were performed in depleted sodium acetategrowth medium (when the current in MECs decreased toalmost 0 mA).

Microbial Community Analysis. The anodic biofilmswere removed from the graphite blocks with sterile pipet tips.The DNA was extracted using the PowerSoil DNA kit (MoBio,CA) according to manufacturer’s instructions. DNA sampleswere sent to Research and Testing Laboratory (www.reasearchandtesting.com) for pyrosequencing. Samples wereamplified for pyrosequencing using a forward primerconstructed with (5′−3′) Roche A linker, an 8−10 bp barcode,and the 28F primer (GAGTTTGATCNTGGCTCAG).24 Thereverse fusion primer was constructed (5′−3′) with a biotinmolecule, the Roche B linker, and the 519R primer (GTN-TTACNGCGGCKGCTG).24 Amplifications were performedin 25 μL reactors under the following thermal profile: 95 °C for5 min, then 35 cycles of 94 °C for 30 s, 54 °C for 40 s, 72 °Cfor 1 min, followed by one cycle of 72 °C for 10 min and thenheld at 4 °C.Amplification products were visualized, cleaned, and size

selected following Roche 454 protocols (454 Life Sciences,Branford, Connecticut). DNA was hybridized to Dynabeads M-270 (Life Technologies) to create single stranded DNAfollowing Roche 454. Single-stranded DNA was diluted, usedin emPCR reactions, and sequenced following establishedmanufacturer protocols (454 Life Sciences).Sequences were processed for quality using the Mothur

(version 1.3)25 software package. Briefly, sequences weredenoised, screened for barcode and primer mismatches,homopolymeric runs, and sequence length. Chimeric sequenceswere detected and removed using the Uchime ChimeraDetection Tool.26 Approximately 6000 (1100 per sample)sequences passed through quality control steps and were usedfor further study. High quality sequences were aligned, groupedinto OTUs (97% similarity), and classified to the phylum and

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581353

genus level using the Silva 108 nonredundant database as thereference taxonomy. Dominant OTUs were further classified atthe species level using the BLAST database. Sequences weredeposited into the NCBI SRA under accession numbersSRP031498. The Jaccard similarity index was used to calculatethe similarity between the bacterial communities.

■ RESULTSCurrent Generation. The currents produced in MECs with

different anode potentials were examined when repeatablecycles appeared after one month using fresh medium (Figure1A). The maximum current and cycle time varied for anodes

with different potentials, but the recovered Coulombs per cyclewere similar (35−39 C) as shown by high Coulombicefficiencies of 90−100% based on the Coulombs contained inadded sodium acetate (39 C). The longer cycle times foranodes set at potentials of −0.25 and 0.81 V indicated muchslower substrate utilization rates of these anodic biofilmscompared to biofilms acclimated at other anode potentials(−0.09, 0.21, and 0.51 V). The maximum current first increasedwith anode potential over the range of −0.25−0.21 V, but thendecreased slightly at 0.51 V and more at 0.81 V (Figure 1B),with values of 0.44 ± 0.05 (−0.25 V), 1.73 ± 0.34 (−0.09 V),2.47 ± 0.07 (0.21 V), 2.22 ± 0.14 (0.51 V), and 0.53 ± 0.21mA (0.81 V). The increase in maximum current with setpotentials up to 0.21 V suggested that the microorganismsbenefitted from the greater differences between the anodepotential and the standard biological redox potential of thesubstrate, that is, a more negative ΔG° (eq 1). A decrease inmaximum current at more positive potentials of 0.51 and 0.81V might indicate the use of less effective metabolic pathways,

changes in the biofilm community, or the induction of chemicalreactions harmful to the bacteria. Additionally, the bell shapedresponse (Figure 1B) might be an example of the “invertedregion” predicted by Marcus theory. According to this theory,the reaction rates usually become higher with increasingexergonicity of the reaction (more negative ΔG°), whereaselectron transfer should become slower in the very negativeΔG° domain because of increasing activation energy.27−29

