Journal of Power Sources 281 (2015) 27e33

Contents lists available at ScienceDirect

Journal of Power Sources

journal homepage: www.elsevier.com/locate/jpowsour

Effect of electricity on microbial community of microbial fuel cell

simultaneously treating sulde and nitrate
Jing Cai a, *, Ping Zheng b, Yajuan Xing b, Mahmood Qaisar c
a

College of Environmental Science and Engineering, Zhejiang Gongshang University, Hangzhou 310012, China
Department of Environmental Engineering, Zhejiang University, Hangzhou 310058, China
c
Department of Environmental Sciences, COMSATS Institute of Information Technology, Abbottabad, Pakistan
b

h i g h l i g h t s

Effect of electricity on microbial community on sulde and nitrate removal was studied in MFC.

All of four MFCs showed similar good capacity to remove substrate simultaneously.

The MFCs displayed different characteristics in electricity generation.

Signicant correlation was existed between Richness of community and generated electricity.

PCA showed that the two MFCs suffered current shock showed similar suspension communities.

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 3 November 2014
Received in revised form
15 January 2015
Accepted 18 January 2015
Available online 28 January 2015

The effect of electric current on microbial community is explored in Microbial Fuel Cells (MFCs)
simultaneously treating sulde and nitrate. The MFCs are operated under four different conditions which
exhibited different characteristics of electricity generation. In batch mode, MFCs generate intermittently
high current pulses in the beginning, and the current density is instable subsequently, while the current
density of MFCs in continuous mode is relatively stable. All operational parameters show good capacity
for substrate removal, and nitrogen and sulfate were the main reaction products. Polymerase Chain
Reaction-Denaturing Gradient Gel Electrophoresis (PCR-DGGE) analysis is employed to obtain proles of
the bacterial communities present in inoculum and suspension of four MFCs. Based on the community
diversity indices and Spearman correlation analyses, signicant correlation exists between Richness of
the community of anode chamber and the electricity generated, while no strong correlation is evident
between other indexes (Shannon index, Simpson index and Equitability index) and the electricity.
Additionally, the results of Principal Component Analysis (PCA) suggest that MFCs suffering from current
shock have similar suspension communities, while the others have diverse microbial communities.
2015 Elsevier B.V. All rights reserved.

Keywords:
Microbial fuel cells
Anaerobic sulde and nitrate removal
Microbial community
Diversity index
Principal Component Analysis

1. Introduction
Microbial fuel cells (MFC) offer a novel approach in the eld of
wastewater treatment, which can generate electricity through
degradation of the substrates [1]. The process has been applied for
the treatment of domestic wastewater in 1991 [2]. MFC technology
is gaining popularity and most of the studies involved the use of
wastewater containing organic substrates like glucose, acetate,

* Corresponding author. College of Environmental Science and Engineering,

Zhejiang Gongshang University, No.18 Xuezheng Street, Hangzhou, Zhejiang Province, China.
E-mail address: caijing@zju.edu.cn (J. Cai).
http://dx.doi.org/10.1016/j.jpowsour.2015.01.165
0378-7753/ 2015 Elsevier B.V. All rights reserved.

sucrose, etc [3].

Generally, wastewaters contain both organic and inorganic
substances such as sulde. Sulde-containing waste streams are
generated by many industries such as tanneries, petrochemical
plants, viscose rayon factories etc. [4]. Sulde is a toxic substance,
which exerts various toxicological inuences on human health and
environmental ecology [5]. The sulde is treated by various physical, chemical and biological technologies. The biological processes
are cost effective as they operate under natural ambient conditions
without any requirement for expensive chemicals and catalysts [6].
It has been demonstrated that nitrate may be employed in the
control of sulde species under anoxic or anaerobic conditions
mediated by some bacterial species [7]. For such reasons, the

