This article was downloaded by: [University of Otago]On: 02 September 2013, At: 00:58Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Environmental TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tent20

A comparison of bioelectricity in microbial fuel cellswith aerobic and anaerobic anodesChih-Yu Chen a , Tzu-Yu Chen b & Ying-Chien Chung ba Department of Tourism and Leisure , Hsing Wu University , Taipei , 244 , Taiwanb Department of Biological Science and Technology , China University of Science andTechnology , Taipei , 115 , TaiwanPublished online: 27 Aug 2013.

To cite this article: Environmental Technology (2013): A comparison of bioelectricity in microbial fuel cells with aerobic andanaerobic anodes, Environmental Technology, DOI: 10.1080/09593330.2013.826254

To link to this article: http://dx.doi.org/10.1080/09593330.2013.826254

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Environmental Technology, 2013http://dx.doi.org/10.1080/09593330.2013.826254

A comparison of bioelectricity in microbial fuel cells with aerobic and anaerobic anodes

Chih-Yu Chena, Tzu-Yu Chenb and Ying-Chien Chungb,∗

aDepartment of Tourism and Leisure, Hsing Wu University, Taipei 244, Taiwan; bDepartment of Biological Science and Technology,China University of Science and Technology, Taipei 115, Taiwan

(Received 7 April 2013; accepted 10 July 2013 )

Microbial fuel cells (MFCs) can, besides running on wastewater, also derive energy directly from certain aquatic plants.However, few studies have focussed on electricity generation using aerobic anodes. This study presents a comparison ofthe MFC performances of an anaerobic-anode MFC (ana-MFC) and an aerobic-anode MFC (aa-MFC), and shows theirindividual conditions for optimal operation. Results show that the maximum power density of 7.07 ± 0.45 mW/m2 for theana-MFC occurred at 500 �, whereas the aa-MFC had a maximum power density of 2.34 ± 0.16 mW/m2 at 2200 �. Theana-MFC generally achieved high electricity generation, and the aa-MFC achieved relatively high electricity generationwhen fed with a diluted substrate. In the ana-MFC, the optimal substrate for electricity generation was glucose (fermentablesubstrate); however, glucose and acetic acid (non-fermentable substrate) were both suitable substrates for the aa-MFC.The optimal gas retention times of the ana-MFC and the aa-MFC were 9 and 120 s, respectively. This retention time is animportant limiting factor of electricity generation for the ana-MFC. The aa-MFCs fed with different substrates exhibitednon-significant differences between bacterial communities. We observed the relative diversities of bacterial communitiesin the ana-MFC fed with various substrates. The results of denaturing gradient gel electrophoresis analysis suggest thatOchrobactrum intermedium, Delftia acidovorans, and Citrobacter freundii may be potential electrogenic bacteria. To ourknowledge, this is the first study comparing the MFC performances of anaerobic and aerobic anodes.

Keywords: aeration; aerobic; anaerobic; microbial fuel cells; substrate

1. IntroductionMicrobial fuel cells (MFCs) have recently emerged as newalternative power sources that are more environmentallyfriendly than fossil fuels.[1,2] In MFCs, micro-organismsconvert chemical energy into electricity. Thus, MFCshave the ability to simultaneously produce electricity anddegrade organic contaminants. The electricity generated inMFCs often uses mixed cultures enriched with domesticwastewater, ocean sediment, animal waste, and anaerobicsewage sludge.[1–5] When organic matter in wastewateror sewage sludge is used to generate electricity by MFCs,it is possible to reduce the treatment cost of wastewaterand sewage sludge.[6] A typical MFC design consists oftwo chambers: one is anaerobic (i.e. the anode) and theother is aerobic (i.e. the cathode). In the anaerobic cham-ber, bacteria oxidize the substrate, generating electrons andprotons. The electrons transfer to the anode either by anexogenous electron carrier, the mediator, or directly fromthe bacterial enzymes to the electrode. The protons transferto the cathode chamber.[7,8] However, energy recovery isoften negatively affected by oxygen in the anode becauseoxygen reduces the number of coulombs available forelectrical current.[9] In addition to electron consumption,oxygen reduction has a high standard potential that can

∗Corresponding author. Email: ycchung@cc.cust.edu.tw

result in an increased anode potential, reducing the cellpotential and energy output. Therefore, only a few stud-ies have addressed electricity generation using an aerobicanode.[10–13] The MFC operation under anaerobic-anodecondition has been recognized as a difficult task, especiallyat large scales. Hence, the feasibility of MFC operationunder aerobic anode or under low dissolved oxygen concen-trations becomes a significant study in the present scenario.Quan et al. used aerobic- and anaerobic-enriched MFCs,and tested both anodes under anaerobic environments.[14]Thus, this system was not really an aerobic-anodeMFC (aa-MFC).