Biomass measurements were consistent with maximumcurrents as higher currents were associated with greater biofilmmasses (Figure 2). The mass of biofilms (μg of protein)

measured on the anodes at different potentials were 6 ± 1(−0.25 V), 294 ± 20 (−0.09 V), 737 ± 70 (0.21 V), 528 ± 42(0.51 V), and 41 ± 1 μg/cm2 (0.81 V). According to previousmeasurements,21,30 a monolayer of G. sulfurreducens containedabout 20 μg/cm2 of protein. This suggests that the biofilm onthe anode set at −0.25 V was sparse and did not completelycover the anode. For the other cases, there could have beencomplete coverage with two layers for the biofilm developed at0.81 V. Much thicker biofilms would have been present on theother anodes. If we assume a minimum thickness of 1 μm perlayer (the approximate diameter of a cell), the biofilms wouldhave been 15 (−0.09 V), 37 (0.21 V), and 26 μm (0.51 V) onthese anodes, which are similar to those previouslyreported.21,30,31 The ratios of current:protein were 48 (−0.25V), 4 (−0.09 V), 2 (0.21 V), 3 (0.51 V), and 9 μA/μg (0.81 V).These ratios suggest that the electron transfer rate by cells wasfaster in thinner biofilms, especially for monolayer. It isreasonable that some cells in the biofilm did not facilitateefficient electron transfer especially in the thicker biofilms.Biomass increased with anode potentials in the range of −0.25to 0.21 V and resulted in greater current, indicating moreenergy was gained for bacterial growth as anode potentials wereincreased over this range. However, biomass decreased whenthe potential was increased from 0.21 to 0.81 V, indicating lessfavorable conditions for biomass growth at these highpotentials.

Electrochemical Analysis. CVs were performed toexamine the electroactivity of biofilms acclimated at differentanode potentials in fresh growth medium (Figure 3A). Allbiofilms started producing currents at −0.3 to −0.2 V, and thencurrents rose with increasing potentials and became stable inhigh potential region. The stabilized peak currents varied forbiofilms acclimated at different potentials, with values of 0.29 ±

Figure 1. (A) Current evolution of MECs developed at differentanode potentials over time for a single cycle, and (B) the maximumcurrent of MECs acclimated at different anode potentials (−0.25,−0.09, 0.21, 0.51, and 0.81 V vs SHE) after stable operation for overone month. Growth medium: 0.82 g/L sodium acetate in bicarbonatebuffer with pH of 6.8.

Figure 2. The relationship between the maximum currents of MECsdeveloped at different anode potentials (−0.25, −0.09, 0.21, 0.51, and0.81 V vs SHE) and the biomasses of anodic biofilms after stableoperation for over one month. Growth medium: 0.82 g/L sodiumacetate in bicarbonate buffer with pH of 6.8.

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581354

0.09 (−0.25 V), 1.62 ± 0.06 (−0.09 V), 2.09 ± 0.05 (0.21 V),2.01 ± 0.08 (0.51 V), and 0.25 ± 0.05 mA (0.81 V). Thebiofilm acclimated at 0.21 V showed the highest electroactivity,followed by biofilms developed at 0.51, −0.09, −0.25, and 0.81V, which was consistent with the maximum currents producedin MECs at these different potentials. The current generation inthe high potential region could be limited by mass transfer.There is no diffusional current decay in these CVs,32,33

indicating that sodium acetate is unlikely depleted whenadded at a concentration of 0.82 g/L, and therefore thedifferent steady-state currents likely reflect transport of redoxspecies through these biofilms.To examine the exocellular electron transfer components

(EETCs) in biofilms with these different set potentials,nonturnover CVs were performed in depleted sodium acetategrowth medium (current decreased to nearly 0 mA) (Figure3B). Four obvious oxidation peaks were observed at −0.18 ±0.02 (P1), −0.11 ± 0.02 (P2), 0.12 ± 0.01 (P3), and 0.77 ±0.02 (P4). The potential of P1 was close to the midpointpotentials reported for multiheme cytochrome OmcZ (−0.22V), OmcB (−0.19 V), and the periplasmic cytochrome C(PpcA, −0.17 V) purified from G. sulfurreducens.34−36 The P2,P3, and P4 have been reported in previous studies, but theEETCs that they represent are still unknown.14,15 Thesenonturnover CV results showed that different predominantEETCs were expressed when the biofilms were acclimated atdifferent potentials. For the biofilms acclimated at lowpotentials (−0.25 and −0.09 V), peaks P1 and P2 werepredominant (Figure 3B), which was consistent with the fastcurrent increase of these biofilms in the potential range of −0.3to −0.1 V in fresh growth medium (Figure 3A). An additionalpeak of P3 was observed for the biofilm incubated at 0.21 V(Figure 3B), resulting in a great increase in current around a

