Effects of anode potentials on bioelectrogenic conversion of xylose and microbial community compositions

1

Effects of anode potentials on bioelectrogenic conversion of xylose and microbial community compositions

Marika E. Kokko*, Annukka E. Mäkinen, Mira L.K. Sulonen, and Jaakko A.

Puhakka

a Department of Chemistry and Bioengineering, Tampere University of Technology, P.O. Box 541, FI-33101Tampere, Finland

* Corresponding author. E-mail address: marika.kokko@tut.fi; Tel.: +358 44 5430862; Fax: +358 3 3115 2869.

2

Abstract

The results on the effects of different anode potentials on current densities, coulombic efficiencies and microbial communities are contradictory and have not been studied with xylose, an important constituent of lignocellulosic materials. In this study, the effects of different anode potentials (+0.2, 0 and -0.2 V vs. Ag/AgCl) on current generation, xylose degradation and microbial communities were examined with an exoelectrogenic enrichment culture originating from anaerobic sludge. Anode potential of +0.2 V (vs. Ag/AgCl) resulted in the highest current density and coulombic

efficiency of 1.5 ± 0.2 A/m² and 62 ± 11 %, respectively, and there was no

accumulation of soluble metabolites. With anode potentials of 0 and -0.2 V the current densities remained low and acetate, butyrate and propionate were detected in the end of batch runs. Different anode potentials resulted in substantial differences in the anodic bacterial species. At more positive anode potentials, Ochrobactrum intermedium reported to be capable of direct electron transfer dominated. At more negative anode potentials, a known mediator-producer, Alcaligenes faecalis, and Desulfitobacterium hafnience, that has been reported to use mediated electron transfer, were detected. This study shows that the anode potential has a substantial effect on microbial communities and on xylose metabolism.

Keywords: microbial fuel cell, anode potential, anaerobic processes, batch processing, bioconversion, biocatalysis

3

1. Introduction

In microbial fuel cells (MFCs), organic compounds are oxidized at the anode producing current. Anode potential is known to affect the performance of the anode.

The anode potential determines the energy available for microbial growth; Theoretical energy gain of anode respiring bacteria (ARB) increases with the difference of redox potentials between substrate and the electrode [1,2]. The energy available for

microorganisms is determined by Gibbs free energy (ΔG°´, Eq.1), where n is the number of electrons per reaction mol, F the Faraday´s constant, and E^0´ the theoretical oxidation potential of the substrate calculated at standard conditions. Thus, more positive anode potential should increase the growth of bacteria and result in higher biocatalyst density and faster start-up of current production [3,4]. However, Geobacter sp., for example, use only a small portion of their net electron flow to ATP production and has been reported to dominate microbial communities at low anode potentials [4].

According to Finkelstein et al. [5], Wei et al. [6] and Wagner et al. [1], the anode potential selects for ARB having redox potentials of the terminal respiratory proteins more negative than the anode potential.

ΔG°´ = -nFΔE⁰´ (1)

The results on the effects of anode potential on current outputs and microbial community compositions are contradictory. For example, higher current densities at more positive anode potentials were reported by Finkelstein et al. [5] and Sun et al. [7], while Torres et al. [8] obtained higher current densities at more negative anode

potentials. Similar debate concerns the effects of anode potential on microbial

4 community composition. Zhu et al. [9] concluded that although the biofilm biomass

increased with increasing anode potential, the bacterial community composition was unaffected. On the contrary, Torres et al. [8] reported that more positive anode potential resulted in more diverse microbial community, while more negative anode potential enriched Geobacter sulfurreducens. Thus, the effects of anode potentials on current production and microbial community compositions remain to be clarified and require further studies (for a review, see Wagner et al. [1]).

In this study, simultaneous fermentation and anaerobic respiration of xylose, an

important constituent of lignocellulosic materials, was studied in fed-batch two-chamber MFCs. The aim was to determine for the first time the effects of different anode

potentials on current production, xylose conversion and bacterial community composition with an exoelectrogenic culture enriched originally from anaerobic digester.