Several factors influence MFC performance: the ratesof fuel oxidation and electron transfer to the electrode bythe micro-organisms, substrate (types and concentrations),hydraulic retention time, circuit resistance, proton transportto the cathode through the membrane, and oxygen supplyand reduction in the cathode.[15–17] When resistors con-nect the anode and the cathode of a fuel cell, the potentialof the cell and the electrical resistance affect the current.The electrical resistances have both external and internalparts: the circuitry powered by the fuel cell is external parts,and internal parts are the fuel cell itself.[18] Hence, iden-tifying the polarization curves of MFCs and optimizing

© 2013 Taylor & Francis

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

2 C.-Y. Chen et al.

MFC operation are important tasks for both aerobic- andanaerobic-anode environments.

This study presents a comparison of the efficiency andperformance of electricity production of an anaerobic-anode MFC (ana-MFC) and an aa-MFC. This study alsoshows the feasibility and applicability of the proposed ana-MFC and aa-MFC designs. We examined several factorsthat affect MFC operation in both anaerobic anode and aer-obic anodes: the external resistance of the circuit, substrateconcentration, type of substrate, and aeration rate in thecathode.

2. Experimental2.1. MaterialsThe mixed bacteria were isolated from sludge obtainedfrom a domestic wastewater treatment plant in TaoyuanCounty. Bacterial strains were enriched in Luria-Bertani(LB) broth at 35◦C. The LB broth was used to immo-bilize the bacterial cells on the carbon felt electrode inthe anode. To evaluate the effects of various substrates(e.g. carbon sources) on the electricity production of theMFC, the mineral medium used in this study containedthe following: 0.4 g/L (NH4)2SO4, 2.098 g/L K2HPO4,1.088 g/L KH2PO4, and trace elements (1 L of mediumcontains 735 mg CaCl2·2H2O, 50 mg MgSO4·7H2O, 5 mgMnSO4·H2O, and 125 mg FeSO4·7H2O). In this case,2000 mg/L of various carbon sources (glucose, fructose,sucrose, lactose, acetic acid, propionic acid, butyric acid,glycerol, ethanol, glycine, phenylalanine, methionine, andisoleucine) were added. The catholyte in the MFC includeda 50 mM phosphate buffer solution (pH = 7) and 100 mMsodium chloride.

2.2. MFC reactors and immobilization processThe structure of the mediator-less MFC in this study was atypical dual-chamber unit. The MFC comprised an 8 × 8 ×8 cm acrylic cube. The volumes of the anode and cathodecompartments were 171.5 mL each, and the compartmentswere separated by a cation exchange membrane (Nafion117, Dupont Co., USA). Graphite felt (70 × 70 × 3 mm,GF series, Electro-synthesis Co., USA) was used to formelectrodes, and platinum wire (ID = 0.5 mm) connected theelectrodes through a variable resistor. A 500 mL sampleof LB broth containing 30 mL of sludge (sourced from adomestic wastewater treatment plant in Taoyuan County)was placed in the glass bottle and continuously recycledin the anode compartment in the MFC by a submersiblepump for cell immobilization at a 10-day retention time.Air was purged into the cathode compartment to supply theO2 needed for the electrochemical reaction. The ana-MFCwas kept anoxic by purging with nitrogen gas, whereas theaa-MFC operated in an aerobic environment supplied withO2 through continuous aeration. The external resistance of

MFC was set at 1500 � during the immobilization period,and gas retention time in the cathode was 60 s.

2.3. Effect of operation parameters on electricitygenerated by ana-MFC and aa-MFC

External resistance: the circuit was adjusted using variableresistance (270–10,000 �) to control the current produc-tion and determine the power generation as a function ofthe load. The medium consisted of original LB (chemicaloxygen demand (COD): 256,000 mg/L), 1/10 LB (COD:25,600 mg/L), 1/100 LB (COD: 2560 mg/L), and 1/1000LB (COD: 256 mg/L). The organic loading rate (OLR)ranged from 0.021952 to 21.952 g-COD/day in the batchmode. The substrate types included carbohydrates (glucose,fructose, sucrose, and lactose), simple organic acids (aceticacid, propionic acid, and butyric acid), amino acids (glycine,phenylalanine, methionine, and isoleucine), and alcohol(glycerol and ethanol) (2000 mg/L). Gas retention times of6–300 s were tested for oxygen aeration in the cathode andthe mineral medium containing glucose (256,000 mg/L)was taken as the anolyte. The MFC was operated in batchmode for 2 days for each operation parameter until it exhib-ited a stable current production. The MFCs were placed ina temperature-controlled chamber controlled at 35◦C.

2.4. AnalysisThe potential between the anode and cathode was measuredusing a multimeter (Model 2700, Keithley Instruments Inc.,OH, USA). Data were recorded digitally on a personal com-puter using an interface card (Model PCI-488, KeithleyInstruments Inc.) every minute. After every new fuel feed-ing or operating condition, the potential signal fluctuatedbefore reaching a new steady state, and the response at thesteady state was averaged. The current (I ) was calculatedat a resistance (R) from the voltage (V ) as I = V /R. Thepower (P) was calculated as P = IV , and normalized bythe surface area of the anode. The power density (mW/m2)and current density (mA/m2) were calculated by relating thepower and current with the surface area (m2) of the anode,respectively. All experiments were conducted using threeseparate MFCs, and all sample analyses were conducted intriplicate, and the mean values were calculated.

