Bioresource Technology xxx (2011) xxx–xxx
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
Full-loop operation and cathodic acidification of a microbial fuel cell operated
on domestic wastewater
Olivier Lefebvre a, Yujia Shen a, Zi Tan a, Arnaud Uzabiaga a, In Seop Chang b, How Yong Ng a,⇑
a
b
Centre for Water Research, Department of Civil and Environmental Engineering, National University of Singapore, 1 Engineering Dr. 2, Singapore 117576, Singapore
Department of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan Gwagi-ro, Buk-gu, Gwangju 500-712, Republic of Korea
a r t i c l e
i n f o
Article history:
Received 3 November 2010
Received in revised form 29 January 2011
Accepted 23 February 2011
Available online xxxx
Keywords:
Microbial fuel cell
Loop
Overflow
Selemion
Wastewater treatment
a b s t r a c t
The present study emphasizes the importance of overcoming proton limitation in a microbial fuel cell
operated on domestic wastewater. When the anode-treated effluent was allowed to trickle into the
cathodic compartment (full-loop operation), high COD and suspended solids removal efficiencies over
75% and 84%, respectively, were achieved while ensuring substantial and sustainable power generation.
Lower removal efficiencies resulted in decreased cell electromotive force caused by excess substrate
crossover. By decreasing the pH in the cathodic compartment to values below 2, we were able to further
increase the maximum power generation by 180% in batch mode and 380% in continuous mode as compared to a negative control (tap water of pH 7.6). Under the optimized conditions, the internal resistance
and electromotive force were 11 X and 0.6 V, respectively, which correspond to the state of the art.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Providing the world’s population with adequate sanitation is
among the major development challenges of this century and
recovering energy from wastewater seems the only way to allow
sanitation programs to maintain their development without drastically increasing energy consumption. Among the different types
of wastewater, management of domestic wastewater is particularly crucial in fast growing and land-limited Singapore. In the recent years, there has been growing interest in microbial fuel cell
(MFC), a technology derived from chemical fuel cell that allows
simultaneous wastewater treatment and energy recovery directly
in the form of electricity. However, the technology is still in its infancy due to scale-up limitations and large applications in wastewater treatment plants are hindered by the inner limitations of
the technology. One of the major limitations of MFCs nowadays
is related to stacking issues. Even though parallel stacking can be
efficient – at least on a limited number of fuel cells (Aelterman
et al., 2006) – series connections of MFCs rapidly lead to energy
losses and reduced stacking efficiencies (Wang and Han, 2009).
Furthermore, series stacking has been shown to result in cell polar-
Abbreviations: HRT, hydraulic retention time; HCl, hydrochloric acid; MEA,
membrane electrode assembly; MFC, microbial fuel cell; OCV, open circuit voltage;
PTFE, polytetrafluoroethylene; SS, suspended solids; VSS, volatile suspended solids.
⇑ Corresponding author. Tel.: +65 6516 4777; fax: +65 6774 4202.
E-mail address: esenghy@nus.edu.sg (H.Y. Ng).
ity reversal (Aelterman et al., 2006). This raises the question of
whether the power should be maximized on a volumetric basis –
as is the consensus nowadays – or on a ‘‘per cell’’ basis, in order
to reduce the number of cells to be connected together.
At first glance, the main difference between conventional chemical fuel cells and MFCs is that MFC anodes are ‘‘alive’’ and rely on
microbial metabolism. However, another major difference between the two technologies lies in the extremely different environmental conditions applied to both systems. In an MFC, the
temperature is ambient and the pH circum-neutral to allow bacterial growth on the anode. On the other hand, in a conventional
hydrogen fuel cell the temperature is above 80 °C and the cathode
is kept under a pressure of 2 bars of oxygen and hydrogen (Barbir,
2005). This makes it possible in hydrogen fuel cells to achieve electromotive force higher than in MFCs. On top of it, the pH in the cation exchange membrane of a hydrogen fuel cell is around 3, which
implies that there are much more protons available to contribute
to the charge transfer between the electrodes, and this transfer is
further facilitated by high temperatures that dramatically increase
proton conductivity (Barbir, 2005). Kinetics are also much faster in
chemical fuel cells, which helps to maintain their internal resistance at low levels, in the order of a few mX.
Because of the disadvantageous environmental conditions applied to MFC systems and described above, the anode is seldom
the limiting factor in a well constructed modern MFC design and
the highest anodic power density achieved so far was obtained in
an MFC where the cathode was 14 times larger than the anode
0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2011.02.098
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
2
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
(Fan et al., 2008). At the cathode, the protons and electrons of an
MFC generally combine with oxygen to form water and in this
sense, proton availability and mobility are of prime importance
and most of the time lacking in MFCs, where electro neutrality is
achieved by transport of other more abundant cation species such
2+
as Na+, K+, NHþ
and Mg2+(Rozendal et al., 2006). Bringing pro4 , Ca
tons directly to the cathode could be a way to improve the electrical performance of the system. One way is by operating the MFC in
a full-loop mode where the anode-treated effluent is allowed to
enter the cathode compartment and, recently, some publications
have presented the advantages of such method (Clauwaert et al.,
2009; Freguia et al., 2008a; Li et al., 2009). From the point of view
of electrochemistry, this helps counterbalancing pH variations in
two-chamber MFCs, in which otherwise cathode alkalinization
and anode acidification with time are observed (Gil et al., 2003).