28

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

simultaneous anaerobic sulde and nitrate removal process has

been recently developed.
0

5HS 2NO
3 /5S N2 6H2 O 7e

DGqm 1264 KJ mol1

E0 1:87 V

Based on stoichiometric considerations, the simultaneous

anaerobic sulde and nitrate removal may be accomplished in
MFCs. However; a few studies have reported this process in MFCs so
far. Lee et al. studied the interactions between denitrifying sulde
removal in a MFC process and conrmed that the MFC was capable
of the simultaneous sulde and nitrate removal using monoculture
of Pseudomonas sp. C27 [8]. We operated a two-chambered MFC
simultaneously treating sulde and nitrate using anaerobic activated sludge process [9,10]. It was concluded that electricity output
enhanced through the microbial enrichment with the operation;
suggesting that an active bacterial consortium capable of substrate
removal and electricity generation had established in the MFCs.
Microorganisms of the anodic chamber play an important role in
substrate removal and electricity generation in MFCs [11]. Many
researchers have investigated the microbial communities on the
electrode surface and those in the anode chamber of MFCs [12e15].
The microbial communities involved in MFC have been characterized in the previous work, generally studies focused on variations of
substrate, reactor design or monocultures of novel organisms [16].
It is well established fact that the microorganism may affect the
electricity production in a MFC. It is speculated that the electricity
would also have adverse inuence on the community of suspended
microorganisms in MFCs. Very little information on the effect of
electricity on the microbial community of the MFCs is available to
date. However, bacterial communities associated with MFC treating
sulde or nitrate in the wastewater are well established [17e19],
still information on the microorganism responsible for simultaneous sulde and nitrate removal in the MFC is limited.
The objective of the current study was to investigate the effect of
electricity on microbial community in MFCs treating sulde and
nitrate removal simultaneously. Four MFCs were operated under
different conditions, in order to assess the impact of electricity.
Denaturing Gradient Gel Electrophoresis (DGGE) of polymerase
chain reaction (PCR) amplied 16S ribosomal RNA (rRNA) gene
fragments to obtain proles of the bacterial communities in inoculum and suspended sludge of four MFCs. Based on the community
diversity indices (such as Richness, Dice index, Shannon Index,
Simpson index and Equitability index) to present a detailed analysis
of the community structure. Spearman correlation analyses were
aimed at understanding the relationship between community diversity and electricity generation. Moreover, Principal Component
Analysis (PCA) was used to investigate visual effects of electricity on
the microbial community diversity in the anodic chamber.

concentrations were administered as KNO3 and Na2S$9H2O,

respectively; their concentrations were controlled according to the
requirements of the experiment conducted. The inuent pH was
maintained at 7.0 0.1 throughout the experiment. The medium in
cathode chamber was a mixture of 50 mM PBS buffer (pH 7.0) and
100 mg L1 KMnO4 solution, which was recycled over the cathode
compartment through a 2.5 L external buffer vessel by peristaltic
pumps.
2.2. MFC operation
Four MFCs of same conguration were used in the experiment,
which were denoted as MFC1, MFC2, MFC3 and MFC4, respectively.
The MFCs were operated under different modes with diverse
electrode materials. MFC1 was operated in continuous mode with
graphite rod as electrodes, MFC2 was operated in batch mode
containing graphite rod electrodes, and MFC3 was operated in
continuous mode having graphite felt as electrodes, while MFC4
was also operated in batch mode equipped with graphite felt as
electrodes. All MFCs were operated at room temperature
(25 2  C).
While operated in batch mode, 200 mL synthetic wastewater
was fed to the anode chamber on daily basis. Sulde (540 mg L1)
was added after the anodic chamber was ushed with N2 for 5 min
to remove O2 from the solution. The nitrate concentration was
added at S/N molar ratio of 5:2 based on stoichiometry of the
chemical reaction.
During the MFCs operation in continuous mode, a peristaltic
pump fed the synthetic wastewater to the system. Inuent substrate concentrations were kept constant (300 mg L1 and
52.5 mg L1, respectively), and the Hydraulic Retention Times
(HRTs) were maintained around 10.6 h.
2.3. Analytical procedures
The inuent and efuent nitrate-nitrogen, pH and sulde were
analyzed during the operation of MFC. Nitrate-nitrogen (NO-3-N)
was analyzed through ultraviolet spectrophotometric screening
method on daily basis using spectrophotometer (Unico UV-2102 PC
and 722S, China) [21]. The sulde was determined by iodometric
method and sulfate was measured through turbimetric method
[21]. The pH was determined following standard method [21]. A
three-point calibration of pH meter was performed on daily basis
before pH determination. Total solids (TS) concentration was
determined according to gravimetric method at 103  C and volatile
solids were analyzed through gravimetric method at 550  C [21].
Voltage across the 1000 U resistor was recorded at an interval of
every 10 min using a digital acquisition system (Agilent 34970A
Data Acquisition/Switch Unit). Current density was normalized by
the net surface areas of anode electrode.