Denaturing gradient gel electrophoresis (DGGE) anal-ysis of gel patterns was used to monitor changes in thebacterial community of the anode of the MFC for varioussubstrates. The structure and intensity of the bacterial com-munity were analysed using a DGGE apparatus (Bio-Rad,Hercules, CA) and Bio-Rad’s image program (QuantityOne 4.5.0). The DGGE gels consisted of 8% acrylamidegel with a 45–60% denaturant gradient at 60◦C and a con-stant voltage of 100 V for 16 h. Cell lysis, DNA extraction,and polymerase chain reaction (PCR) amplification wereperformed as described by Chung.[19] PCR primers forF968GC and R1401 were used to amplify the segment of

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

Environmental Technology 3

0

20

40

60

80

100

120

140

160

180

200

0 10 20 30 40 50 60Time (h)

Vol

tage

(m

V)

anaerobic anode MFC

aerobic anode MFC

Figure 1. Voltage during the operation of ana-MFC and aa-MFCwith the function of time (operating temperature: 35◦C; pHof anolyte: 7.0; external resistance: 1500 �; anolyte: LB; gasretention time in the cathode: 60 s).

eubacterial 16S rDNA. The various bands were identifiedby excising the bands from the DGGE gel and then elut-ing, re-amplifying, and sequencing them. The analysis ofthe DNA sequences and homology searches in this studywere submitted for comparison to the GenBank databaseby using the BLAST algorithm.

3. Results and discussion3.1. The potential change between the ana-MFC and

the aa-MFCAfter a 42-day immobilization process, the potentialreached a steady state in the MFCs, suggesting that thebiofilm in the MFC anode was mature. Hence, the feedsolution was replaced with fresh LB broth and the sys-tem was operated in batch mode. Both MFCs were testedunder 1500 � external resistance. Figure 1 shows that thevoltages of both MFCs immediately increased, dramati-cally declined, slowly rose, and reached a new steady statebetween 24 and 36 h. These results show that the volt-age of the ana-MFC was higher than that of the aa-MFC.At the steady state, the ana-MFC and aa-MFC exhibitedan average of 174.6 ± 1.2 mV (4.14 ± 0.9 mW/m2, 23.7 ±1.5 mA/m2) and 107.8 ± 0.6 mV (1.57 ± 0.5 mW/m2,14.6 ± 1.2 mA/m2), respectively. The microbial growthrate was more rapid under aerobic oxidation than underanaerobic fermentation. This means that the fast growth rateof bacteria also causes rapid substrate degradation underaerobic conditions. However, the experimental results ofthis study indicate that a higher oxidation rate does not

0

50

100

150

200

250

300

350

400

0 10 20 30 40 50 60 70

Current density (mA/m2)

Vol

tage

(m

V)

1

2

3

4

5

6

7

8

Pow

er d

ensi

ty (

mW

/m2 )

Voltage

Power density

(a)

0

50

100

150

200

250

300

0 5 10 15 20 25Current density (mA/m2)

Vol

tage

(m

V)

1

1.2

1.4

1.6

1.8

2

2.2

2.4

Pow

er d

ensi

ty (

mW

/m2 )

Voltage

Power density

(b)

Figure 2. Voltage and power density produced from (a)ana-MFC and (b) aa-MFC at 270–10,000 � external resistances(operating temperature: 35◦C; pH of anolyte: 7.0; anolyte: LB;gas retention time in the cathode: 60 s).

lead to higher electricity production. This might be becauseof the less-than-expected electron transfer from the anode tothe cathode in the aa-MFC. Ringeisen et al. suggested thatthe decrease in voltage of the aa-MFC is likely becauseof the presence of dissolved oxygen in the anode.[10]This oxygen scavenges electrons to form water, therebyreducing the number of electrons transferred to the cath-ode. Sharma and Li suggested that bacterial growth mayconsume substrates instead of generating electricity.[20]

3.2. MFC performance as a function of current densityThis study shows that the voltage and power density of theMFCs were a function of the measured steady-state currentsunder various external resistances. Figure 2 shows that thepotential decreased as the current density increased, andthe ana-MFC and aa-MFC achieved maximum potentialsof 340.75 ± 2.5 and 254.96 ± 1.2 mV, respectively. Thedifferences in MFC performance with diverse external resis-tances may be associated with variations in activation losses

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

4 C.-Y. Chen et al.

in the MFC. These activation losses are a function of elec-trochemical activity of anode-reducing micro-organisms.These results also show the relationship between the powerdensity and current density (external resistance). It indicatedthat the maximum power density was 7.07 ± 0.45 mW/m2

for the ana-MFC at 500 � and 2.34 ± 0.16 mW/m2 forthe aa-MFC at 2200 �. Apparently, the ana-MFC achieveda higher electricity output than the aa-MFC did. A possi-ble reason for this increase is the high internal resistances(approximately 4.4 times) in the aa-MFC. Thus, we testedthe ana-MFC and the aa-MFC under external resistances of500 and 2200 �, respectively, in the following experiments.