potential of 0.1 V for this biofilm in the presence of substrate(Figure 3A). When anode potential further increased to 0.51 V,the peak P3 disappeared and P4 appeared (Figure 3B),consistent with the slow current increase of this biofilm in thepotential range of 0.1 to 0.7 V in fresh growth medium (Figure3A). For the biofilm acclimated at 0.81 V, peaks P1 and P2 alsodisappeared and P4 became apparent (Figure 3B), resulting in avery slow current increase over the whole potential range of−0.3 to 0.7 V with fresh growth medium (Figure 3A). Thechanges of the EETCs predominance for the biofilmsacclimated at different potentials demonstrated either bacteriaaltered their extracellular electron transfer pathways to adapt todifferent anode potentials, or that different bacteria becamepredominant in the biofilm that expressed different EETCs.

Microbial Community Analysis. The microbial commun-ities of biofilms developed at different anode potentials were 79± 8% similar based on OTUs (defined as >97% sequencesimilarity) shared between samples, with high percentages ofthe sequences indicating dominance of Proteobacteria: 85%(−0.25 V), 78% (−0.09 V), 78% (0.21 V), 67% (0.51 V), and80% (0.81 V) (Figure 4A). The Proteobacteria in each reactor

were all most similar to G. sulfurreducens (94%−99% ofProteobacterial sequences, with a 97% or higher similarity)(Figure 4B). G. sulfurreducens was dominant in all samples withan average percentage of 75 ± 6%. The lack of a change inbacterial community composition suggests that the changes inthe EETCs noted in CVs were not due to changes incommunity. This conclusion is supported by pure culturestudies using G. sulfurreducens which shows that differentEETCs, at potentials similar to those here, similarly evolve withchanges in anode potentials.15

Figure 3. Cyclic voltammograms of the biofilms developed at differentanode potentials (−0.25, −0.09, 0.21, 0.51, and 0.81 V vs SHE) (A) infresh growth medium with 0.82 g/L sodium acetate and (B) indepleted sodium acetate growth medium with scan rates of 1 mV/safter stable operation for over one month.

Figure 4. (A) Microbial community distribution for biofilmsdeveloped at different anode potentials (−0.25, −0.09, 0.21, 0.51,and 0.81 V vs SHE) at phylum level and (B) the composition ofphylum Proteobacteria for biofilms developed at different anodepotentials at the genus level.

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581355

■ DISCUSSION

These results conclusively demonstrate that set anode potentialdoes not alter the structure of the microbial community whenthese potentials are set in separate, single-chamber MECs. Thecommunities were predominantly Proteobacteria at all thedifferent anode potentials, with sequences most similar to G.sufurreducens. This similarity of the communities is somewhatdifferent than that reported by Torres et al.18 in their tests overa smaller range of potentials. Although Torres et al. similarlyreported that lower anode potentials of −0.15, −0.09, and 0.02V produced communities that were most similar (>97%similarity for 92−99% of clones) to G. sulfurreducens, a higheranode potential of 0.37 V produced a very diverse communitywith a predominance of the phyla Bacteriodetes, Actinobacteria,Firmicutes, and Proteobacteira. We suggest that their results mayhave been affected by placing the four electrodes set at differentpotentials in the same chamber which induced “crosstalk”among these electrodes. Electric fields surrounding theseelectrodes would influence the preferential flux of negativelycharge species (substrate and bacteria), and consequently thecolonization and growth of bacteria on these electrodes.Biofilms growing on one electrode could produce mediatorsthat affect growth and current production of biofilms on otherelectrodes. Here, we used separate reactors at each potential sothat there was no interaction between the different anodes.The use of multiple electrodes in the same chamber also will

affect the anode potential that will produce the highest current.Torres et al.18 found that the most negative anode potential(−0.15 V compared to −0.09, 0.02, and 0.37 V) produced thehighest current. Parot et al.19 also placed three electrodes in asingle chamber using different potentials (0.54, 0.74, and 0.97V) and found higher currents were obtained at less positivepotentials. These results are different from that obtained here,as well as in many other studies, where higher current wasobtained at more positive potentials when electrodes wereplaced in separate reactors.10,13,14