2. Materials and methods

2.4. DNA extraction and PCReDGGE analyses

2.1. MFC construction

The activated sludge was collected from the anodic chamber of

four MFCs during their optimal operation to analyze the dominant
bacterial species. The activated sludge and inoculum were washed
several times with the phosphate buffer solution before DNA
extraction. Genomic DNA was extracted using 3S DNA isolation Kit
for environmental samples (Shanghai Biocolors Company, China)
according to the manufacturer's instructions. Quality of DNA extracts was detected by 1% (wt/vol) agarose gel electrophoresis. The
extracted DNA was preserved at 20  C.
The DGGE bacterial primers were GC-BAC338-F (ACT CCT ACG
GGA GGC AG), added to the 50 end of a 40-bp GC clamp (50 -CGC CCG
CCG CGC CCC GCG CCC GTC CCG CCC CCG CCC-30 ), and BAC805-R

The MFCs consisted of anodic and cathodic chambers, each

could accommodate total volume of 350 mL (300 mL net volume) as
reported in our previous study [9]. Inoculum was collected from a
UASB reactor simultaneously treating sulde and nitrate operated
under anaerobic lithoautotrophic conditions. The enriched sludge
(100 mL) was inoculated in the anodic chambers of the MFCs. The
anode chambers was also lled with synthetic inuent containing
NaHCO3, MgCl2, KH2PO4, (1 g L1 each), (NH4)2SO4 (0.24 g L1) and
trace element solution (1 mL L1). The trace element solution was
prepared according to Mahmood et al. [20]. The nitrate and sulde

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

(GAC TAC CAG GGT ATC TAA TCC), which targeted the 16S rRNA
gene of bacteria and generated a PCR product of 468 bp [22]. PCR
reactions were performed in 25 mL reaction volumes containing
0.25 mL Taq DNA polymerase (Takara), 2.5 mL 10  PCR reaction
buffer (Takara), 2 mL dNTPs mixture (2.5 mM) (Takara), 0.75 mL of
each of the primers (20 mM) (Takara), and 10e100 ng of the
extracted DNA. The PCR conditions were as follows: an initial
denaturation at 94  C for 5 min; 35 cycles of denaturation (40 s at
94  C), annealing (40 s at 55  C) and extension (30 s at 72  C); a nal
extension at 72  C for 10 min and then stored at 4  C. The amplied
products were checked on 1% (m/v) agarose TAE gels and nally
visualized under UV light [23]. Then the DGGE prole was analyzed
by Quantity One Software (BioRad Company).
The PCR products were separated using DGGE, performed with
the Bio-Rad DCode Universal Mutation Detection System (BioRad, USA) on 8% (w/v) polyacrylamide gels with a gradient of
denaturant 30e60%. Electrophoresis was carried out at 60  C, 110 V
for 10 h (with an initial electrophoresis for 30 min at 60 V) in
1  TAE buffer. Polyacrylamide gels were stained by silver staining
and scanned on gel documentation system (Gel Doc-XR; Bio-Rad).
The dominant bands excised and eluted from the gels were reamplied using DGGE primer set without GC clamp and then
cloned and sequenced. The sequences were compared with the
existing sequences of the National Center for Biotechnology Information database (NCBI) using the BLASTN program. The MEGA
Software version 6.0 was used to construct phylogenetic trees
based on the neighbor-joining method. Bootstrap re-sampling
analysis of 1000 replicates was performed to estimate the degree
of condence in the tree topologies [24]. All nucleotide sequences
data reported in this study were deposited in the GenBank database
under accession numbers KM925019-KM925039.

2.5. Microbial community diversity analyses

Richness is the simplest way to measure the community diversity; which is dened as the number of distinct species present
within a given community [16].
Dice index (Cs) of similarity was used to evaluate the similarity
of bacterial communities of both bioreactors [25]. The equation
used was Cs 2j/(ab), where j was the number of common bands
between samples A and B; a and b were the total number of bands
in samples A and B, respectively.
DGGE prole was also examined using Shannon-Weaver index
of diversity (H) [26]. The index was calculated using the following
equation:

H

N
X
ni=Nlnni=N
i1

The Simpson index l is the sum of the squares of the fractional

species abundances, and is used to assess diversity [26]. This index
was calculated using the following equation:

l

N
X

ni=N2

i1

where ni/N was the proportion of community which was made up

by species i (brightness of the band i/total brightness of all bands in
the lane).
Equitability index (EI) was introduced to analyze the species
distribution during the operational period [27]. This index was
calculated using the following equation:

29

EI H=ln n
where, n is the total number of species in the sample.
3. Results
3.1. Substrate removal in the MFCs
Table 1 showed the substrate removal in the MFCs. MFC1 was
operated in continuous mode containing graphite rod electrodes.
Keeping the inuent substrate concentrations constant (sulde and
nitrate concentrations were 300 mg S L-1and 52.5 mg N L1,
respectively), the HRT was maintained around 10.6 h in MFC1. The
efuent sulde and nitrate concentration were 0.30 mg L1 and
1.23 mg L1, respectively. And sulde and nitrate removal percentages were 99.9% and 97.66%, respectively.
MFC2 was operated in batch mode having graphite rod electrodes, where the inuent sulde and nitrate concentration were
540 mg L1 and 94.5 mg L1, respectively. The efuent sulde and
nitrate concentration were 1.95 mg L1 and 3.34 mg L1, respectively. MFC2 accomplished sulde and nitrate removal percentages
of 99.64% and 96.47%, respectively.
MFC3 was operated in continuous mode containing graphite felt
electrodes, while the other operating conditions were similar to
MFC1. The efuent sulde and nitrate concentration were
0.25 mg L1 and 1.13 mg L1, respectively. Sulde and nitrate
removal percentages of MFC3 were 99.92% and 97.07%, respectively.
MFC4 was operated in batch mode with graphite felt as electrodes; its operating conditions were similar to MFC2. The efuent
sulde and nitrate concentrations were 0.85 mg L1 and
0.04 mg L1, respectively; while its sulde and nitrate removal
percentages were 99.84% and 99.96%, respectively.
In spite of minor differences in substrate removal of four MFCs,
good capacity was evident to remove sulde and nitrate simultaneously. Sulde and nitrate removal percentages were higher than
99.64% and 96.47%, respectively. In the tested substrate concentrations and HRT, about 68.66%e79.19% sulde in the inuent
converted to sulfate, and about 96.28%e98.44% nitrate in the
inuent transformed to nitrogen. It was inferred that nitrogen and
sulfate were the main end products by microbial communities of
MFCs.
3.2. Electricity generation in the MFCs
Fig. 1 showed the electricity generation in the MFCs. The current
density of MFC1 was stable in the operating period with the
external resistance of 1000 U and the HRT of 10.6 hand it maintained between 24 mA m2 and 35 mA m2 (Fig. 1). The average
current density of MFC1 was 29 mA m2, and the generated Electronic Quantity of MFC1 was 1946.56C m2. The current density
prole of MFC3 was similar to that of MFC1 due to the similarity in
their operational modes [28]. The current density of MFC3 was
maintained between 148 mA m2 and 158 mA m2 during its
operation (Fig. 1). The average current density of MFC3 was
153 mA m2, while its Electronic Quantity of MFC3 was
10313.26C m2.
The proles of current density in MFC2 and MFC4 were entirely
different in comparison to MFC1 and MFC3. The maximum current
density of MFC2 was 456 mA m2, which was observed during the
rst 10 min (Fig. 1). The current density decreased rapidly onwards,
and it dropped to 92 mA m2 in the 4th hour (down by 79.8%).
Subsequently, it slowly descended to about 34 mA m2 in the 14th
hour. Finally, it reached at the steady point (28 mA m2). During
18.5h operation, MFC2 produced 6117.88C m2 of charges. The

30

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

Table 1
Substrate removal in the MFCs.
MFC

Eff. Sulde/(mg L1)

Eff. Sulfate/(mg L1)

Sulde
removal/(%)

Sulfate
production/(%)

Eff. Nitrate/(mg L1)

Eff. Nitrite/(mg L1)

Nitrate
removal/(%)

Nitrogen
production/(%)

MFC1
MFC2
MFC3
MFC4

0.30
1.95
0.25
0.85

220.34
384.18
237.56
370.78

99.90
99.64
99.92
99.84

73.45
71.14
79.19
68.66

1.23
3.34
1.13
0.04

3.77
0.18
0.41
1.43

97.66
96.47
97.07
99.96

90.48
96.28
97.07
98.44

Fig. 1. Electricity generation in the MFCs.

proles of current density for MFC2 were similar with that in MFC4
due to similarity of batch mode. The maximum current density of
MFC4 was 180 mA m2 which appeared in the beginning and lasted
for a short time. The current density rapidly decreased afterwards
and it dropped to a relatively stable value (78 mA m2), which
lasted for more than 10 h. The generated Electronic Quantity of
MFC4 was 7583.84C m2.
The results suggested that the four MFCs displayed different
electricity generation proles. The current density slightly uctuated during the operation of MFC1 and MFC3. However, the current
density of MFC1 in continuous mode was relatively low and stable;
while that of MFC3 operated in the same mode was high and stable.
By contrast, MFC2 and MFC4 generated intermittently high current
pulses in the beginning, while the current density was instable
subsequently. The current density of MFC2 in batch mode was
relatively low and unstable, while that of MFC3 operated in the
same mode was relatively high but unstable.