Furthermore, from the point of view of wastewater treatment engineering, the cathodic compartment occupies a large footprint that
is not directly used for wastewater treatment in most cases. With
domestic wastewater, MFC is known to produce an effluent of
insufficient quality for discharge after passing through the anodic
compartment (Cheng et al., 2006; Lefebvre et al., 2008). The situation even worsens if the MFC is optimized for electricity generation, because the power output increases when the HRT
decreases but the trend is opposite for COD removal (Ahn and
Logan, 2010; Liu et al., 2004). This means that effluent polishing
will be required in an MFC-based wastewater treatment plant. If
the anolyte is introduced into the cathodic compartment, the latter
has the potential to provide aerobic post-treatment for the anodetreated effluent, while protons are transported directly by the anolyte to the cathode of the MFC. Another way to overcome proton
limitation at the cathode is by bringing them intentionally into
the cathode compartment, which is the main focus of the present
study.
In this paper, we disclose a prototype MFC that is suitable for
continuous treatment of domestic wastewater, according to the research objectives of our laboratory. This implies that the MFC has
to be of a reasonable size (a few litres). In a larger reactor, power
generation is further maximized per cell used and not on a volumetric basis, which – as explained above – might prove useful until
the stacking technology matures. In order to overcome proton limitation at the cathode, two strategies were employed – introduction of the anolyte into the cathodic compartment of the MFC,
and acidification of the cathode using HCl. For that purpose, the
MFC made use of a membrane electrode assembly (MEA) incorporating a proton selective membrane, in order to prevent accidental
acidification of the anodic compartment.
2. Methods
2.1. MEA-MFC design
The reactor vessel was a vertical cylinder (length = 90 cm, diameter = 7 cm) made of transparent polyacrylic plastic (Thermoplastics, Singapore) and the MEA was incorporated in the middle. The
MEA (length = 90 cm, diameter = 3 cm)was wrapped around a
stainless steel grid acting as the cathodic current collector and a
stainless steel mesh was tightened over the anode acting as the
anodic current collector. The anodic compartment had a volume
of 2.9 L and the cathodic compartment capacity was of 0.6 L. The
experimental design is shown in Fig. 1.
The MEA consisted of an anode and a cathode wrapped on
opposite sides of an ion exchange membrane. A proton-selective
Selemion ion exchange membrane (model HSF, Asahi, Japan) was
selected over the comparable Nafion membrane due to its more
competitive price. Selemion HSF membrane, originally designed
for electrodialysis, is characterized by a thickness of 150 lm, a
burst strength of 0.2 MPa and a resistivity of 0.3 X cm2 in
0.5 mol LÀ1 HCl or H2SO4 (manufacturer data). Two layers of reverse osmosis spacer were incorporated, between the anode and
the Selemion membrane, and between the cathode and the membrane. The purpose of such spacer was to prevent partial shortcircuit between the anode and the cathode in the MEA. In the
absence of spacer indeed, the resistance between the anode and
the cathode was only of around 200 X after tightening the MEA,
but with the use of spacer this value could be increased to over
1 MX, indicating that the partial short-circuit was largely
overcome. This emphasizes that, even though the anode and the
cathode should be maintained as close as possible to one another
in an MFC system (Cheng et al., 2006), there is a limitation in
MEA designs due to the risk of short-circuits. Both the anode and
the cathode were made of carbon cloth (designation B, E-Tek,
USA) and the cathode was further coated with Pt (0.5 mg cmÀ2)
on one side. A detail of the MEA is shown in Fig. 1c.
2.2. Experimental conditions
The fuel (domestic wastewater) was circulated continuously in
an upflow mode into the anodic compartment at a flow rate of
20 mL minÀ1 (HRT = 2.4 h).Unless specified otherwise, aeration
was provided by actively bubbling air into the cathodic compartment using an air compressor and regulating the airflow with a
valve at 5 L minÀ1. In early experiments, the anode-treated effluent
collected from the top of the reactor was introduced into the top of
the cathodic compartment where it was allowed to trickle along
the cathodic wall (see Fig. 1a). Domestic wastewater was used
simultaneously as the fuel and the inoculum for the reactor, which
was collected from the primary decantation basin of the Ulu Pandan Reclamation Plant in Singapore. The pH, COD, suspended solids
(SS) and volatile suspended solids (VSS) content ranged from 7.3 to
7.9240 to 459 mg LÀ1, 152 to 288mg LÀ1 and 138 to 426 mg LÀ1,
respectively. The VSS to SS ratio was ranged between 0.6 and 0.9,
indicating good biodegradability. The reactor was set-up at ambient temperature (25 °C) and wrapped with aluminum foil to prevent algae growth.
Cathodic acidification was performed in batch and in continuous modes. Batch acidification used 500 mL of hydrochloric acid
(HCl) solutions diluted in tap water at different pH, ranging from
6 to 1, and pumped into the cathodic compartment, which had
been previously closed (Fig. 1b). Air was allowed to bubble into
the acidic solution at a flow rate of 5 L minÀ1. The voltage was recorded every 10 s across an external resistance of 40 O and, finally,
the cathodic compartment was reopened to collect the catholyte
and measure its pH. Tap water with a pH of 7.6 was used as a negative control. Continuous acidification used HCl solution diluted in
tap water at different pH, ranging from 1.2 to 6, and pumped at a
flowrate of 250 mL minÀ1 into the cathodic compartment. Airflow
rate was similarly fixed at 5 L minÀ1. The experiment began under
the open circuit voltage (OCV) conditions. After OCV stabilization,
the external load was progressively decreased in order to record
polarization curves. Samples of the cathodic outlet were collected
regularly during the experimental period. Tap water (pH 7.6) and
phosphate buffer solutions (pH 7) having ionic strengths of 10À2
and 10À4 M, respectively, were used as negative controls.