5

2. Materials and methods

2.1 MFC experiments

The experiments were conducted in fed-batch two-chamber MFCs [10]. The anode and cathode chambers were separated with anion exchange membrane (AMI-7000,

Membrane International, USA) and had working volumes of 75 mL. Plain graphite electrodes (38.5 cm², McMaster-Carr, USA) were used. Anode and cathode chambers were kept anaerobic by flushing the chambers with N2 in the beginning of the

experiments and during the feeding. MFCs were grown at 37°C without mixing. The pH of the anode solution was adjusted to 7.0 ± 0.1 in the beginning and during each MFC feeding. Cathode solution was 50 mM K3Fe(CN)6 in 100 mM Na2HPO4 buffer (pH 7).

Synthetic growth medium contained 6.7 mM (1 g/L) xylose and was prepared as described earlier [11].

The MFCs were inoculated with an anaerobic digester culture (10 % (v/v)) enriched for electricity production on xylose in similar reactor configurations and with external resistance of 100 Ω [12]. The experiments were done as duplicate. Anode potentials of +0.2 V, 0 V and -0.2 V were adjusted against Ag/AgCl reference electrode (-205 mV vs. SHE; Sentek, UK). If not otherwise mentioned, all the anode potentials are reported against Ag/AgCl. The performance of the MFCs with anode potential control was compared to duplicate MFCs with uncontrolled anode potential run with external resistance of 100 Ω. In addition, control MFCs with external resistance of 100 Ω were run i) with inoculum and without substrate, and ii) without inoculum and with substrate.

The MFCs were fed regularly by replacing 10 mL of the anodic solution with fresh medium containing xylose (final concentration 6.7 mM). Prior to feeding, samples were

6 taken for the analysis of volatile fatty acids (VFAs), alcohols, and soluble chemical

oxygen demand (CODs) that was used to estimate the remaining xylose concentration.

During the feeding, the catholyte solution was also replenished.

2.2 Electrochemical analyses

Data acquisition system (Agilent 34970A, Agilent Technologies, Santa Clara, USA) was used to monitor and record the voltage and anode potential every 2 min. Current and power densities were calculated against the area of the anode electrode. Coulombic efficiency was calculated against the initial xylose concentration (6.7 mM) according to Logan et al. [13]. Potentiostat (μSTAT 8000, DropSens, Asturias, Spain) was used to adjust the anode potential every three minutes. Anode electrode was working electrode, cathode counter electrode, and Ag/AgCl reference electrodes in 3 M KCl solution connected to the anode chambers through glass capillaries (ProSense) were used as references. A complete coulombic analysis was done for each MFC run according to Huang and Logan [14].

2.3 Chemical analyses

VFAs (acetate, butyrate, isobutyrate, propionate, valerate) and alcohols (ethanol, butanol) were analysed with a gas chromatograph (GC-2010 Plus, Shimadzu, Kyoto, Japan) equipped with ZB-WAX plus column (Phenomenex, USA) and flame ionization detector (FID). The injector and detector temperatures were kept at 250°C. The oven temperature was programmed as follows: 40°C for 2 min, increasing to 160°C at 20°C/min, increasing to 220°C at 40°C/min, and then constant at 220°C for 2 min.

Helium was used as carrier gas at a flow rate of 1.0 mL/min. Samples were filtered (0.2 µm) and acidified with oxalic acid. Soluble COD was analysed from filtrated samples

7 (0.2 μm) with dichromate method according to standard SFS5504 [15]. Possible

interference of chloride ions from the reference electrode assembly on CODs analysis was prevented by using HgSO4 during analysis. The concentration of non-degraded xylose was calculated by subtracting the produced VFAs and alcohols (converted to COD equivalents with Eq. 3) from the remaining CODs.