3.3. Microbial community in the MFCs

The microbial proles of activated sludge from four MFCs were
ngerprinted using DGGE (Fig. 2). The afliation of nucleotide sequences determined by BLAST analysis was shown in Table 2, and a
phylogenetic tree was constructed based on the 16S rRNA gene
sequences (Fig. 3). The microbial diversity was not rich in the
inoculum sludge. Most of the microbial species (12 of 13 bands)
belonged to the class Betaproteobacterium, followed by Gammaproteobacterium (1 of 13 bands). Majority of the microorganisms
came from Betaproteobacterium belonging to the genus Thiobacillus
(6 of 12 bands) and Thauera (3 of 12 bands).
Similar results were obtained by screening the sequences from
the MFC1-MFC4. In MFC1, most of the microbial species (14 of 15
bands) belonged to the class Betaproteobacterium, followed by
Bacilli (1 of 15 bands). Majority of the microorganisms were related
to Betaproteobacterium, belonging to the genus Thiobacillus (9 of 14

bands) and Thauera (2 of 14 bands). In MFC2, most of the microbial

species (12 of 13 bands) were related to the class Betaproteobacterium, followed by Gammaproteobacterium (1 of 13 bands). Mainly
the microorganisms came from Betaproteobacterium in the genus
Thiobacillus (8 of 12 bands). In MFC3, a good number of the microbial species (14 of 15 bands) came from the class Betaproteobacterium, followed by Bacilli (1 of 15 bands). The majority of the
microorganisms were related to Betaproteobacterium in the genus
Thiobacillus (11 of 14 bands). In MFC4, most of the microbial species
(12 of 13 bands) came from the class Betaproteobacterium, followed
by Bacilli (1 of 13 bands). Most of the microorganisms came from
Betaproteobacterium belonging to the genus Thiobacillus (8 of 12
bands) and Thauera (2 of 12 bands).
In general, most of the microorganisms belonged to the genera
Thiobacillus and Thauera in the tested sludge. The results were
similar to the reported literature. Zhang et al. [18] used doublechamber MFCs to remove sulde and organics simultaneously.
Community analysis showed that Thiobacillus sp. was responsible
for electricity production and sulde oxidation in anodic biolms of
the MFCs. Kondaveeti et al. studied the bacterial communities in a
bioelectrochemical denitrication system [19]. Thauera a nitrate
utilizing bacterium was prevalent in the system. Thiobacillus species is capable of oxidizing a variety of sulfur compounds, namely
sulde, sulfur, thiosulfate and sulte, which are obligate chemolithotroph organisms [29]. Thauera species are known as nitratereducing microorganisms [19]. Members of this genus have been
suggested to oxidize sulde under anoxic conditions, and some of
these microorganisms are putatively involved in nitrate and sulfur
transformations have also been detected in suldogenic wastewater biolms. It is plausible that they were involved in the consumption of sulde and nitrate as observed in the tests [30].
4. Discussion
All MFCs showed good capacity of simultaneous sulde and
nitrate removal. Hence, the effect of substrate removal on the
bacterial community could be ignored. The differences in the microbial communities of four MFCs would be attributed to the inuence of the electricity through the MFCs. It seemed that the
microorganisms in the anodic chambers of all MFCs were exposed
to different types of current pressure. The current through inoculum was thought to be 0 mA m2; while the current through MFC1
was low (29 mA m2) and stable and the ratio of the highest current
to the lowest was 1.20. The current through MFC2 was low
(28 mA m2) and unstable; whereby the ratio of the highest current
to the lowest could reach 16.29. The current through MFC3 was
high (153 mA m2) and stable; the ratio of the highest current to
the lowest was only 1.03. The current through MFC4 was relatively
high (78 mA m2) and unstable, where the ratio of the highest
current to the lowest was 2.30.
The similarity of DGGE banding patterns for bacterial communities in inoculum sludge and suspension of four MFCs was quantied using the Dice index of similarity was shown in Table S1. The
Dice index of similarity of the inoculum sludge was 1.0 while values
for the MFC1, MFC2, MFC3 and MFC4 were 0.71, 0.77, 0.64 and 0.69,

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

Fig. 2. DGGE proles of bacterial 16S rRNA gene fragments (MFC1 continuous mode
with graphite rod, MFC2 batch mode with graphite rod, MFC3 continuous mode with
graphite felt and MFC4 batch mode with graphite felt.).

31

respectively indicating moderately similar bacterial communities.