2.3. Analytical methods and calculations
COD, SS, VSS and pH were analyzed according to the standard
methods (APHA, 2005). The cell voltage was monitored with a multimeter (M3500A, Array Electronic, Taiwan) connected to a computer by a data acquisition system (PC1604, TTi, RS, Singapore).
Starting from the OCV, polarization curves were obtained by
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
3
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
(a)
(b)
air
HCl in
effluent
out
anode
compartment
spacer
cathode
compartment
(c)
cathode
current
collector
Selemion
effluent
in
effluent
in
effluent out
spacer
lumen
anode
current
collector
HCl out
Fig. 1. Experimental design (a) operation in full-loop where the effluent from the anodic compartment is introduced into the cathodic compartment, (b) operation with
acidification of the cathodic compartment by HCl, (c) detail of the membrane electrode assembly.
decreasing the applied external resistance and recording the pseudo steady-state voltage. The current was then determined using
the Ohm’s law and the Coulombic efficiency was calculated based
on the current and COD removal following Logan et al. (2006). The
cell electromotive force (Eemf, V) and internal resistance (Rint, O)
were determined using a linear regression (least squares method)
on the linear part of the polarization curve that corresponds to
the Ohmic zone. The electromotive force was estimated as the
intercept of the regression with the Y-axis whereas the internal
resistance was the opposite of its slope. The maximum power
(Pmax, W) supplied by the MFC was calculated according to Eq. 1
Pmax ¼
ðEemf Þ2
4Rint
ð1Þ
Data analyzes were assessed statistically by Student’s t-tests for unpaired samples. Two sets of data were considered significantly different when the P value was inferior than 0.05.
tion 3.1. Stage 4 (75 d) corresponds to the MFC operation in a single-chamber mode with an open-air cathode (i.e., full-loop
operation was ceased). Cathodic acidification experiments were
carried out during stage 4 between days 95 and 120. Details of
operation during stage 4 are detailed in Section 3.2.
3.1. Operation in a full-loop mode (stages 1–3)
The full-loop experiments were carried over a period of 65 d
and the experimental protocol is shown in Fig. 1a. Three distinct
stages could be identified. Stage 1 (7 d) corresponded to the
start-up of the reactor followed by stage 2 (20 d) characterized
by a deterioration of the reactor performance. Finally, in Stage 3
(38 d), the anodic compartment was filled with graphite granules
resulting in improved performance. The detail of the three stages
is given below and the electrochemical characteristics of the reactor and of the treatment performance at each stage are summarized in Figs. 2 and 3.
2.4. Performance assessment as compared to the literature
The performance of our MFC set-up was compared to the literature available using the following criteria: OCV, Eemf, Rint, Pmax,
and the volumetric power (Pvmax). Eemf and Rint were determined
using the polarization curves provided in the selected articles
and Pmax, was calculated using Eq. 1. Furthermore, in an effort of
standardization, Pvmax was calculated for each study taking into account the total working volume of the MFCs, which corresponds to
the sum of the volume of the anodic and cathodic compartments.
In our opinion, this shall reflect more accurately the impact of
the footprint of the treatment system.
3. Results and discussion
The experiments were conducted over a period of 140 d. Four
distinct stages could be identified. Stages 1 to 3 (65 d) correspond
to the MFC operation in a full-loop mode and are described in Sec-
3.1.1. Stage 1
The MEA-MFC was operated initially on domestic wastewater in
a full-loop mode in which the effluent flowed upward into the anodic compartment then trickled into the inner cathodic compartment (see Fig. 1a). The power rose constantly over the first week
of operation and attained 1.74 ± 0.16 mW (Fig. 2). The OCV at that
time was higher than 0.7 V, which indicates proper functioning of
the system as an electricity generation device. The cell electromotive force averaged 0.59 ± 0.08 V and the internal resistance was
estimated at 52 ± 19 X. After passing through the anodic compartment, only 39 ± 11% of the COD and 57 ± 10% of the SS were removed (Fig. 3). However, the treated effluent collected from the
outlet of the cathodic compartment appeared much clearer in color
and the COD removal averaged 71 ± 5%. In addition, 78 ± 2% of the
SS were removed in the process. This confirms the potential of
using the aerobic cathode compartment for effluent polishing in
the MFC technology. The Coulombic efficiency was estimated to
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
4
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
Fig. 2. Evolution of the electromotive force (Eemf), maximum power (Pmax), internal resistance (Rint) and Coulombic efficiency (CE) of the microbial fuel cell during the four
stages of the experimental period. Stage 1 corresponds to the start-up of the reactor. Stage 2 is a period during which the reactor performance deteriorated. Stage 3
corresponds to the operation of the anodic compartment with graphite granules. Stage 4 corresponds to the operation in a single-chamber mode with an open air–cathode.
Cathodic acidification experiments were carried out during stage 4.
average 0.2 ± 0.1%, which reflects that most of the COD was
removed by ways that did not contribute to electricity generation.
Mostly, there was a tendency to observe sludge accumulation at
the bottom of the anodic compartment, where fermentation was
probably taking place.