3.3. Effects of substrate concentration on electricitygeneration

Figure 3 and Table 1 show that the electricity generationincreases with the substrate concentration or OLR, regard-less of the MFC type. This trend is in agreement withprevious research showing that increased organic loadingenhances the power generation of MFCs.[21,22] The highpower generation occurred at a high organic concentrationor OLR may be the result of the increased bioactivity ofelectrogenic bacteria and the increased conductivity of theanode solution.[20,23]

The voltage output of the ana-MFC was an average of1.53 times greater than that of the aa-MFC when the sub-strate concentration was 1–1/10 LB. When the substrateconcentration was 1/100 LB, the voltage output of the ana-MFC was only 1.22 times that of the aa-MFC. However,when the substrate concentration was reduced to 1/1000

0

50

100

150

200

250

300

LB 1/10 LB 1/100 LB 1/1000 LB

LB concentration

Vol

tage

(m

V)

ana-MFC

aa-MFC

Figure 3. Effect of substrate concentration on voltage generationin ana-MFC and aa-MFC (operating temperature: 35◦C; pH ofanolyte: 7.0; external resistance: 500 � for ana-MFC and 2200 �for aa-MFC; anolyte: LB; gas retention time in the cathode: 9 s forana-MFC and 120 s for aa-MFC).

LB, the voltage output of the ana-MFC was only 0.6 timesthat of the aa-MFC. Jadhav and Ghangrekar indicated thatelectricity generation was directly proportional to substrateconsumption.[24] Thus, the low voltage observed in theana-MFC was because the 1/1000 LB substrate concen-tration was too low to generate enough ATP. The resultssuggest that a concentrated substrate favours electricity gen-eration in the ana-MFC. Conversely, a diluted substratefavours electricity generation in the aa-MFC.

Table 1. List of average voltage, power density, and current density of aa-MFC and ana-MFC under different operatingconditions.

aa-MFC ana-MFC

Power Current Power CurrentVoltagea density density Voltage density density

(mV) (mW/m2) (mA/m2) (mV) (mW/m2) (mA/m2)

LB concentration (mg/L)/OLR (g-COD/day)256,000/21.952 165.3 ± 3.5 2.53 ± 0.3 15.3 ± 0.8 251 ± 2.8 25.7 ± 2.6 102.5 ± 6.425,600/2.1952 121.0 ± 0.8 1.36 ± 0.5 11.2 ± 0.4 182 ± 1.6 13.5 ± 1.2 74.3 ± 2.52560/0.21952 58.5 ± 1.6 0.32 ± 0.03 5.42 ± 0.2 71.8 ± 1.4 2.1 ± 0.8 29.3 ± 1.3256/0.021952 20.6 ± 0.5 0.04 ± 0.005 1.91 ± 0.05 12.3 ± 0.6 0.06 ± 0.008 5.02 ± 0.3Substrate type (2000 mg/L)Glucose 70.0 ± 1.5 0.46 ± 0.05 6.5 ± 0.8 29.1 ± 0.8 0.35 ± 0.04 11.9 ± 1.0Glycine 34.5 ± 0.5 0.11 ± 0.02 3.2 ± 0.9 15.7 ± 0.3 0.10 ± 0.02 6.4 ± 1.1Glycerol 51.7 ± 1.2 0.25 ± 0.03 4.8 ± 0.7 17.4 ± 0.3 0.12 ± 0.01 7.1 ± 1.4Acetate 66.8 ± 1.8 0.41 ± 0.06 6.2 ± 0.6 16.9 ± 0.5 0.11 ± 0.01 6.9 ± 0.6Gas retention time (s)9 205.8 ± 5.8 3.93 ± 0.6 19.1 ± 2.6 152.6 ± 1.3 9.51 ± 1.2 62.3 ± 1.660 206.9 ± 6.2 3.97 ± 0.5 19.2 ± 2.2 132.5 ± 0.8 7.17 ± 0.6 54.1 ± 2.4120 223.1 ± 7.1 4.62 ± 0.8 20.7 ± 2.4 129.1 ± 1.2 6.80 ± 0.4 52.7 ± 2.2180 195.1 ± 1.2 3.53 ± 0.3 18.1 ± 2.8 120.3 ± 0.6 5.91 ± 0.8 49.1 ± 2.8300 184.3 ± 2.5 3.15 ± 0.5 17.1 ± 1.2 106.6 ± 1.2 4.64 ± 0.6 43.5 ± 3.6

Note: aCalculated from values at the plateau of output voltage.