The other indices like, Richness, Shannon Index, Simpson index
and Equitability index were also used to calculate the diversity of
bacterial communities, which were shown in Table 3. In order to
evaluate the effect of electricity on the microbial diversity,
spearman correlation analyses were applied to explore the correlation between bacterial diversity indices and the electricity production. In the current study, steady current density and Electronic
Quantity were used to demonstrate the electricity passage through
the MFCs as shown in Table 4.
If the electricity through the inoculum sludge was dened as 0,
it was suggested that the relation between Richness (or Simpson
index) and electricity generation was negative; while signicant
negative correlation existed between Richness and steady current
diversity (or Electronic Quantity). Spearman correlation analyses
showed that Shannon index and Equitability index were positively
related to steady current diversity and electronic quantity, while
the two indices did not show signicant correlation with the
electricity. Probably the most interesting observation was a strong
link between richness of the community in anode chamber and the
electricity generated; while there were not strong links between
the other indices and the electricity. The results were slightly
different from the other studies. Stratford et al. [16] revealed that
suspension diversities were not signicantly affected by the values
of current through MFCs. They examined the relationship between
the diversity of microbial consortia and their electrogenic potential
in the anodes of MFCs using different diversity measures as predictors. The bivariate correlation showed that there was lack of a
signicant correlation between suspension diversities (Richness,
Shannon index and Simpson index) and power output.
PCA was also used to investigate the effect of electricity on the
microbial community diversity in the anode chamber (Fig. 4). PCA
is an ordination technique widely used to provide a visual determination of the similarities among samples in community data. The
relationship amongst samples in a plot representing PCA 1 and PCA
2 showed four well dened groups; MFC2 and MFC4 formed one
group whereas inoculum, MFC1 and MFC3 constituted the other
distinct groups. In this analysis, PCA1 and PCA2 explained 38.04%
and 27.06% of the total variations, respectively.
PCA analysis revealed that the two MFCs suffered from current

Table 2
Phylogenetic afliation of the 16S rRNA gene sequences from DGGE bands.
Band

Accession number

Nearest species (Accession number)

Similarity (%)

Bacterial class

Relative abundance (%)

Inoculum

MFC1

MFC2

MFC2

MFC4

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

KM925019
KM925020
KM925021
KM925022
KM925023
KM925024
KM925025
KM925026
KM925027
KM925028
KM925029
KM925030
KM925031
KM925032
KM925033
KM925034
KM925035
KM925036
KM925037
KM925038
KM925039

Thauera aromatica (NR 026153.1)

Burkholderia pseudomultivorans (NR 117661.1)
Thiobacillus thiophilus NR 044555.1
Vulcanibacillus modesticaldus (NR 042421.1)
Pseudoxanthomonas mexicana (NR 113973.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117864.1)
Thauera aromatica (NR 026153.1)
Diaphorobacter nitroreducens (NR 024782.1)
Thauera terpenica (NR 025284.1)
Thiobacillus aquaesulis (NR 044793.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thioparus (NR 117817.1)
Thiobacillus aquaesulis (NR 044793.1)
Dechloromonas agitata (NR 024884.1)
Thiobacillus thioparus (NR 117864.1)
Thiobacillus thiophilus (NR 044555.1)

99
91
99
97
97
99
98
99
99
99
99
98
95
95
99
98
99
96
96
99
99

Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Bacilli
Gammaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria
Betaproteobacteria

5.69
9.99
7.60
ND
6.08
ND
ND
4.23
ND
ND
4.06
4.26
6.34
5.76
ND
ND
6.98
7.96
10.84
20.20
ND

3.73
4.38
4.77
5.58
ND
6.38
7.35
6.98
6.34
ND
5.43
4.99
ND
6.89
6.45
ND
7.12
ND
15.47
8.15
ND

3.98
7.25
7.64
ND
5.42
ND
ND
4.15
ND
4.36
ND
5.32
ND
4.96
ND
6.76
8.98
ND
12.49
24.55
4.15

ND
7.44
7.80
5.44
ND
ND
4.36
4.28
4.57
6.54
ND
3.79
ND
5.48
5.64
ND
6.22
6.70
9.10
10.96
11.69

7.26
14.09
3.77
6.93
ND
ND
ND
5.04
ND
ND
ND
ND
4.18
11.51
9.59
9.77
8.32
ND
10.72
5.17
3.65

32

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

Fig. 3. The phylogenetic tree based on the 16S rRNA gene sequences from DGGE bands (The numbers at the nodes indicate the levels of bootstrap support based on neighbor-joining
analysis of 1000 re-sampled datasets).

shock which was evident by similar suspension communities,

while the other three were distinct from each other. It can be
inferred that current shock had a greater impact on communities
than current values. In other words, the effect of current values was
little on suspension communities. The other studies proved that the
suspension community was similar, when the value and behavior
of the current was similar. El-Chakhtoura et al. tested the start-up
of two sets air-cathode MFCs inoculated with wastewater sludge
or cattle manure [31]. When the manure MFCs exhibited almost the