3.1.2. Stage 2
After 14 d, the power generation was found to be much lower
and the OCV was reduced to below 0.4 V. The electromotive force
dropped drastically to 0.13 ± 0.04 V while the internal resistance
was affected to a lesser extent, averaging 70 ± 13 X (Fig. 2). Consequently, the maximum power output (0.06 ± 0.04 mW) was found
to be significantly lower than that during stage 1 (P = 0.005; Student’s t-test). This drop in performance was accompanied by a
poorer quality of the treated effluent that looked darker in color
and the COD and SS removal efficiencies dropped to 54 ± 11 and
32 ± 30% (Fig. 3), respectively, resulting in unchanged Coulombic
efficiency of 0.2 ± 0.1%. A probable reason for the observed drop
of electromotive force could be attributed to excess substrate
crossover from the anode to the cathode, resulting in a cathodic
potential mixed between that of O2 and that of the abovementioned substrate, as explained by Harnisch et al. (2009). Furthermore, another explanation could be related to aerobic bacteria
growing on the cathode due to excess substrate and further limiting oxygen access for the cathodic reaction. Similar problem was
also noted by Freguia et al. (2008a) who used acetate as a substrate. When they increased the loading rate in their MFC, overloading of the cathodic compartment and a drop in power
occurred. However, for practical reasons, it was not possible to access the MFC cathode in the course of our experiment to evaluate
biofilm growth.
3.1.3. Stage 3
In order to improve the quality of the anode-treated effluent,
the anodic compartment was filled with graphite granules having
an estimated projected surface of between 500 and 3000 m2 mÀ3
(Carbone Lorraine, Belgium). In addition, the cathode compartment
was washed with plenty of water. As a result, the COD and SS removal efficiencies after passing through the anodic compartment
increased to 75 ± 21% and 84 ± 19%, respectively (Fig. 3). The high
variability of COD and SS removal efficiencies could be attributed
to the variability of the influent itself depending on the weather
conditions (i.e., rainfall). For instance, during stage 3 the influent
SS dropped and this was accompanied by a drop in the SS removal
efficiency that followed the same pattern. Introducing the anolyte
into the cathode further improved the COD removal to 92 ± 4%,
whereas the SS removal remained unchanged. In terms of electrochemical performance, the fuel cell electromotive force rose to
0.62 ± 0.06 V while the internal resistance remained unaffected at
70 ± 9 O (Fig. 2). These results confirm the hypothesis that the drop
in power observed in stage 2 could be attributed to substrate
crossover as explained by Harnisch et al. (2009), since only the
electromotive force seemed to be affected by the quality of
the anode-treated effluent. The maximum power was then of
1.41 ± 0.21 mW (an increase by 2250% as compared to stage 2;
P = 0.00003 using Student’s t-test) and remained stable over the
next 37 d. The Coulombic efficiency was also considerably enhanced, averaging 11 ± 1% (an increase by 5400% from stage
2).These results confirm the benefits of letting the anolyte flow into
the cathode compartment to increase proton availability at the
cathode. However, this should be done with care to avoid overloading the cathode with excess substrate.
3.2. Effect of cathodic acidification (stage 4)
The cathodic acidification experiments were carried out without granules in the anode compartment. This phase corresponds
to stage 4 in Figs. 2 and 3. In the first place and before adding
HCl, full-loop operation and active aeration were stopped so the
cathode became an open-air cathode. Under steady state
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
5
(a)
(b)
Fig. 3. Evolution of the (a) COD and (b) suspended solids (SS) concentrations in the influent, anodic outlet, cathodic outlet and corresponding total removal efficiencies during
the four stages of the experimental period.
conditions, the maximum power averaged 2.29 ± 0.32 mW, an increase by 62% as compared to stage 3 (P = 0.0002; Student’s t-test).
The internal resistance was reduced to 48 ± 11 O and the electromotive force reached 0.66 ± 0.04 V. However, the COD and SS
removal efficiencies were reduced to 50 ± 9% and 61 ± 12%, respectively, as compared to stage 3 (Fig. 3), and the Coulombic efficiency
dropped to 0.9 ± 0.4%, due to the absence of granules. The
increased electrochemical performance of the MFC reactor in
open-air cathode conditions comes as no surprise, knowing that
this mode of operation results in increased oxygen availability at
the cathode and reduced internal resistance, as it has largely been
documented in the literature (Liu et al., 2004). After assessing the
performance without HCl, the actual cathodic acidification experiments were carried out between days 95 and 120 subsequently in
batch mode and in continuous manner.
3.2.1. Batch experiments
In the beginning, cathodic acidification was conducted over a
period of 14 d under different pH conditions in a batch mode by
pumping 500 mL of diluted HCl solution into the cathodic compartment. The power response is shown in Fig. 4, from which it
can be seen that the injection of diluted HCl solution resulted in
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
Power (mW)
4
pH 7.6
pH 6
pH 4
pH 3
pH 2
pH 1
3
(a) 8
5
4
3
2
1
0
0
10
20
Time (min)
Power (mW)
5
Power (mW)
6
30
2
6
4
2
1
0
0
100
200
Time (min)
300
400
Fig. 4. Power response after batch acidification of the cathode (Rext = 40 O). Insert:
details of the first 30 min.
an immediate power increase. However, the improvement caused
by the decreasing pH was not steady and the power rapidly decreased again after the protons were consumed. As a result the
catholyte had a pH around 8 at the end of the experimental time,
regardless of the initial pH applied. It should be noted that both
the maximum power and the time taken to stabilize increased
when the pH was decreased. Ultimately, the power went back to
the steady state conditions under neutral pH after a period of time
varying from 20 min to 8 h depending on the initial pH applied.