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

Environmental Technology 5

3.4. Effect of substrate types on current generationThe experiments in this study were conducted to investigatevarious types of carbon sources, including carbohydrates,amino acids, short-chain fatty acids, and alcohol at thesame OLR (0.1715 g/day). These results show a relativelyhigh current density in the ana-MFC compared with the aa-MFC when the same substrate was used (except for theshort-chain fatty acids: data not shown). The ana-MFChad a 10–12 mA/m2 current density for carbohydrates,5.5–6.5 mA/m2 current density for amino acids, 5.2–6.8mA/m2 current density for alcohol, 1.8–2.1 mA/m2 cur-rent density for propionic acid and butyric acid, and6.8 mA/m2 current density for acetic acid. The aa-MFChad a 3.5–6.5 mA/m2 current density for carbohydrates,1.2–3.5 mA/m2 current density for amino acids, 2.8–4.8mA/m2 current density for alcohol, and 5.1–6.2 mA/m2

current density for short-chain fatty acids.Figure 4 shows the effects of representative substrates

on current generation in both MFCs and Table 1 lists theMFC’s performance at different feeding substrates. In theana-MFC, glucose was the optimal substrate for electricitygeneration and its substrate utilization rate (SUR) achieved0.138915 g/day. This finding is consistent with previousresearch by Chae et al.[25] In the aa-MFC, both glucoseand acetic acid were suitable substrates for electricity gen-eration and their SURs were 0.14749 and 0.145775 g/day,respectively. Substrate types significantly affected electric-ity generation.[26] In this study, the electricity generationand SUR showed a linear relationship (data not shown).Among the compounds tested in this study, propionic acidand butyric acid had difficulty generating current in the ana-MFC, which is consistent with Xing et al.[27] However,propionic acid and butyric acid easily generated currentin the aa-MFC. This is likely because of the various sub-strate types (fermentable or non-fermentable) and the typesof electrogenic bacteria in the MFCs. Fermentable sub-strates (e.g. glucose, fructose, lactose, and sucrose) wereeasily converted into simple organics under the anaerobiccondition.[28] Thus, the ana-MFC produced a high currentdensity when fed with different carbohydrate types. Con-versely, the aa-MFC generated a high current density fornon-fermentable substrates (short-chain fatty acids), whichwere easily metabolized under aerobic conditions. Besides,since the diluted substrate (2000 mg/L) was unfavourableto electricity generation of the ana-MFC, the power densitywas low in the ana-MFC (Table 1).

3.5. Effects of aeration in the cathode compartment oncurrent generation

Aeration affected the amount of dissolved oxygen in thecathode compartment of the MFCs in this study. Introduc-ing a higher gas flow rate into the cathode compartmentproduced higher levels of dissolved oxygen.[29] The opti-mal gas retention times of the ana-MFC and the aa-MFC

0

2

4

6

8

10

12

14

Substrate

Cur

rent

den

sity

(m

A/m

2 )

glucose glycine glycerol acetate

(a)

0

1

2

3

4

5

6

7

8

Substrate

Cur

rent

den

sity

(m

A/m

2 )

glucose glycine glycerol acetate

(b)

Figure 4. Effect of substrate types on current density in (a)ana-MFC and (b) aa-MFC (operating temperature: 35◦C; pHof anolyte: 7.0; external resistance: 500 � for ana-MFC and2200 � for aa-MFC; anolyte: mineral medium with various car-bon sources; gas retention time in the cathode: 9 s for ana-MFCand 120 s for aa-MFC).

were 9 and 120 s, respectively (Figure 5 and Table 1). Thegas retention time significantly affected the current den-sity of the ana-MFC (43.54–62.26 mA/m2 with a 30.1%variation), but only slightly affected that of the aa-MFC(17.82–20.71 mA/m2 with a 14.0% variation). Also, theeffect of gas retention time on the power density of MFCalso revealed a similar tendency (Table 1). This differencemay be attributed to more electrons being transferred tothe cathode in the ana-MFC, producing a faster oxygenconsumption rate in the ana-MFC than in the aa-MFC.Thus, the gas retention time for the cathode compartmentof the ana-MFC is a limiting factor of current generation.Besides, the proton exchange membrane usually played arole in stopping the passage of air from cathode to the anodecompartment. Sevda et al. have demonstrated the advan-tages of MFC with Zirfon®, and the membrane might bea good option for the future scale up of MFCs.[30] Theair cathode MFCs developed by Sevda et al. presented the

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

6 C.-Y. Chen et al.

0

10

20

30

40

50

60

70

0 30 60 90 120 150 180 210 240 270 300 330

Gas retention time(s)

Cur

rent

den

sity

(m

A/m

2 )

ana-MFC

aa-MFC

Figure 5. Effect of gas retention time on current density inana-MFC and aa-MFC (operating temperature: 35◦C; pH ofanolyte: 7.0; external resistance: 500 � for ana-MFC and 2200 �for aa-MFC; anolyte: glucose solution).

potential to treat the wastewater with high COD and producebioelectricity as well.[31]

3.6. Bacterial community in the MFCThe DGGE technique was used to analyse the bacterialcommunities of the MFCs in this study. Pieces of elec-trodes from ana-MFCs enriched with various substrates(glycine, glycerol, and glucose) were used to extract thewhole genomic DNA. Figure 6 shows the results of bacterialcommunities in the MFCs for various substrates. The glu-cose samples shown in Lane 3 shared more bands than otherlanes. However, the aa-MFC showed non-significant differ-ences between bacterial communities (data not shown); thismay be because the active electrode-reducing communityis a much smaller portion of the community that is usingup the oxygen and these organisms are not seen by DGGEanalysis. These results show that the electricity-producingbacteria in the ana-MFC favoured a fermentable substrateof glucose as the carbon source.