𝐶𝑂𝐷 = (8 ∙ (4𝑥 + 𝑦 − 2𝑧) (12𝑥 + 𝑦 + 16𝑧)) 𝑔 𝐶𝑂𝐷 𝑔 𝐶⁄ ⁄ _𝑥𝐻_𝑦𝑂_𝑧 (3)

2.4 Bacterial community analysis

Duplicate samples from the anode solution and biofilm were taken from the end points of the MFC runs and the samples were stored at -20ºC. DNA extraction, polymerase chain reaction – denaturing gradient gel electrophoresis (PCR-DGGE), sequencing and analysing of partial 16S rRNA genes were done as described earlier by Mäkinen et al.

[16].

3. Results and discussion

The exoelectrogenic culture enriched from anaerobic digester [12] produced current densities of 0.37 A/m² and 19.1 A/m³ in similar two-chamber MFCs used in this study with an external resistance of 100 Ω. The coulombic efficiency (CE) with 6.7 mM xylose was 13 %. Ethanol was the sole soluble metabolite from xylose and based on the coulombic balance, 40 % of the electrons from xylose were directed towards ethanol production while 48 % of the electrons were lost to unknown processes. In the end of the enrichment, the culture contained, among others, denitrifying bacteria, Comamonas denitrificans and Paracoccus pantotrophus, as well as a xylanolytic species,

8 Ruminobacillus xylanolyticum [12]. With this culture, the effects of controlled anode

potentials on current production, xylose degradation and bacterial community compositions of anode biofilms and solutions were studied.

3.1 Effects of anode potentials on current production

The effects of different anode potentials poised at +0.2, 0 and -0.2 V (vs. Ag/AgCl) on current production were delineated. In control MFCs run either without inoculum or without substrate, no current was produced (results not shown). The maximum current density and CE of 1.47 ± 0.23 A/m² and 63 ± 15 %, respectively, were obtained with poised anode potential of +0.2 V, while low current densities and CEs were produced at anode potentials of -0.2 and 0 V (Figure 1). The anode potential of the MFC with uncontrolled anode potential (with 100 Ω external resistance) was at the highest +0.1 V during stable current production. The maximum current density and CE of the MFC with uncontrolled anode potential were 0.40 ± 0.01 A/m² and 31 ± 3 %, respectively.

Figure 1, Table 1

Controlling the anode potential at +0.2 V resulted in 2.5 times higher current densities and 2 times higher CEs than obtained in the MFC with 100 Ω external resistance. In the MFC with uncontrolled anode potential, the anode potential was the highest (+0.1 V) after feeding at high xylose concentration and the anode potential decreased (close to - 0.2 V) with decreasing xylose concentration. The more positive anode potential is more beneficial for the bacteria due to higher energy available for the bacteria [3,4]. Thus, the maintenance of anode potential at +0.2 V likely increased the growth and activity of exoelectrogenic bacteria resulting in higher current densities. Furthermore, the electron

9 transfer from the microorganisms to the anode electrode depends on the potential of the outer membrane protein involved in the electron transport chain or the potential of the mediator and the anode potential; The anode potential has to be more positive than the protein or mediator [1]. Thus, anode potential in MFCs with external resistance might have decreased below the point at which the bacteria are capable of donating electrons to the electrode. The results are in accordance with Wang et al. [17] and Sun et al. [7]

reporting higher current densities with applied anode potentials of +0.2 and -0.05 V vs.

Ag/AgCl, respectively, than with anodes without applied anode potential. Thus,

controlling anode potential can be used as a tool for enhancing the performance and the start-up of the MFCs. Further experiments are required to evaluate whether similar results are obtained from original culture before enrichment and whether the MFC performance can be sustained, if the anode potential is no longer controlled.