According to Fig. 4, the optimal power production was attained
when the cathodic pH was decreased to 2. Under these conditions,
a maximum power of 4.5 mW (Rext = 40 X) was achieved after
2.5 h, an increase by 180% as compared to a negative control (tap
water of pH 7.6). In a comparable test, also in batch mode but at
a cathodic pH of 1, Erable et al. (2009) found an increase of power
by 250% in batch mode as compared to neutral conditions, achieving 3.5 mW in their 700 mL MFC system consisting of two chambers with an open-air cathode. However, in our case when the
catholyte pH was further decreased to 1, the power production
reached a maximum of 3.6 mW after 2.3 min, followed by a rapid
decrease. The catholyte was then characterized by a green color
and an acidic pH of 2, which indicated that the protons had not
been consumed in their entirety by the reaction at the cathode.
The resistance to HCl of the stainless steel current collector used
at the cathode and that of the platinum coated carbon cloth cathode were tested by exposing samples at a pH of 1 and, within two
days, stainless steel was completely degraded resulting in a green
color solution, whereas the carbon cloth was unaffected. The green
compound was identified to be FeCl2 – a greenish tetrahydrate
responsible for the green colored solution – and it can be hypothesized that Fe and HCl combined to form FeCl2, a reaction that is
shown in Eq. 2.
Fe þ 2HCl ! FeCl2 þ H2
ð2Þ
Such redox reaction was detrimental to the cathodic reaction
and resulted in decreased power production at pH 1. It should be
noted here that in the study of Erable et al. (2009), this problem
was not encountered as they utilized brass as current collector.
3.2.2. Continuous experiments
Continuous acidification experiments were carried out over a
period of 10 d between day 110 and 120. Under continuous acidification, stable power could be generated. Power curves at different
catholyte pH confirmed the trend of increasing power generation
with decreasing pH (Fig. 5a). Continuous cathodic acidification at
pH = 1.2 allowed the generation of power supply 380% as high as
what could be obtained with tap water (pH 7.6). Under such conditions, the reactor generated a maximum power of 7.4 mW
0
0.01
(b) 0.8
Voltage (V)
0
0.02
Current ( A)
pH 7.6
pH 5
pH 3
pH 1.5
Buffer 10 M
0.6
0.03
0.04
pH 6
pH 4
pH 2
pH 1.2
Buffer 10 M
0.4
0.2
0
0
0.01
0.02
Current ( A)
0.03
0.04
Fig. 5. (a) Power curves under continuous acidification of the cathode at different
pH; (b) polarization curves under continuous acidification of the cathode at
different pH.
(Rext = 15 X), the maximum observed in this study. Two phosphate
buffer solutions with ionic concentrations of 10À2 and 10À4 M to
cover the entire range of our acidic solutions were used as negative
controls to demonstrate that the improved concentration is due to
increased proton access and not simply due to increased ionic
strength. It appears clearly from Fig. 5 that the effect of ionic
strength was limited as compared to that of pH.
The polarization curves provide a deeper insight into the impact
of acidification on the electrical performance of the MFC (Fig. 5b).
First, the internal resistance decreased with the pH from 58 O at pH
7.6 to 11 X at pH 1.2. The effect on the electromotive force was
more complex. Eemf first increased when pH was decreased and
reached an optimum value of 0.7 V at a pH of between 3 and 2.
The Nernstian effects, whereby the cathodic potential increases
with the proton concentration, can explain this. However, further
decrease in pH resulted in a decreasing electromotive force that
reached 0.58 V at a pH of 1.2 and this was related to the corrosion
reaction previously described that affected the potential of the
cathode when the pH became too acidic. However, in terms of
power generation, higher values were obtained at pH 1.2 than 2,
which shows that, overall, the impact of acid on the internal resistance was stronger than that on the electromotive force.
Finally, we analyzed the pH of the solution in the outlet of the
cathodic compartment and it appeared that the pH was slightly
higher than that of the feed to the cathodic compartment for feed
pH higher than 2. For a solution of pH 2 or less, the pH in the outlet
was similar to that of the inlet, indicating that the proton limitation might be overcome under these conditions in our MFC system.
At pH 2, the maximum power was achieved at a current of 15 mA
(Fig. 5a) and, since 1 mol of electron reacts with 1 mol of protons in
the cathodic reaction, the quantity of pH 2 solution to maintain the
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
7
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
current at 15 mA was estimated at 1.3 L dÀ1. That corresponds
roughly to as little as 2 mL of 37% HCl solution, as commonly available for sale. Considering the HRT of the reactor (2.4 h), this corresponds roughly to 0.07 mL of 37% HCl solution used per litre of
wastewater treated.
3.2.3. Reversibility of the cathodic acidification experiments
With the use of the Selemion membrane that allows protons to
be transferred only in one direction from the anode towards the
cathode, the pH at the anode was never affected by the drastic
change of pH in the cathode compartment. The MFC operation in
a reversible mode from cathodic acidification back to open-air
cathode operation was further assessed at the end of stage 4
(day 120–140). As can be seen from Fig. 2, the MFC immediately reverted to performance similar to what was observed before the
cathodic acidification experiments. The internal resistance and
electromotive force averaged 47 ± 8 O and 0.63 ± 0.05 V, respectively, resulting in a maximum power of 2.17 ± 0.01 mW, not
significantly different from the values obtained before the acidification experiments (P = 0.6; Student’s t-test). From Fig. 3, it can
be further seen that the COD and SS removal efficiencies were also
largely unaffected by the cathodic acidification experiments, with
mean values of 53 ± 7% and 62 ± 2%, respectively. As a result the
Coulombic efficiency also remained stable at 0.4 ± 0.2%. This
further shows that bacteria possibly present as a biofilm on the
cathode played little to no role in power generation as compared
to abiotic processes catalyzed by platinum. This is because any biofilm on the cathode would have been damaged by the acidic treatment but nevertheless, the electrical performance remained stable.