To identify the differences in bacterial diversity amongsamples, the discriminable bands were excised, PCR-amplified, and sequenced. Table 2 lists the phylogeny,closest relatives, nucleotide sequence similarity, and rel-ative abundance of sequenced DGGE bands. Eight dis-criminable bands (A–H) were individually identified asmembers of different eubacterial phyla. Two bands (A andF) were grouped with the phylum γ -Proteobacteria, namelyPseudomonas putida and Citrobacter freundii, respectively.Two bands (B and D) were clustered within the phylumβ-Proteobacteria, namely Alcaligenes faecalis and Delf-tia acidovorans, respectively. Four other bands (C, E, G,and H) were clustered within the phylum α-Proteobacteria,

Figure 6. DGGE analysis of 16S rDNA amplified from thegenomic DNA of the ana-MFCs enriched with different sub-strates (glycine, glycerol, and glucose). Lane 1: enriched withglycine; Lane 2: enriched with glycerol; and Lane 3: enriched withglucose.

ε-Proteobacteria, Actinobacteria, and Firmicutes, and theirclosest relative showed homology to Ochrobactrum inter-medium, Arcobacter butzleri, Microbacterium laevanifor-mans, and Clostridium acetobutylicum, respectively. Theseresults clearly show that the phylum Proteobacteria pre-dominated and accounted for 87.3–100% of the ana-MFCbacteria for various substrates. Oh et al. presented similarresults.[32]

Based on the presence of their DGGE bands, P. putida,A. faecalis, and O. intermedium (bands A–C) were consis-tently present in various substrates. They or their relativestrains are capable of degrading glycine, glycerol, andglucose.[33–35] These characteristics provide a possibleexplanation for the predominance of P. putida, A. faecalis,and O. intermedium in various substrates. A second bac-terial group, including D. acidovorans, A. butzleri, andC. freundii (bands D–F), appeared in glycerol and glucosesubstrates. These strains can metabolize both substrates,[36–38] thereby explaining the presence of D. acidovorans,A. butzleri, and C. freundii in the ana-MFC. Microbac-terium sp. has a strong ability to ferment glucose but aweak ability to ferment glycerol and glycine.[39] Hence,M. laevaniformans (band G) appeared only in the glucose

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

Environmental Technology 7

Table 2. Nucleotide sequence similarity and relative abundance of sequenced DGGE bands.

Banda Phylogeny Closest relatives Similarity (%) Relative abundance of bands (%)

Glycine Glycerol Glucose

A γ−Proteobacteria Pseudomonas putida 99.3 23.92 12.62 15.91B β−Proteobacteria Alcaligenes faecalis 98.6 14.50 16.81 7.82C α−Proteobacteria Ochrobactrum intermedium 99.2 51.38 32.95 38.81D β−Proteobacteria Delftia acidovorans 96.5 0 8.43 4.65E ε−Proteobacteria Arcobacter butzleri 96.7 0 14.35 15.29F γ−Proteobacteria Citrobacter freundii 98.9 0 15.84 4.82G Actinobacteria Microbacterium laevaniformans 99.1 0 0 9.14H Firmicutes Clostridium acetobutylicum 99.3 10.20 0 3.56

Note: aThe bands are designated as shown in Figure 6.

substrate. The C. acetobutylicum was unable to grow onglycerol as the sole carbon source,[40] which explainedits absence in the glycerol substrate. The strains found inthe ana-MFCs, including P. putida, A. faecalis, A. butz-leri, M. laevaniformans, and C. acetobutylicum, confirmtheir power-generation ability.[14,37,41–43] Conversely,O. intermedium, D. acidovorans, and C. freundii mightbe identified as electrogenic bacteria but a further studyis required.

4. ConclusionsIn this study, the power density of the ana-MFC was higherthan that of the aa-MFC. This difference may be attributedto the types of electrogenic bacteria, internal resistances inthe MFC, and presence of dissolved oxygen in the anode.A concentrated substrate favours electricity generation inthe ana-MFC. Conversely, a diluted substrate is suitable forgenerating electricity in an aa-MFC. The substrate typessignificantly affect current generation. Fermentable andnon-fermentable substrates are suitable for ana-MFCs andaa-MFCs, respectively. The gas retention time for the cath-ode compartment of the ana-MFC is a limiting factor ofelectricity generation, but the aa-MFC performance is inde-pendent of this factor. Diversities of bacterial communitiesare non-significant in the aa-MFC for various substrates, butthey vary in the ana-MFC. This study shows that O. inter-medium, D. acidovorans, and C. freundii may be newbioelectricity-producing species. The electricity productionability of the proposed design requires further research toconfirm the role of individual metabolic functions.

FundingThis study was partially supported by a grant from theNational Science Council, ROC [NSC 99-2313-B-157-001-MY3].

References[1] Liu H, Cheng S, Logan BE. Power generation in fed-batch

microbial fuel cells as a function of ionic strength, tem-perature, and reactor configuration. Environ Sci Technol.2005;39:5488–5493.