The initial culture was enriched in a two-chamber MFC with an external resistance of 100 Ω resulting in anode potentials above 0 V. Thus, the low current production at anode potentials of -0.2 and 0 V was likely a consequence of the acclimation of the previously enriched culture to more positive anode potentials. The enrichment culture likely contained no or low concentrations of bacteria capable of extracellular electron transfer at lower anode potentials as suggested by Torres et al. [8]. The acclimation times were also longer at more negative anode potentials. With anodes poised at -0.2 and 0 V, it took 13 days before the current production was stable between the

consecutive feeding cycles. With anode poised at +0.2 V and in MFCs with 100 Ω external resistance, current production stabilised after 5 days suggesting that the microbial culture was already acclimated to the higher anode potentials.

10 As can be seen from Table 1, the optimal anode potential changes considerably between studies depending on, e.g., the original inoculum, substrate and cell configuration. More positive anode potential has been reported to increase current production in many studies [5,6,7]. Furthermore, more positive anode potential has resulted in faster start- up, enhanced the bacterial growth and resulted in more diverse bacterial community on anode biofilm [5,6,8,18]. Our results are in agreement with these studies, since more positive anode potential accelerated the start-up times and increased the current densities. However, Torres et al. [8] reported lower anode potential to increase current production, result in thicker biofilm formation and favour the enrichment of G.

sulfurreducens in mixed culture.

3.2 Effects of anode potentials on bacterial metabolism

The xylose degradation during stable current production was 48 ± 14 %, 30 ± 2 %, 15 ± 2 % and 24 ± 0 % with poised anode potentials of -0.2, 0 and +0.2 V and with MFC with 100 Ω external resistance, respectively. The results show that the highest

degradation efficiencies were obtained with the most negative anode potential, although the current densities and coulombic efficiencies (calculated against added xylose) were the highest with the most positive anode potential. Thus, at the most positive anode potential of +0.2 V, the degraded xylose fraction was efficiently used for current generation and there was no accumulation of soluble metabolites during stable current production (Table 2). This also indicates that exoelectrogenic bacteria capable of donating electrons to the electrode at higher anode potentials (> 0 V) were present. At more positive anode potentials, further improvement of xylose degradation would result in higher current densities and coulombic efficiencies.

11 Although xylose degradation was the highest at the lowest anode potentials, only small

amount of the electrons were directed to current production. Instead, electrons were directed into soluble metabolites. Acetate was the main soluble metabolite with small concentrations of propionate (< 1.5 mM) and butyrate (< 0.5 mM). Acetate is often regarded as the most potential substrate for exoelectrogenic bacteria, which suggests that at more negative anode potentials there were not enough exoelectrogens that could oxidize acetate and donate electrons to the electrode at such low potentials. According to coulombic analysis, 34 – 55 % of the electrons were lost to unknown processes at more negative anode potentials (Table 2). The high losses are likely due to soluble metabolites, such as lactate or formate, that cannot be detected in the gas

chromatographic analysis. Since the concentration of non-degraded xylose was calculated by subtracting the COD equivalents of the measured metabolites from the remaining COD, the values of remaining xylose may be inaccurate due to unmeasured organic metabolites.

Table 2

3.3 Effects of anode potentials on microbial communities

The microbial community results were as shown in Figure A.1 and Table 3. Microbial communities were diverse, containing α-, β- and γ-proteobacteria, Bacteroidetes, Firmicutes and Sprirochaetes. Bacterial species capable of xylose degradation were detected and they were closely related to Ruminobacillus xylanolyticum (97.1 %

similarity) and Bacteroides sp. (100 % similarity) [19] These species belonging to class Bacteroidetes were present in every anode sample. However, in MFC with the anode potential of +0.2 V having the lowest xylose degradation efficiency, known xylose

12 degrading bacteria were only detected in the anode solution and not in the biofilm. In

other MFCs, strong bands of xylose degraders were also visible in the biofilm samples.