3.3. Performance assessment as compared to the literature
Table 1 compares the performance of our MFC set-up to the literature available.
Under the optimized conditions, the internal resistance of our
MFC system was 11 X and the electromotive force was 0.6 V. These
values correspond to the state of the art as showcased in Table
1.The maximum power attained was 7.4 mW, which corresponded
to a volumetric power of 2 W mÀ3 by considering the total volume
of our MFC being of 3.5 L (anodic compartment of 2.9 L and cathodic compartment of 0.6 L). This can appear low considering the
state of the art presented in Table 1. However, in terms of power
generated on a per cell basis, our MFC was in the highest range.
This is because the volumetric power is a decreasing function of
the geometrical parameters. However, by reducing the size too
much, the raw power output is also expected to decrease and this
mathematically increases the number of cells that would have to
be stacked together in order to provide enough power to be of
use. Considering the very low maturity of MFC-stacking, this is a
serious obstacle to MFC application. Furthermore, the electrical
resistivity and the volume of all the connectors that will be used
to perform stacking will also affect the electrical performance of
the system and its total volume. The impact on a volumetric power
based on the total volume and not on the working volume may become negative.
Moreover, the substrate used to operate an MFC – ranging
from acetate or glucose to complex industrial wastewater – is
known to impact widely on its performance and the best performance from the electrical as well as from the wastewater treatment point of views is usually obtained with synthetic and
simple substrates (see Table 1). However, in view of optimizing
MFCs for wastewater treatment, it is also useful to assess their
performance with domestic wastewater, but this does not play
in favor of our system. It is interesting to see that our system
could remove in average above 50% of the COD and 56% of the
SS at a HRT of 2.4 h. With domestic wastewater, Ahn and Logan
(2010) found COD removal higher than 88% in batch mode.
However, in that same study, during continuous operation maximizing power generation and at ambient temperature, COD removal was reduced to 19%, with a Coulombic efficiency of 0.7%
at a HRT of 4.2 min. In our study, the COD removal efficiency
was higher but Coulombic efficiency also remained below 1%,
Table 1
State of the art in microbial fuel cell research.
Substrate
OCV (V)
Pmax (mW cellÀ1)
Eemf (V)
Rint (X)
Two chamber design, ferricyanide catholyte
Lactate
0.6
Lactate
0.8
Sewage sludge
0.75
0.05
0.7
0.7
22,727
130
40
Two chamber design, aerated cathode
Glucose
0.7
Acetate
0.6
Acetate
0.8
0.7
0.6
0.8
122
11
9
1
7.8
18
Two chamber design, open-air cathode
Acetate
0.5
0.5
65
Single chamber design, open-air cathode
Domestic ww
0.8
Acid mine Drainage
0.6
Acetate
0.4
Domestic ww
NA
Acetate
0.8
Acetate
NA
Acetate
NA
Industrial ww
NA
Acetate
0.6
Acetate
0.8
Brewery ww
0.65
Molasses
0.55
Acetate
0.8
Acetate
NA
Acetate
0.85
Glucose
0.75
0.6
0.6
0.4
0.5
0.5
0.7
0.9
0.8
0.6
0.6
0.55
0.5
0.6
0.6
0.75
0.7
408
329
85
108
93
169
277
213
75
71
57
42
53
23
34
26
0.00002
0.94
3.06
V (mL)
0.0055
2.4
510
Pvmax (W mÀ3)
Ref.
4.4
392.6
6
Qian et al. (2009)
Ringeisen et al. (2006)
Jiang et al. (2009)
970
700
960
1
11.1
19
Deng et al. (2010)
Freguia et al. (2007)
Freguia et al. (2008a)
0.9
700
1.4
0.22
0.27
0.42
0.57
0.67
0.72
0.72
0.75
1.2
1.26
1.32
1.45
1.68
3.78
4.13
4.63
28
28
85
388
12
28
145
900
200
26
170
3280
26
2.5
336
210
7.8
9.7
4.9
1.4
55.9
25.8
5
0.8
6
48.4
7.7
0.4
64.6
1512
12.3
22
Erable et al. (2009)
Ahn and Logan (2010)
Cheng et al. (2007)
Lefebvre et al. (2009)
Liu et al. (2004)
Fan et al. (2008)
Wang et al. (2009)
Li et al. (2009)
Sun et al. (2009)
Kim et al. (2009)
Cheng and Logan (2007)
Zhuang et al. (2009)
Zhang et al. (2009)
Logan et al. (2007)
Fan et al. (2007)
Rabaey et al. (2008)
Feng et al. (2010)
NA, not available.
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
8
O. Lefebvre et al. / Bioresource Technology xxx (2011) xxx–xxx
except when the anodic compartment was filled with graphite
granules. In the latter case, Coulombic efficiency was much higher at 11%.