Another important variable that can affect microbialcommunity development is the use of single- or two-chamberreactors. Kumar et al.12 compared the performance of biofilmsgrown at different set anode potentials (−0.09, 0.01, and 0.41V), and found the maximum current density was obtained at0.41 V in the single-chamber reactor, but at 0.01 V with thetwo-chamber system. The biofilm community showed ≥76%similarity among all the different set potentials for samples fromanodes in single-chamber reactors, and >65% similarity forsamples from anodes in two-chamber reactors, but the anodecommunities were different from each other in these twosystems. Proteobacteria dominated on the anodes in single-chamber reactors, while Tenericutes was the most dominantphylum for the two-chamber reactors. The reasons for thedifferences were not clear, and it may be that communitydevelopment was affected by hydrogen gas production andaccumulation in the anode chamber of the single-chambersystem. G. sulfurreducens can use hydrogen gas as an electrondonor, and experiments in single-chamber systems with thismicrobe showed use of hydrogen gas based on Coulombicefficiencies larger than 100% due to hydrogen cycling from thecathode to anode.6,37 In addition, anode communities fed aceticacid decreased in species richness and diversity, and increasedin numbers of G. sulfurreducens, when reactors were shiftedfrom single-chamber MFCs (little or no H2 evolution) toMECs (H2 generated).

38 Acetate utilization by G. sulfurreducens

is inhibited by high concentrations of H2, and thus biomassproduction could be adversely affected as this species cannot fixCO2.

39 However, it appears that the net result of hydrogen gasevolution in a single-chamber system is to favor the growth ofG. sulfurreducens and other hydrogen oxidizing exoelectro-gens.40 In both single- and two-chamber systems, growth andcurrent generation of G. sulfurreducens were improved at anodepotentials more positive than 0 V compared to more negativepotentials.10,15 Thus, while the chamber configuration is afactor in community development with mixed cultures, thegrowth of G. sulfurreducens was favored in both configurationsat more positive potentials. The presence of hydrogen gas inour experiments, versus that conducted with multiple anodes intwo-chamber systems,18 is another important factor to accountfor the differences in optimal potentials and communityevolution, and should be examined further.The responses of mixed cultures obtained here to the

different anode potentials is not the same as that previouslyobtained using pure cultures of G. sulfurreducens (SupportingInformation Figure S1). Previous tests15 in our laboratory usingG. sulfurreducens at the same anode potentials and with thesame reactors used here showed the maximum currentincreased with anode potentials up to 0.21 V, and then becameconstant at the higher anode potentials (Supporting Informa-tion Figure S1). This observation on the effect of anodepotential was consistent with another study10 on G.sulfurreducens in a two-chamber MEC, where current increasedat potentials of −0.16 to 0 V, with no further increase at 0.4 V.The different optimum potentials found in our study (0.21 V)and the other study (0 V) could be due to the use of the single-chamber design in our study, versus two-chamber design in theother study. The different responses of mixed culture(predominantly G. sulfurreducens) and pure cultures of G.sulfurreducens indicated that the presence of other bacteria canimpact current generation at these relatively high set potentials.Community development in both MECs and MFCs is

complex, and clearly affected by a number of factors in additionto anode potentials. For example, in MFCs anode communitydevelopment and power production can adversely be affectedby oxygen transfer from the cathode,41 or the resistance used toacclimate the reactors.42 A comparison of set potentialscompared to fixed resistances used in MFCs with anodesclose to the cathodes, with a separator used to reduce oxygentransfer, showed that a set anode potential of −0.2 V workedbetter for biofilm acclimation than 0.2 V or a fixed resistance(1000 Ω).43 However, multiple tests at −0.2 V often producederratic results in startup. The most consistent method toproduce good performance in the MFCs was to use a transferof solution from an existing MFC.43,44 Such transfers toinoculate MFCs and MECs are commonly reported in manystudies.43−46 Microbial community development is also affectedby the substrate used,47−49 although it appears that thecommunity that produces good current densities is inevitablyassociated with the predominance of various Geobacter specieswhen acetate is used.20,50

■ ASSOCIATED CONTENT

*S Supporting InformationMaximum currents of MECs with mixed culture and G.sulfurreducens developed at potentials of −0.25, −0.09, 0.21,0.51, and 0.81 V vs SHE. This material is available free ofcharge via the Internet at http://pubs.acs.org.

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581356

■ AUTHOR INFORMATION

Corresponding Author*Telephone: +1 814 863 7908. Fax: +1 814 863 7304. E-mail:blogan@psu.edu.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

This research was supported by Award KUS-I1-003-13 fromthe King Abdullah University of Science and Technology(KAUST). The authors also want to thank the threeanonymous reviewers for their constructive comments on theoriginal manuscript.