Table 3
Diversity indexes of microbial communities.
Diversity index

Inoculum

MFC1

MFC2

MFC3

MFC4

Richness
Shannon index
Simpson index
Equitability index

9
2.45
0.10
0.95

8
2.64
0.08
0.98

8
2.38
0.11
0.93

7
2.65
0.07
0.98

8
2.48
0.09
0.97

same voltage behavior as the wastewater MFCs, bacterial communities had most likely evolved to the same bacterial diversity. Quan
et al. [13] compared the difference in microbial community and
power generation capacity of air-cathode MFCs enriched under
anode facultative conditions. Although the aerobic enriched MFC
produced slightly more electricity using acetate, glucose and
ethanol compared to the anaerobic MFC; the two MFCs showed a
very similar microbial community (a similarity of 97%) for suspended culture in anode chamber.
In the literature reported, many researchers have studied the
microbial community of anodic biolm in MFCs, and they observed
that several factors could have inuences on microbial community
structure including but not limited to inoculum source [32], cathode types [33], substrate used [15] and Anodic potential [12].
However, Sun et al. tested four materials as packed anodic materials
to seek a potentially practical material for MFCs, and explored the
microbial community of anodic biolm and its correlation with the
electricity generation performance [34]. They found that electrode

Table 4
Spearman correlation analyses.
Diversity index

Richness
Shannon index
Simpson index
Equitability index

Spearman correlation
Maximum current density

Signicance

Steady current density

Signicance

Electronic quantity

Signicance

0.447
0.300
0.300
0.359

0.450
0.624
0.624
0.553

0.894*
0.800
0.800
0.718

0.041
0.104
0.104
0.172

0.894*
0.500
0.500
0.359

0.041
0.391
0.391
0.553

*p < 0.05 as determined by SPSS version 18.0 program (SPSS, Chicago, Illinois, USA).

J. Cai et al. / Journal of Power Sources 281 (2015) 27e33

33

Foundation of China (No. 51278457) for nancial support of this

study.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jpowsour.2015.01.165.

References

Fig. 4. PCA analysis of inoculum, MFC1, MFC2, MFC3 and MFC4.

materials had important effects on the microbial diversity, and

microbial species were also related to the power generation capability. But there was no apparent relationship between the microbial diversity and power generation capability was observed.
Maybe it reects that the electricity does not have signicant inuence on the microbial community from one perspective.
Moreover, in the experiment, the inoculum sludge was enriched
in UASB reactor as long as three years, which caused microbial
population simple. Most of the microorganisms (about 70%)
belonged to the family Thiobacillus and Thauera, and the two families were closely related to simultaneous sulde and nitrate
removal. The typical exoelectrogens were not found in the anode
chamber of MFCs, such as Shewanella putrefaciens, Shewanella
oneidensis, Desulfuromonas acetoxidans, Geobacter sulfurreducens,
Rhodoferax ferrireducens, Pseudomonas aeruginosa, Ochrobactrum
anthropiwere [35]. Generally speaking, the species on the surface of
electrode can participate in electron transfer, while the species in
suspension may divert available energy from substrate towards
other metabolic activities. It might be able to explain why suspension communities were not sensitive to current values, because
the microorganisms in suspended sludge of MFC were not directly
related to generate electricity.
5. Conclusion
The MFCs treating sulde and nitrate simultaneously under four
different conditions displayed different behavior in electricity
generation. All MFCs showed good capacity for substrate removal
where nitrogen and sulfate were the main end products.16S rRNA
PCR-DGGE analysis determined the bacterial composition of
inocula and respective suspension within four MFCs. Based on the
community diversity indices and Spearman correlation analyses, a
signicant relation was observed between community Richness
and the electricity generated; while weak relation was noted
among the other indices (Shannon index, Simpson index and
Equitability index) and power output. The results of PCA inferred
that based on similar microbial communities, two MFCs suffered
from current shock, while the remaining communities were
distinct from each other.
Acknowledgment
The authors wish to thank the National Natural Science