Table 3

Bacteria reported to be capable for current production were different at different anode potentials. MFCs with anode potential of +0.2 V or run with external resistance of 100 Ω produced the highest current densities. In these MFCs, a bacterium from class α- proteobacteria and closely related to Ochrobactrum intermedium (99 - 99.8 % similarity), was detected both in the solution and biofilm. O. intermedium is a nitrate reducer [20] that has been shown to produce electricity from, e.g., acetate, propionate, butyrate and glucose without external mediators [21]. Zuo et al. [21] reported that current densities produced with O. intermedium varied between 0.3 A/m² and 0.7 A/m² with 1 g/L propionate or 1 g/L acetate as substrate, respectively. These current densities are similar to the results obtained in this study, i.e., 0.4 A/m² with external resistance of 100 Ω and 1.5 A/m² with anode potential of +0.2 V. Direct current production from xylose with O. intermedium has not been reported and thus, it is not clear whether electricity in this experiment was produced directly from xylose or from the soluble metabolites. In addition to O. intermedium, a denitrifying bacterium closely related to Citrobacter amalonaticus (100 % similarity) was detected. C. amalonaticus has been earlier observed in MFC degrading petroleum hydrocarbons, although its role in the MFC was not eludicated [22].

O. intermedium or other exoelectrogenic bacteria reported to be capable of direct electron transfer were not detected in MFCs with anode potentials of -0.2 or 0 V.

13 Instead, β-proteobacteria closely related to Alcaligenes faecalis (97.7 % similarity) and a dechlorinating, Fe(III) reducing bacterium Desulfitobacterium hafnience (100 % similarity) [23] belonging to Firmicutes were identified in these MFCs. A. faecalis has been reported to excrete a mobile electron carrier, plastocyanin, in MFC fed with glucose [24]. In addition, D. hafnience has been shown to produce electricity with mediators from, e.g., hydrogen, pyruvate and ethanol [25]. Thus, mediators produced by A. faecalis may have been responsible for the low current production in these MFCs.

Mediator production is, however, energy-intensive [26] and often leads to lower current production than direct electricity transfer, which was also observed in this study. The microbial community and current production results demonstrate that the enrichment culture did not contain bacteria capable of efficiently transferring electrons to the electrode at more negative anode potentials. This is also supported by the fact that the concentrations of acetate, that should be the easiest substrate for exoelectrogens, increased at more negative anode potentials.

4. Conclusions

With exoelectrogenic culture enriched from anaerobic digester, applied anode potential had a substantial effect on exoelectrogenic xylose conversion and microbial

communities. The most positive anode potential of +0.2 V (vs. Ag/AgCl) resulted in the highest coulombic efficiency of 63 ± 15 % and no soluble metabolites were detected.

Anode potentials of 0 V and -0.2 V resulted in low current densities and higher

concentrations of soluble metabolites indicating the absence of efficient exoelectrogens.

More positive anode potentials selected for bacterium reported to be capable of direct

14 electron transfer, O. intermedium, while at more negative anode potentials two bacteria

reported to conduct mediated electron transfer were present.

Acknowledgements

The work was funded by The Finnish Doctoral Programme of Environmental Science and Technology (A.E.M.) and Tampere University of Technology Graduate School (M.E.N.).

References

[1] R.C. Wagner, D.F. Call, B.E. Logan, Optimal set anode potentials vary in bioelectrochemical systems, Environ. Sci. Technol. 44 (2010) 6036-6041.

[2] J. Bosch, K.Y. Lee, S.F. Hong, F. Harnisch, U. Schröder, R.U. Meckenstock, Metabolic efficiency of Geobacter sulfurreducens growing on anodes with different redox potentials, Curr. Microbiol. 68 (2014) 763-768.

[3] P. Aelterman, S: Freguia, J. Keller, W. Verstraete, K. Rabaey, The anode potential regulates bacterial activity in microbial fuel cells, Appl. Microbiol. Biotechnol. 78 (2008) 409-418.