■ REFERENCES(1) Logan, B. E., Microbial Fuel Cells; John Wiley & Sons: New York,2008.(2) Call, D.; Logan, B. E. Hydrogen production in a single chambermicrobial electrolysis cell lacking a membrane. Environ. Sci. Technol.2008, 42, 3401−3406.(3) Cao, X. X.; Huang, X.; Liang, P.; Xiao, K.; Zhou, Y. J.; Zhang, X.Y.; Logan, B. E. A new method for water desalination using microbialdesalination cells. Environ. Sci. Technol. 2009, 43, 7148−7152.(4) Pant, D.; Singh, A.; Van Bogaert, G.; Olsen, S. I.; Nigam, P. S.;Diels, L.; Vanbroekhoven, K. Bioelectrochemical systems (BES) forsustainable energy production and product recovery from organicwastes and industrial wastewaters. RSC Adv. 2012, 2, 1248−1263.(5) Liu, H.; Ramnarayanan, R.; Logan, B. E. Production of electricityduring wastewater treatment using a single chamber microbial fuel cell.Environ. Sci. Technol. 2004, 38, 2281−2285.(6) Call, D. F.; Wagner, R. C.; Logan, B. E. Hydrogen production byGeobacter species and a mixed consortium in a microbial electrolysiscell. Appl. Environ. Microbiol. 2009, 75, 7579−7587.(7) Kim, Y.; Logan, B. E. Hydrogen production from inexhaustiblesupplies of fresh and salt water using microbial reverse-electrodialysiselectrolysis cells. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 16176−16181.(8) Nam, J. Y.; Cusick, R. D.; Kim, Y.; Logan, B. E. Hydrogengeneration in microbial reverse-electrodialysis electrolysis cells using aheat-regenerated salt solution. Environ. Sci. Technol. 2012, 46, 5240−5246.(9) Wagner, R. C.; Call, D. I.; Logan, B. E. Optimal set anodepotentials vary in bioelectrochemical systems. Environ. Sci. Technol.2010, 44, 6036−6041.(10) Wei, J.; Liang, P.; Cao, X.; Huang, X. A new insight intopotential regulation on growth and power generation of Geobactersulfurreducens in microbial fuel cells based on energy viewpoint.Environ. Sci. Technol. 2010, 44, 3187−3191.(11) Logan, B. E.; Hamelers, B.; Rozendal, R. A.; Schrorder, U.;Keller, J.; Freguia, S.; Aelterman, P.; Verstraete, W.; Rabaey, K.Microbial fuel cells: Methodology and technology. Environ. Sci.Technol. 2006, 40, 5181−5192.(12) Kumar, A.; Siggins, A.; Katuri, K.; Mahony, T.; O’Flaherty, V.;Lens, P.; Leech, D. Catalytic response of microbial biofilms grownunder fixed anode potentials depends on electrochemical cellconfiguration. Chem. Eng. J. 2013, 230, 532−536.(13) Finkelstein, D. A.; Tender, L. M.; Zeikus, J. G. Effect ofelectrode potential on electrode-reducing microbiota. Environ. Sci.Technol. 2006, 40, 6990−6995.(14) Busalmen, J. P.; Esteve-Nunez, A.; Feliu, J. M. Whole cellelectrochemistry of electricity-producing microorganisms evidence anadaptation for optimal exocellular electron transport. Environ. Sci.Technol. 2008, 42, 2445−2450.(15) Zhu, X. P.; Yates, M. D.; Logan, B. E. Set potential regulationreveals additional oxidation peaks of Geobacter sulfurreducens anodicbiofilms. Electrochem. Commun. 2012, 22, 116−119.