Low Coulombic efficiency has to be related to the complex nature of domestic wastewater allowing fermentation and other
biological processes to interfere with electricity generation. Furthermore, it seems that electrochemically active bacteria can only
use a limited range of products. For example, Freguia et al. (2008b)
showed that glucose had to be converted into acetate before it can
be utilized for electricity generation. It is consequently not surprising that butyrate and acetate, respectively, generated 62% and 82%
more power than domestic wastewater in the study of Ahn and
Logan (2010). Even though the low Coulombic efficiency could be
seen as a problem, it is merely the result of the MFC having reached
its maximum capacity as an electricity generation device. The electromotive force cannot be expected to go much beyond the values
observed in the present study – the theoretical maximum voltage
achievable in open circuit being approximately of 1.1 V in typical
MFC conditions (Logan et al., 2006) – and our values of internal
resistance are also competitive. As a result, any organic matter in
excess is further removed by other means. The vertical configuration of the reactor allowed sludge accumulation at the bottom of
the anodic compartment, where fermentation and even probably
methane production were allowed to take place, even though our
design did not allow for any gas collection and analysis to verify
that hypothesis. This made our MFC a hybrid system that overall
contributed to improve the elimination of organic matter, which
is the primary goal of a wastewater treatment plant. Overall, it
seems more important to optimize the effluent COD and SS
removal efficiencies in the anodic compartment rather than the
Coulombic efficiency in wastewater treatment application, which
further renders possible the introduction of the anolyte inside
the cathode compartment without reducing the performance of
the reactor, as shown in this study.
4. Conclusion
The present study shows the importance of overcoming proton
limitation in an MFC system and this was achieved first by letting
the anolyte flow into the cathode compartment (full-loop operation) and second by acidifying the cathode. The latter experiment
led to the best electrochemical performance and it appeared
clearly that the anolyte quality must be good enough before being
allowed to enter the cathodic compartment. We believe that the
future of the MFC technology as a treatment plant lies in the design
of hybrid systems that will harvest part of the energy in the form of
electricity and the rest in the form of biogas.
Acknowledgements
This work was supported by a Grant from the Environment &
Water and Industry Development Council, Singapore (MEWR
651/06/159).
References
Aelterman, P., Rabaey, K., Pham, H.T., Boon, N., Verstraete, W., 2006. Continuous
electricity generation at high voltages and currents using stacked microbial fuel
cells. Environ. Sci. Technol. 40, 3388–3394.
Ahn, Y., Logan, B.E., 2010. Effectiveness of domestic wastewater treatment using
microbial fuel cells at ambient and mesophilic temperatures. Bioresour.
Technol. 101, 469–475.
APHA., 2005. Standard Methods for Water and Wastewater Examination. 21 ed,
Washington.
Barbir, F., 2005. PEM Fuel Cells: Theory and Practice. Elsevier Academic, Amsterdam.
Cheng, S., Dempsey, B.A., Logan, B.E., 2007. Electricity generation from synthetic
acid-mine drainage (AMD) water using fuel cell technologies. Environ. Sci.
Technol. 41, 8149–8153.
Cheng, S., Liu, H., Logan, B.E., 2006. Increased power generation in a continuous flow
MFC with advective flow through the porous anode and reduced electrode
spacing. Environ. Sci. Technol. 40, 2426–2432.
Cheng, S.A., Logan, B.E., 2007. Ammonia treatment of carbon cloth anodes to
enhance power generation of microbial fuel cells. Electrochem. Commun. 9,
492–496.
Clauwaert, P., Mulenga, S., Aelterman, P., Verstraete, W., 2009. Litre-scale microbial
fuel cells operated in a complete loop. Appl. Microbiol. Biotechnol. 83, 241–247.
Deng, Q., Li, X.Y., Zuo, J.E., Ling, A., Logan, B.E., 2010. Power generation using an
activated carbon fiber felt cathode in an upflow microbial fuel cell. J. Power
Sources 195, 1130–1135.
Erable, B., Etcheverry, L., Bergel, A., 2009. Increased power from a two-chamber
microbial fuel cell with a low-pH air-cathode compartment. Electrochem.
Commun. 11, 619–622.
Fan, Y.Z., Hu, H.Q., Liu, H., 2007. Sustainable power generation in microbial fuel cells
using bicarbonate buffer and proton transfer mechanisms. Environ. Sci. Technol.
41, 8154–8158.
Fan, Y.Z., Sharbrough, E., Liu, H., 2008. Quantification of the internal resistance
distribution of microbial fuel cells. Environ. Sci. Technol. 42, 8101–8107.
Feng, Y.J., Lee, H., Wang, X., Liu, Y.L., He, W.H., 2010. Continuous electricity
generation by a graphite granule baffled air-cathode microbial fuel cell.
Bioresour. Technol. 101, 632–638.
Freguia, S., Rabaey, K., Yuan, Z., Keller, J., 2007. Non-catalyzed cathodic oxygen
reduction at graphite granules in microbial fuel cells. Electrochim. Acta 53, 598–
603.
Freguia, S., Rabaey, K., Yuan, Z.G., Keller, J., 2008a. Sequential anode-cathode
configuration improves cathodic oxygen reduction and effluent quality of
microbial fuel cells. Water Res. 42, 1387–1396.
Freguia, S., Rabaey, K., Yuan, Z.G., Keller, J., 2008b. Syntrophic processes drive the
conversion of glucose in microbial fuel cell anodes. Environ. Sci. Technol. 42,
7937–7943.
Gil, G.C., Chang, I.S., Kim, B.H., Kim, M., Jang, J.K., Park, H.S., Kim, H.J., 2003.
Operational parameters affecting the performance of a mediator-less microbial
fuel cell. Biosens. Bioelectron. 18, 327–334.
Harnisch, F., Wirth, S., Schroder, U., 2009. Effects of substrate and metabolite
crossover on the cathodic oxygen reduction reaction in microbial fuel cells:
Platinum vs. iron(II) phthalocyanine based electrodes. Electrochem. Commun.
11, 2253–2256.