[4] D.R. Bond, Electrodes as electron acceptors and the bacteria who love them, in: L.L.

Barton, M- Mandel, A. Lay (Eds.), Geomicrobiology: molecular and environmental perspective, Springer, Netherlands, 2010, pp. 385-399.

[5] D.A. Finkelstein, L.M. Tender, J.G. Zeikus, Effect of electrode potential on electrode-reducing microbiota, Environ. Sci. Technol. 40 (2006) 6990-6995.

15 [6] J. Wei, P. Liang, X. Cao, W. Huang, A new insight into potential regulation on

growth and power generation of Geobacter sulfurreducens in microbial fuel cells based on energy viewpoint, Environ. Sci. Technol. 44 (2010) 3187-3191.

[7] D. Sun, D.F. Call, P.D. Kiely, A. Wang, B.E. Logan, Syntrophic interactions improve power production in formic acid fed MFCs operated with set anode potentials or fixed resistances, Biotechnol. Bioeng. 109 (2012) 405-414.

[8] C.I. Torres, R. Krajmalnik-Brown, P. Parameswaran, A.K. Marcus, G. Wanger, Y.A. Gorby, B.E. Rittmann, Selecting anode-respiring bacteria based on anode potential: phylogenetic, electrochemical, and microscopic characterization, Environ.

Sci. Technol. 43 (2009) 9519-9524.

[9] X. Zhu, M.D. Yates, M.C. Hatzell, H.A. Rao, P.E. Saikaly, B.E. Logan, Microbial community composition is unaffected by anode potential, Environ. Sci. Technol. 48 (2014) 1352-1358.

[10] A.-M. Lakaniemi, O.H. Tuovinen, J.A. Puhakka, Production of electricity and butanol from microalgal biomass in microbial fuel cells, BioEnergy Res. 5 (2012) 481- 491.

[11] A.E. Mäkinen, C.-H. Lay, M.E. Nissilä, J.A. Puhakka, Bioelectricity production on xylose with a compost enrichment culture, Int. J. Hydrogen Energy 38 (2013) 15608- 15612.

[12] M. Nissilä, Biohydrogen, bioelectricity and bioalcohols from cellulosic materials, Publication 1103, Tampere University of Technology, Tampere, Finland, 2013.

[13] B.E. Logan, B. Hamelers, R. Rozendal, U. Schröder, J. Keller, S. Freguia, P.

Aelterman, W. Verstraete, K. Rabaey, Microbial fuel cells: Methodology and technology, Environ. Sci. Technol. 40 (2006) 5181-5192.

[14] L. Huang, B.E. Logan, Electricity production from xylose in fed-batch and continuous-flow microbial fuel cell, Appl. Microbiol. Biotechnol. 80 (2008) 655-664.

16 [15] SFS 5504, Determination of chemical oxygen demand (COD Cr) in water with the

closed tube method. Oxidation with dichromate, Finnish Standards Association (SFS), 1998.

[16] A.E. Mäkinen, M.E. Nissilä, J.A. Puhakka, Dark fermentative hydrogen production from xylose by a hot spring enrichment culture, Int. J. Hydrogen Energy 37 (2012) 12234-12240.

[17] X. Wang, Y. Feng, N. Ren, H. Wang, H. Lee, N. Li, Q. Zhao, Accelerated start-up of two-chambered microbial fuel cells: Effect of anodic positive poised potential, Electrochimica Acta 54 (2009) 1109-1114.

[18] J.P. Busalmen, A. Esteve-Nuñez, J.M. Felium, Whole cell electrochemistry of electricity-producing microorganisms evidence an adaptation for optimal exocellular electron transport, Environ. Sci. Technol. 42 (2008) 2445–2450.

[19] H.N. Shah, The genus Bacteroides and related taxa, in: A. Balows, H.G. Trüper, M.