(16) Aelterman, P.; Freguia, S.; Keller, J.; Verstraete, W.; Rabaey, K.The anode potential regulates bacterial activity in microbial fuel cells.Appl. Microbiol. Biotechnol. 2008, 78, 409−418.(17) Wang, X.; Feng, Y. J.; Liu, J.; Lee, H.; Ren, N. Q. Performanceof a batch two-chambered microbial fuel cell operated at differentanode potentials. J. Chem. Technol. Biotechnol. 2011, 86, 590−594.(18) Torres, C. I.; Krajmalnik-Brown, R.; Parameswaran, P.; Marcus,A. K.; Wanger, G.; Gorby, Y. A.; Rittmann, B. E. Selecting anode-respiring bacteria based on anode potential: phylogenetic, electro-chemical, and microscopic characterization. Environ. Sci. Technol. 2009,43, 9519−9524.(19) Parot, S.; Delia, M. L.; Bergel, A. Forming electrochemicallyactive biofilms from garden compost under chronoamperometry.Bioresour. Technol. 2008, 99, 4809−4816.(20) Logan, B. E. Exoelectrogenic bacteria that power microbial fuelcells. Nat. Rev. Microbiol. 2009, 7, 375−381.(21) Marsili, E.; Sun, J.; Bond, D. R. Voltammetry and growthphysiology of Geobacter sulfurreducens biofilms as a function of growthstage and imposed electrode potential. Electroanal. 2010, 22, 865−874.(22) Call, D. F.; Logan, B. E. A method for high throughputbioelectrochemical research based on small scale microbial electrolysiscells. Biosens. Bioelectron. 2011, 26, 4526−4531.(23) Ishii, S.; Watanabe, K.; Yabuki, S.; Logan, B. E.; Sekiguchi, Y.Comparison of electrode reduction activities of Geobacter sulfurredu-cens and an enriched consortium in an air-cathode microbial fuel cell.Appl. Environ. Microbiol. 2008, 74, 7348−7355.(24) Ishak, H. D.; Plowes, R.; Sen, R.; Kellner, K.; Meyer, E.; Estrada,D. A.; Dowd, S. E.; Mueller, U. G. Bacterial diversity in Solenopsisinvicta and Solenopsis geminata ant colonies characterized by 16Samplicon 454 pyrosequencing. Microbial Ecol. 2011, 61, 821−831.(25) Schloss, P. D.; Westcott, S. L.; Ryabin, T.; Hall, J. R.; Hartmann,M.; Hollister, E. B.; Lesniewski, R. A.; Oakley, B. B.; Parks, D. H.;Robinson, C. J.; Sahl, J. W.; Stres, B.; Thallinger, G. G.; Van Horn, D.J.; Weber, C. F. Introducing Mothur: Open-source, platform-independent, community-supported software for describing andcomparing microbial communities. Appl. Environ. Microbiol. 2009,75, 7537−7541.(26) Edgar, R. C.; Haas, B. J.; Clemente, J. C.; Quince, C.; Knight, R.UCHIME improves sensitivity and speed of chimera detection.Bioinformatics 2011, 27, 2194−2200.(27) Marcus, R. A. On the theory of oxidation-reduction reactionsinvolving electron transfer. J. Chem. Phys. 1956, 24, 966−978.(28) Marcus, R. A. Electrostatic free energy and other properties ofstates having nonequilibrium polarization. J. Chem. Phys. 1956, 24,979−989.(29) Marcus, R. A. Chemical and electrochemical electron-transfertheory. Annu. Rev. Phys. Chem. 1964, 15, 155−&.(30) Marsili, E.; Rollefson, J. B.; Baron, D. B.; Hozalski, R. M.; Bond,D. R. Microbial biofilm voltammetry: Direct electrochemical character-ization of catalytic electrode-attached biofilms. Appl. Environ. Microbiol.2008, 74, 7329−7337.(31) Reguera, G.; Nevin, K. P.; Nicoll, J. S.; Covalla, S. F.; Woodard,T. L.; Lovley, D. R. Biofilm and nanowire production leads toincreased current in Geobacter sulfurreducens fuel cells. Appl. Environ.Microbiol. 2006, 72, 7345−7348.(32) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:Fundamentals and Applications; Wiley: New York, 2000.(33) Zhu, X. P.; Tokash, J. C.; Hong, Y. Y.; Logan, B. E. Controllingthe occurrence of power overshoot by adapting microbial fuel cells tohigh anode potentials. Bioelectrochemistry 2013, 90, 30−35.(34) Katuri, K. P.; Kavanagh, P.; Rengaraj, S.; Leech, D. Geobactersulfurreducens biofilms developed under different growth conditions onglassy carbon electrodes: Insights using cyclic voltammetry. Chem.Commun. 2010, 46, 4758−4760.(35) Lloyd, J. R.; Leang, C.; Myerson, A. L. H.; Coppi, M. V.; Cuifo,S.; Methe, B.; Sandler, S. J.; Lovley, D. R. Biochemical and geneticcharacterization of PpcA, A periplasmic c-type cytochrome inGeobacter sulfurreducens. Biochem. J. 2003, 369, 153−161.