Jiang, J.Q., Zhao, Q.L., Zhang, J.N., Zhang, G.D., Lee, D.J., 2009. Electricity generation
from bio-treatment of sewage sludge with microbial fuel cell. Bioresour.
Technol. 100, 5808–5812.
Kim, J.R., Premier, G.C., Hawkes, F.R., Dinsdale, R.M., Guwy, A.J., 2009. Development
of a tubular microbial fuel cell (MFC) employing a membrane electrode
assembly cathode. J. Power Sources 187, 393–399.
Lefebvre, O., Al-Mamun, A., Ooi, W.K., Tang, Z., Chua, D.H.C., Ng, H.Y., 2008. An
insight into cathode options for microbial fuel cells. Water Sci. Technol. 57,
2031–2037.
Lefebvre, O., Ooi, W.K., Tang, Z., Abdullah-Al-Mamun, M., Chua, D.H.C., Ng, H.Y.,
2009. Optimization of a Pt-free cathode suitable for practical applications of
microbial fuel cells. Bioresour. Technol. 100, 4907–4910.
Li, Z.J., Zhang, X.W., Zeng, Y.X., Lei, L.C., 2009. Electricity production by an overflowtype wetted-wall microbial fuel cell. Bioresour. Technol. 100, 2551–2555.
Liu, H., Ramnarayanan, R., Logan, B.E., 2004. Production of electricity during
wastewater treatment using a single chamber microbial fuel cell. Environ. Sci.
Technol. 38, 2281–2285.
Logan, B., Cheng, S., Watson, V., Estadt, G., 2007. Graphite fiber brush anodes for
increased power production in air-cathode microbial fuel cells. Environ. Sci.
Technol. 41, 3341–3346.
Logan, B.E., Hamelers, B., Rozendal, R., Schrorder, U., Keller, J., Freguia, S., Aelterman,
P., Verstraete, W., Rabaey, K., 2006. Microbial fuel cells: methodology and
technology. Environ. Sci. Technol. 40, 5181–5192.
Qian, F., Baum, M., Gu, Q., Morse, D.E., 2009. A 1.5 mu L microbial fuel cell for onchip bioelectricity generation. Lab on a Chip 9, 3076–3081.
Rabaey, K., Read, S.T., Clauwaert, P., Freguia, S., Bond, P.L., Blackall, L.L., Keller, J.,
2008. Cathodic oxygen reduction catalyzed by bacteria in microbial fuel cells.
Isme J. 2, 519–527.
Ringeisen, B.R., Henderson, E., Wu, P.K., Pietron, J., Ray, R., Little, B., Biffinger, J.C.,
Jones-Meehan, J.M., 2006. High power density from a miniature microbial fuel
cell using Shewanella oneidensis DSP10. Environ. Sci. Technol. 40, 2629–2634.
Rozendal, R.A., Hamelers, H.V.M., Buisman, C.J.N., 2006. Effects of membrane cation
transport on pH and microbial fuel cell performance. Environ. Sci. Technol. 40,
5206–5211.
Sun, J., Hu, Y.Y., Bi, Z., Cao, Y.Q., 2009. Improved performance of air-cathode singlechamber microbial fuel cell for wastewater treatment using microfiltration
membranes and multiple sludge inoculation. J. Power Sources 187, 471–479.
Wang, B., Han, J.I., 2009. A single chamber stackable microbial fuel cell with air
cathode. Biotechnol. Lett. 31, 387–393.
Wang, X., Cheng, S.A., Feng, Y.J., Merrill, M.D., Saito, T., Logan, B.E., 2009. Use of
carbon mesh anodes and the effect of different pretreatment methods on power
production in microbial fuel cells. Environ. Sci. Technol. 43, 6870–6874.
Zhang, B.G., Zhao, H.Z., Zhou, S.G., Shi, C.H., Wang, C., Ni, J.R., 2009. A novel UASBMFC-BAF integrated system for high strength molasses wastewater treatment
and bioelectricity generation. Bioresour. Technol. 100, 5687–5693.
Zhuang, L., Zhou, S.G., Wang, Y.Q., Liu, C.S., Geng, S., 2009. Membrane-less cloth
cathode assembly (CCA) for scalable microbial fuel cells. Biosens. Bioelectron.
24, 3652–3656.
Please cite this article in press as: Lefebvre, O., et al. Full-loop operation and cathodic acidification of a microbial fuel cell operated on domestic wastewater. Bioresour. Technol. (2011), doi:10.1016/j.biortech.2011.02.098
... al (20 10) Freguia et al (20 07) Freguia et al (20 08a) 0.9 700 1.4 0 .22 0 .27 0. 42 0.57 0.67 0. 72 0. 72 0.75 1 .2 1 .26 1. 32 1.45 1.68 3.78 4.13 4.63 28 28 85 388 12 28 145 900 20 0 26 170 328 0 26 2. 5... 0.75 0.7 408 329 85 108 93 169 27 7 21 3 75 71 57 42 53 23 34 26 0.000 02 0.94 3.06 V (mL) 0.0055 2. 4 510 Pvmax (W mÀ3) Ref 4.4 3 92. 6 Qian et al (20 09) Ringeisen et al (20 06) Jiang et al (20 09) 970... 2. 5 336 21 0 7.8 9.7 4.9 1.4 55.9 25 .8 0.8 48.4 7.7 0.4 64.6 15 12 12. 3 22 Erable et al (20 09) Ahn and Logan (20 10) Cheng et al (20 07) Lefebvre et al (20 09) Liu et al (20 04) Fan et al (20 08) Wang