[2] Kim M, Hyun MS, Gadd GM, Kim GT, Lee SJ, Kim HJ.Membrane-electrode assembly enhances performance of amicrobial fuel cell type biological oxygen demand sensor.Environ Technol. 2009;30:329–336.

[3] Huang L, Cheng S, Rezaei F, Logan BE. Reducing organicloads in wastewater effluents from paper recycling plantsusing microbial fuel cells. Environ Technol. 2009;30:499–504.

[4] Zhang L, Mao Y, Ma J, Li D, Shi H, Liu Y, Cai L. Effect of thechemical oxidation demand to sulfide ratio on sulfide oxida-tion in microbial fuel cells treating sulfide-rich wastewater.Environ Technol. 2013;34:269–274.

[5] Pant D, Singh A, Bogaert GV, Olsen SI, Nigam PS, Diels L,Vanbroekhoven K. Bioelectrochemical systems (BES) forsustainable energy production and product recovery fromorganic wastes and industrial wastewaters. R Soc Chem Adv.2012;2:1248–1263.

[6] Rulkens W. Sewage sludge as a biomass resource for the pro-duction of energy: overview and assessment of the variousoptions. Energy Fuel. 2008;22:9–15.

[7] Min B, Cheng S, Logan BE. Electricity generation usingmembrane and salt bridge microbial fuel cells. Water Res.2005;39:1675–1686.

[8] Xiao B, Yang F, Liu J. Enhancing simultaneous electricityproduction and reduction of sewage sludge in two-chamberMFC by aerobic sludge digestion and sludge pretreatments.J Hazard Mater. 2011;189:444–449.

[9] Harnisch F, Schröder U. Selectivity versus mobility: separa-tion of anode and cathode in microbial bioelectrochemicalsystems. Chem Sus Chem. 2009;2:921–926.

[10] Ringeisen BR, Ray R, Little B. A miniature microbial fuelcell operating with an aerobic anode chamber. J PowerSources. 2007;165:591–597.

[11] Nevin KP, Zhang P, Franks AE, Woodard TL, LovleyDR. Anaerobes unleashed: aerobic fuel cells of Geobactersulfurreducens. J Power Sources. 2011;196:7514–7518.

[12] Mohan SV, Mohanakrishna G, Sarma PN. Effect of anodicmetabolic function on bioelectricity generation and substratedegradation in single chambered microbial fuel cell. EnvironSci Technol. 2008;42:8088–8094.

[13] Biffinger JC, Byrd JN, Dudley BL, Ringeisen BR. Oxy-gen exposure promotes fuel diversity for Shewanella onei-densis microbial fuel cells. Biosens Bioelectron. 2008;23:820–826.

[14] Quan XC, Quan YP, Tao K, Jiang XM. Comparative investi-gation on microbial community and electricity generation inaerobic and anaerobic enriched MFCs. Bioresour Technol.2012;128:259–265.

[15] Gil GC, Chang IS, Kim BH, Kim M, Jang JK, Park HS,Kim HJ. Operational parameters affecting the performance

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013

8 C.-Y. Chen et al.

of a mediator-less microbial fuel cell. Biosens Bioelectron.2003;18:327–334.

[16] Liu H, Ramnarayanan R, Logan BE. Production of elec-tricity during wastewater treatment using a single cham-ber microbial fuel cell. Environ Sci Technol. 2004;38:2281–2285.

[17] Pant D, Bogaert GV, Diels L, Vanbroekhoven K. A reviewof the substrates used in microbial fuel cells (MFCs) for sus-tainable energy production. Bioresour Technol. 2010;101:1533–1543.

[18] Menicucci J, Beyenal H, Marsili E, Veluchamy RA, DemirG, Lewandowski Z. Procedure for determining maximumsustainable power generated by microbial fuel cells. EnvironSci Technol. 2006;4:1062–1068.

[19] Chung YC. Evaluation of gas removal and bacterialcommunity diversity in a biofilter developed to treatcomposting exhaust gases. J Hazard Mater. 2007;144:377–385.

[20] Sharma Y, Li B. The variation of power generation withorganic substrates in single-chamber microbial fuel cells(SCMFCs). Bioresour Technol. 2010;101:1844–1850.

[21] Kim JR, Premier GC, Hawkes FR, Rodrıguez J, Dins-dale RM, Guwy AJ. Modular tubular microbial fuel cellsfor energy recovery during sucrose wastewater treatmentat low organic loading rate. Bioresour Technol. 2010;101:1190–1198.

[22] Jiang D, Curtis M, Troop E, Scheible K, McGrath J, Hu B,Suib S, Raymond D, Li B. A pilot-scale study on utilizingmulti-anode/cathode microbial fuel cells (MAC MFCs) toenhance the power production in wastewater treatment. Int JHydrogen Energy. 2011;36:876–884.

[23] Zhang B, Zhao H, Zhou S, Shi C, Wang C, Ni J. Anovel UASB–MFC–BAF integrated system for high strengthmolasses wastewater treatment and bioelectricity generation.Bioresour Technol. 2009;100:5687–5693.