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581357

(36) Inoue, K.; Qian, X. L.; Morgado, L.; Kim, B. C.; Mester, T.;Izallalen, M.; Salgueiro, C. A.; Lovley, D. R. Purification andcharacterization of OmcZ, an outer-surface, octaheme c-typecytochrome essential for optimal current production by Geobactersulfurreducens. Appl. Environ. Microbiol. 2010, 76, 3999−4007.(37) Lee, H. S.; Rittmann, B. E. Significance of biological hydrogenoxidation in a continuous single-chamber microbial electrolysis cell.Environ. Sci. Technol. 2010, 44, 948−954.(38) Kiely, P. D.; Cusick, R.; Call, D. F.; Selembo, P. A.; Regan, J. M.;Logan, B. E. Anode microbial communities produced by changingfrom microbial fuel cell to microbial electrolysis cell operation usingtwo different wastewaters. Bioresour. Technol. 2011, 102, 388−394.(39) Brown, D. G.; Komlos, J.; Jaffe, P. R. Simultaneous utilization ofacetate and hydrogen by Geobacter sulfurreducens and implications foruse of hydrogen as an indicator of redox conditions. Environ. Sci.Technol. 2005, 39, 3069−3076.(40) Kimura, Z.; Okabe, S. Acetate oxidation by syntrophicassociation between Geobacter sulfurreducens and a hydrogen-utilizingexoelectrogen. ISME J. 2013, 7, 1472−1482.(41) Quan, X. C.; Quan, Y. P.; Tao, K. Effect of anode aeration onthe performance and microbial community of an air−cathodemicrobial fuel cell. Chem. Eng. J. 2012, 210, 150−156.(42) Katuri, K. P.; Scott, K.; Head, I. M.; Picioreanu, C.; Curtis, T. P.Microbial fuel cells meet with external resistance. Bioresour. Technol.2011, 102, 2758−2766.(43) Zhang, F.; Xia, X.; Luo, Y.; Sun, D.; Call, D. F.; Logan, B. E.Improving startup performance with carbon mesh anodes in separatorelectrode assembly microbial fuel cells. Bioresour. Technol. 2013, 133,74−81.(44) Liu, Y.; Harnisch, F.; Fricke, K.; Sietmann, R.; Schroder, U.Improvement of the anodic bioelectrocatalytic activity of mixed culturebiofilms by a simple consecutive electrochemical selection procedure.Biosens. Bioelectron. 2008, 24, 1006−1011.(45) Rabaey, K.; Ossieur, W.; Verhaege, M.; Verstraete, W.Continuous microbial fuel cells convert carbohydrates to electricity.Water Sci. Technol. 2005, 52, 515−523.(46) Wang, X.; Cheng, S. A.; Feng, Y. J.; Merrill, M. D.; Saito, T.;Logan, B. E. Use of carbon mesh anodes and the effect of differentpretreatment methods on power production in microbial fuel cells.Environ. Sci. Technol. 2009, 43, 6870−6874.(47) Jung, S.; Regan, J. M. Comparison of anode bacterialcommunities and performance in microbial fuel cells with differentelectron donors. Appl. Microbiol. Biot. 2007, 77, 393−402.(48) Chae, K. J.; Choi, M. J.; Lee, J. W.; Kim, K. Y.; Kim, I. S. Effectof different substrates on the performance, bacterial diversity, andbacterial viability in microbial fuel cells. Bioresour. Technol. 2009, 100,3518−3525.(49) Xing, D. F.; Cheng, S. A.; Regan, J. M.; Logan, B. E. Change inmicrobial communities in acetate- and glucose-fed microbial fuel cellsin the presence of light. Biosens. Bioelectron. 2009, 25, 105−111.(50) Yates, M. D.; Kiely, P. D.; Call, D. F.; Rismani-Yazdi, H.; Bibby,K.; Peccia, J.; Regan, J. M.; Logan, B. E. Convergent development ofanodic bacterial communities in microbial fuel cells. ISME J. 2012, 6,2002−2013.

Environmental Science & Technology Article

dx.doi.org/10.1021/es404690q | Environ. Sci. Technol. 2014, 48, 1352−13581358