[1] B.E. Logan, B. Hamelers, R.A. Rozendal, U. Schrorder, J. Keller, S. Freguia,

P. Aelterman, W. Verstraete, K. Rabaey, Environ. Sci. Technol. 40 (2006)
5181e5192.
[2] W. Habermann, E.H. Pommer, Appl. Microbiol. Biotechnol. 35 (1991)
128e133.
[3] D. Pant, G. Van Bogaert, L. Diels, K. Vanbroekhoven, Bioresour. Technol. 101
(2010) 1533e1543.
[4] Q. Mahmood, P. Zheng, J. Cai, Y. Hayat, M.F. Hassan, D.L. Wu, B.L. Hu,
J. Zhejiang University-Science A 8 (2007) 1126e1140.
[5] R. Jin, G.F. Yang, Q.Q. Zhang, C. Ma, J.J. Yu, B.S. Xing, Water Res. 47 (2013)
1459e1469.
[6] D.G. Cirne, F.P. van der Zee, M. Fernandez-Polanco, F. Fernandez-Polanco, Rev.
Environ. Sci. Bio/Technol. 7 (2008) 93e105.
[7] J. Garcia-de-Lomas, A. Corzo, M.C. Portillo, J.M. Gonzalez, J.A. Andrades, C. SaizJimenez, E. Garcia-Robledo, Water Res. 41 (2007) 3121e3131.
[8] C. Lee, K. Ho, D. Lee, A. Su, J. Chang, Int. J. Hydrogen Energy 37 (2012)
15827e15832.
[9] J. Cai, P. Zheng, Bioresour. Technol. 128 (2013) 760e764.
[10] J. Cai, P. Zheng, J. Zhang, Z. Xie, W. Li, P. Sun, Bioresour. Technol. 129 (2013)
224e228.
[11] W. Zhi, Z. Ge, Z. He, H. Zhang, Bioresour. Technol. 171 (2014) 461e468.
[12] T.-Y. Chou, C.G. Whiteley, D.-J. Lee, Int. J. Hydrogen Energy 39 (2014)
19225e19231.
[13] X.-c. Quan, Y.-p. Quan, K. Tao, X.-m. Jiang, Bioresour. Technol. 128 (2013)
259e265.
[14] B. Xiao, Y. Han, X. Liu, J. Liu, Int. J. Hydrogen Energy 39 (2014) 16419e16425.
[15] Y. Zhang, B. Min, L. Huang, I. Angelidaki, Bioresour. Technol. 102 (2011)
1166e1173.
[16] J.P. Stratford, N.J. Beecroft, R.C.T. Slade, A. Grning, C. Avignone-Rossa, Bioresour. Technol. 156 (2014) 84e91.
[17] M. Sun, Z.-H. Tong, G.-P. Sheng, Y.-Z. Chen, F. Zhang, Z.-X. Mu, H.-L. Wang,
R.J. Zeng, X.-W. Liu, H.-Q. Yu, L. Wei, F. Ma, Biosens. Bioelectron. 26 (2010)
470e476.
[18] B. Zhang, J. Zhang, Y. Liu, C. Hao, C. Tian, C. Feng, Z. Lei, W. Huang, Z. Zhang, Int.
J. Hydrogen Energy 38 (2013) 14348e14355.
[19] S. Kondaveeti, S.-H. Lee, H.-D. Park, B. Min, Water Res. 51 (2014) 25e36.
[20] Q. Mahmood, P. Zheng, J. Cai, D.L. Wu, B.L. Hu, J.Y. Li, J. Hazard Mater. 147
(2007) 249e256.
[21] APHA, AWWA, WPCF, Standard Methods for the Examination of Water and
Wastewater, twentieth ed., American Public Health Association, Washington,
DC, 1998.
[22] Y. Yu, C. Lee, J. Kim, S. Hwang, Biotechnol. Bioeng. 89 (2005) 670e679.
[23] Y.-j. Xing, J.-y. Ji, P. Zheng, J.-q. Zhang, A. Ghulam, Appl. Microbiol. Biot. 98
(2014) 1357e1366.
[24] B. Hu, Z. He, S. Geng, C. Cai, L. Lou, P. Zheng, X. Xu, Appl. Microbiol. Biot. 98
(2014) 7983e7991.
[25] T.M. LaPara, C.H. Nakatsu, L.M. Pantea, J.E. Alleman, Water Res. 36 (2002)
638e646.
[26] C.J. Keylock, Oikos 109 (2005) 203e207.
[27] H.C. Ramrez-Saad, A. Sessitsch, A.D.L. Akkermans, Microbiol. Res. 158 (2003)
47e54.
[28] J. Cai, P. Zheng, M. Qaisar, Y. Xing, J. Ind. Microbiol. Biotechnol. 41 (2014)
795e802.
[29] M. Mirzaei, G. Amoabediny, F. Yazdian, M. Sheikhpour, E. Ebrahimi,
B.E.H. Zadeh, Process Biochem. 49 (2014) 380e385.
[30] O. Auguet, M. Pijuan, H. Guasch-Balcells, C.M. Borrego, O. Gutierrez, Water
Res.
[31] J. El-Chakhtoura, M. El-Fadel, H.A. Rao, D. Li, S. Ghanimeh, P.E. Saikaly, Biomass
Bioenergy 67 (2014) 24e31.
[32] Y. Zhang, B. Min, L. Huang, I. Angelidaki, Appl. Environ. Microbiol. 75 (2009)
3389e3395.
[33] G. Zhang, K. Wang, Q. Zhao, Y. Jiao, D.-J. Lee, Bioresour. Technol. 118 (2012)
249e256.
[34] Y. Sun, J. Wei, P. Liang, X. Huang, Bioresour. Technol. 102 (2011)
10886e10891.
[35] Q. Xu, J. Sun, Y.-y. Hu, J. Chen, W.-j. Li, Bioresour. Technol. 142 (2013)
101e108.