[24] Jadhav GS, Ghangrekar MM. Performance of microbialfuel cell subjected to variation in pH, temperature, exter-nal load and substrate concentration. Bioresour Technol.2009;100:717–723.

[25] Chae KJ, Choi MJ, Lee JW, Kim KY, Kim IS. Effect of dif-ferent substrates on the performance, bacterial diversity, andbacterial viability in microbial fuel cells. Bioresour Technol.2009;100:3518–3525.

[26] Pant D, Bogaert GV, Porto-Carrero C, Diels L, Van-broekhoven K. Anode and cathode materials characterizationfor a microbial fuel cell in half cell configuration. Water SciTechnol. 2011;63:2457–2461.

[27] Xing D, Zuo Y, Cheng S, Regan JM, Logan BE. Electricitygeneration by Rhodopseudomonas palustris DX-1. EnvironSci Technol. 2008;42:4146–4151.

[28] Jung S, Regan JM. Comparison of anode bacterial commu-nities and performance in microbial fuel cells with differ-ent electron donors. Appl Microbiol Biotechnol. 2007;77:393–402.

[29] Cha J, Choi S, Yu H, Kim H, Kim C. Directly applicablemicrobial fuel cells in aeration tank for wastewater treatment.Bioelectrochemistry. 2010;78:72–79.

[30] Sevda S, Dominguez-Benetton X, Vanbroekhoven K,Sreekrishnan TR, Pant D. Characterization and comparisonof the performance of two different separator types in air-cathode microbial fuel cell treating synthetic wastewater.Chem Eng J. 2013;228:1–11.

[31] Sevda S, Dominguez-Benetton X, Vanbroekhoven K, WeverHD, Sreekrishnan TR, Pant D. High strength wastewatertreatment accompanied by power generation using air cath-ode microbial fuel cell. Appl Energy. 2013;105:194–206.

[32] Oh ST, Kim JR, Premier GC, Lee TH, Kim C, Sloan WT.Sustainable wastewater treatment: how might microbial fuelcells contribute. Biotechnol Adv. 2010;28:871–881.

[33] Wu J, Zhan X, Liu H, Zheng Z. Enhanced production ofcurdlan by Alcaligenes faecalis by selective feeding withammonia water during the cell growth phase of fermentation.Sheng Wu Gong Cheng Xue Bao. 2008;24:1035–1039.

[34] Zuo Y, Xing D, Regan JM, Logan BE. Isolation of theexoelectrogenic bacterium Ochrobactrum anthropi YZ-1 byusing a U-tube microbial fuel cell. Appl Environ Microbiol.2008;74:3130–3137.

[35] Sharma PK, Fu J, Cicek N, Sparling R, Levin DB. Kineticsof medium-chain-length polyhydroxyalkanoate productionby a novel isolate of Pseudomonas putida LS46. Can JMicrobiol. 2012;58:982–989.

[36] Shigematsu T, Yumihara K, Ueda Y, Numaguchi M,Morimura S, Kida K. Delftia tsuruhatensis sp. nov., aterephthalate-assimilating bacterium isolated from activatedsludge. Int J Syst Evol Microbiol. 2003;53:1479–1483.

[37] Fedorovich V, Knighton MC, Pagaling E, Ward FB, Free A,Goryanin I. Novel electrochemically active bacterium phy-logenetically related to Arcobacter butzleri, isolated froma microbial fuel cell. Appl Environ Microbiol. 2009;75:7326–7334.

[38] Xu S, Liu H. New exoelectrogen Citrobacter sp. SX-1 isolated from a microbial fuel cell. J Appl Microbiol.2011;111:1108–1115.

[39] Kim KK, Lee KC, Oh HM, Lee JS. Microbacteriumaquimaris sp. nov., isolated from seawater. Int J Syst EvolMicrobiol. 2008;58:1616–1620.

[40] González-Pajuelo M, Meynial-Salles I, Mendes F, SoucailleP, Vasconcelos I. Microbial conversion of glycerol to 1,3-propanediol: physiological comparison of a natural producer,Clostridium butyricum VPI 3266, and an engineered strain,Clostridium acetobutylicum DG1(pSPD5). Appl EnvironMicrobiol. 2006;72:96–101.

[41] Rabaey K, Boon N, Siciliano SD, Verhaege M, VerstraeteW. Biofuel cells select for microbial consortia that self-mediate electron transfer. Appl Environ Microbiol. 2004;70:5373–5382.

[42] Mathuriya AS, Sharma VN. Electricity generation by Sac-charomyces cerevisae and Clostridium acetobutylicum viamicrobial fuel cell technology: a comparative study. AdvBiol Res. 2010;4:217–223.

[43] Friman H, Schechter A, Nitzan Y, Cahan R. Effect of externalvoltage on Pseudomonas putida F1 in a bio electrochem-ical cell using toluene as sole carbon and energy source.Microbiology. 2012;158:414–423.

Dow

nloa

ded

by [

Uni

vers

ity o

f O

tago

] at

00:

58 0

2 Se

ptem

ber

2013