Dworkin, W. Harder, K.H. Schleifer (Eds.), The prokaryotes, Springer-Verlag, New York, 1992, pp. 3593-3607.

[20] J. Velasco, C. Romero, I. Lópey-Goni, J. Leiva, R. Díaz, I. Moriyón, Evaluation of the relatedness of Brucella spp. and Octrobactrum antrhopi and description of

Ochrobactrum intermedium sp. nov., a new species with a close relationship to Brucella spp, Int. J. Syst. Bacteriol. 48 (1998) 759-768.

[21] Y. Zuo, D. Xing, J.M. Regan, B.E. Logan, Isolation of the exoelectrogenic bacterium Ochrobactrum anthropic YZ-1 by using a U-tube microbial fuel cell, Appl.

Environ. Microbiol. 74 (2008) 3130-3137.

[22] J.M. Morris, J. Song, B. Crimi, A. Pruden, Microbial fuel cell in enhancing anaerobic biodegradation of diesel, Chem. Eng. J. 146 (2000) 161-167.

17 [23] N. Christiansen, B.K. Ahring, Desulfitobacterium Hafniense sp. nov., an anaerobic, reductively dechlorinating bacterium, Int. J. Syst. Bacteriol. 46 (1996) 442-448.

[24] K. Rabaey, N. Boon, S.D. Siciliano, M. Verhaege, W. Verstraete, Biofuel cells select for microbial consortia that self-mediate electron transfer, Appl. Environ.

Microbiol. 70 (2004) 5373-5382.

[25] C.E. Milliken, H.D. May, Sustained generation of electricity by the spore-forming, Gram-positive, Desulfitobacterium hafniense strain DCB2, Appl. Microbiol.

Biotechnol. 73 (2007) 1180-1189.

[26] S.E. Childers, S. Ciufo, D.R. Lovley, Geobacter metallireducens accesses insoluble Fe(III)oxide by hemotaxis, Nature 416 (2002) 767-769.

[27] A. Kumar, A. Siggins, K. Katuri, T. Mahony, V. O’Flaherty, P. Lens, D. Leech, Catalytic response of microbial biofilms grown under fixed anode potentials depends on electrochemical cell configuration, Chem. Eng. J. 230 (2013) 532-536.

18 Figure legends

Figure 1 Average current densities (A) and coulombic efficiencies (B) with different poised anode potentials and MFC with uncontrolled anode potential (100 Ω external resistance).

Standard deviations are shown in the figures.

19 A

B

Figure 1

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8

100 Ω -0.2 V 0 V 0.2 V

Current density (A/m2)

0 10 20 30 40 50 60 70 80

100 Ω -0.2 V 0 V 0.2 V

Coulombic efficiency (%)

20 Table 1 Effects of anode potentials on current production in selected MFC studies.

Anode potential (V vs. Ag/AgCl)

Inoculum Substrate Current density (A/m²)

Reference -0.36

-0.2 +0.

G. sulfurreducens Acetate 1.3 2.0 2.0

[6]

+0.2

uncontrolled (+0.29 – - 0.36)

Domestic wastewater and anaerobic sludge

Glucose 0.6

0.08

[17]

-0.058 +0.103 +0.618

Sediment Acetate 0.11

0.13 0.17

[5]

-0.35 -0.29 -0.18 +0.17

Activated and anaerobic digester sludge

Acetate 10.3

6.0 2.0 0.6

[8]

-0.4 -0.2 0

Exoelectrogenic culture enriched on acetate

Acetate 620^a 910^a 730^a

[3]

-0.50 -0.35 -0.05

uncontrolled (1000 Ω)

Domestic wastewater Formate 0^a 89^a 102^a

20^a

[7]

+0.2 -0.3

Anaerobic sludge Acetate > 5.0

< 2.0

[27]

-0.2 0 +0.2 uncontrolled (≤ +0.1)

Exoelectrogenic culture enriched from anaerobic

digester

Xylose 0.08

0.14 1.47 0.40

This study

a A/m³

21 Table 2 Coulombic analysis of MFCs poised with different anode potentials and MFC run with external resistance of 100 Ω. The results are calculated from the periods with stable current production. The results show the distribution of electrons in produced electricity (coulombic efficiency, CE), measured soluble metabolites (acetate, propionate and butyrate), calculated remaining xylose, and other processes (losses).

Anode potential (V vs. Ag/AgCl)

CE (%) Cacetate (%) Cpropionate (%) Cbutyrate (%) Cxylose (%) Closses (%)

-0.2 0.7 ± 0.2 11 ± 2 2.2 ± 1.4 2.6 ± 1.8 28 ± 21 55 ± 21

0 5.2 ± 1.3 7.8 ± 6.9 3.2 ± 2.4 0.8 ± 1.6 49 ± 21 34 ± 25

+0.2 63 ± 15 - - - 66 ± 14 -

Uncontrolled (0 – 0.1) 31 ± 3 2.7 ± 0.5 1.5 ± 2.3 - 55 ± 12 6.2 ± 8.5

22 Table 3 Affiliation of DGGE fragments determined by their 16S rRNA sequence from

exoelectrogenic enrichment cultures grown on xylose and on different poised anode potentials.

Band^a Affiliation (acc)^b Family^c Sim (%)^d SL (bp)^e

A Uncultured bacterium (JQ439904) - 98.4 435

B Ochrobactrum intermedium (KC525437) α-proteobacteria 99-99.8 404-439 C Ruminobacillus xylanolyticum (DQ178248) Ruminobacillus 97.1 342 D Bacteroides sp. (AY554420) Bacteroidetes 100 437-450 E Ruminobacillus xylanolyticum (DQ178248) Ruminobacillus 97.3 400 F Shigella sp. (FJ405328)

Sporosarcina pasteurii (FR719721)

γ-proteobacteria Firmicutes

99.3 437 G Alcaligenes sp. (AY238499) β-proteobacteria 96.1 429 H Alcaligenes faecalis (KF056900) β-proteobacteria 97.7 347 I Desulfitobacterium hafnience (NR_074996) Clostridia 100 428 J Kerstersia gyiorum (JX316030)

Bordetella sp. (AF227829)

β-proteobacteria 98.4-100 382-459 K Citrobacter amalonaticus (KC689304) γ-proteobacteria 100 470 L Azoarcus indigens (GU592532) β-proteobacteria 98.9 453 M Uncultured spirochete (JF736651) Spirochaetes 100 430-431

a Band mark in Fig. A.1, ^b Accession number of closest species in GenBank, ^c Family according to Ribosomal Database project II, ^d Similarity (%), ^e Sequence length (base pairs)

23

Appendix 1:

Effects of anode potentials on bioelectrogenic conversion of xylose and microbial community compositions

Marika E. Kokko*, Annukka E. Mäkinen, Mira L.K. Sulonen, and Jaakko A. Puhakka

a Department of Chemistry and Bioengineering, Tampere University of Technology, P.O.

Box 541, FI-33101Tampere, Finland

* Corresponding author. E-mail address: marika.kokko@tut.fi; Tel.: +358 44 5430862; Fax:

+358 3 3115 2869

24 Figure A.1 Bacterial DGGE profiles of cultures grown on xylose at different poised anode

potentials (vs. Ag/AgCl) or with external resistance of 100 Ω. Samples were taken both from solution (S) and biofilm (B). Inoculum (Inoc.) refers to the enrichment culture that was used to inoculate the MFCs. See Table 3 for the labeled bands.

A> <B

<C D>

E>

F>

<G

<J

<K

<L

<M M>

< D > <B

B>

<B B>

H>

I>

J>

Inoc. 100Ω S

100Ω B

-0.2 V S

-0.2 V B

0 V S

0 V B

+0.2 V B +0.2 V

S