Bioresource Technology 136 (2013) 707710

Contents lists available at SciVerse ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Short Communication

Effect of shear rate on the response of microbial fuel cell toxicity sensor
to Cu(II)
Yujia Shen a, Meng Wang 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
School of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheomdan-gwagiro, Buk-gu, Gwangju 500-712, Republic of Korea

h i g h l i g h t s
" The sensitivity of MFC sensors to toxicants is affected by the biolm structure.
" Most inuential parameters of MFC sensitivity: biolm porosity, density and EPS.
" Low ow rate reduced the biolm density and increased its porosity.
" Intermittent nitrogen sparging reduced the EPS content of the biolm.
" Low ow rate and intermittent nitrogen sparging increased the sensitivity of MFC.

a r t i c l e

i n f o

Article history:
Received 15 November 2012
Received in revised form 19 February 2013
Accepted 20 February 2013
Available online 26 February 2013
Keywords:
Microbial fuel cell
Biosensor
EPS
Toxicity
Cu(II)

a b s t r a c t
A microbial fuel cell (MFC) was successfully developed as a toxicity biomonitoring system, giving a quick
response to Cu(II) toxic events. The objective was to increase MFC sensitivity to Cu(II) toxicity by evaluating the impact of shear rate caused by mixing and intermittent nitrogen sparging on the biolm structure. Low shear rate achieved by continuously feeding the wastewater into the MFC at a low ow rate of
1.3 mL min 1 during the enrichment period resulted in low biomass density (124 g VSS L 1 of biolm),
high porosity and reduced levels of extracellular polymeric substances (EPS). Consequently, the sensitivity was improved. Scattered nitrogen sparging also increased the sensitivity by reducing the EPS level. It
suggested that MFC enriched under low ow rate with intermittent nitrogen sparging could produce an
anodic biolm that was less dense, more porous, contained less EPS and ultimately displayed higher sensitivity to Cu(II) toxicity.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
In the anode chamber of a microbial fuel cell (MFC), electrochemically active bacteria (EAB) oxidize organic matter and transfer the electrons to the anode where they pass through an external
circuit to reach the cathode of the system, generating current in the
process (Lovley, 2008). The presence of a toxic substance in the
system originated from the feed water can potentially inhibit the
metabolic activity of EAB and reduce the electron transfer as well
as the current output. This suggests that the current can be used
as a signal to monitor the occurrence and the intensity of a toxic
event, and consequently, MFC can potentially be used as an online
biosensor for the detection of toxic compounds in water (Kim et al.,
2007).

Corresponding author. Tel.: +65 6516 4777; fax: +65 6779 1635.
E-mail address: howyongng@nus.edu.sg (H.Y. Ng).
0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2013.02.069

Heavy metals often constitute the leading cause for disturbance

of wastewater treatment processes (Altas, 2009). Kim et al. (2007)
successfully used an MFC to detect Pb(II) and Hg(II) at a concentration of 1 ppm; however, Patil et al. (2010) showed that in the presence of Pb(II) and Hg(II) at concentrations in the range of 0.41
12.48 and 0.838.33 ppm, respectively, planktonic cells were
inhibited, while the electrochemically active biolm was not affected. The contradiction between these two studies suggests that
the behavior of the electrochemically active biolm and thus the
sensitivity of an MFC as a toxicity sensor can be inuenced by several factors, among which biolm structure (i.e., its density, porosity and composition) could be a key parameter.
Hydrodynamic shear rates are known to affect mass transfer
conditions, biolm structure and the production of extracellular
polymeric substances (EPS) (Celmer et al., 2008), all of which are
factors that can affect the diffusivity of toxicants and their interaction with the biolm (Henriques and Love, 2007; Liu and Tay,
2002). As such, shear rates play an important role in biolm reactors and shear rates have already been shown to impact on the

708

Y. Shen et al. / Bioresource Technology 136 (2013) 707710

density of MFC biolms (Pham et al., 2008). Both intensive mixing

and nitrogen sparging have been validated as effective methods to
regulate the shear rate (Celmer et al., 2008).
Consequently, the objective of this study was to assess the characteristics of the electrochemically active biolm of an MFC under
different shear rates controlled by the ow rate and intermittent
nitrogen sparging and these characteristics were further correlated to the MFC sensitivity to Cu(II). The ultimate goal was to improve the sensitivity of the MFC toxicity sensor through
modication of the biolm structure.
2. Methods

(5000 rpm; 1 min) according to Xing et al. (2010). VSS were analyzed according to the Standard Methods (APHA, 2005). The protein and carbohydrate concentrations of EPS were determined by
the modied Lowry method (Frolund et al., 1995) and the Dubois
phenolsulfuric acid method (Dubois et al., 1956), respectively.
Scanning electron microscopy (SEM) was carried out to investigate
both the surface and the thickness of the biolm. The samples were
soaked in a 2.5% glutaraldehyde solution for 30 min at room temperature (25 C) for xation. They were then dehydrated through
an ascending ethanol series from 25% to 100% at room temperature. Prior to observation, the samples were nally critical point
dried and coated with gold in a sputtering device. Three different
points were observed for each piece.

2.1. MFC construction and operation

2.4. Calculations
The MFCs used in this study were of single-chamber aircathode designs (Liu and Logan, 2004) equipped with a proton exchange membrane (Selemion, Asahi Glass Co., Japan). The
working volume of each cylindrical MFC was 28 mL. Both the anode and the cathode (each having a surface area of 28 cm2) consisted of carbon cloth (E-Tek, USA) and the cathode was coated
with platinum catalyst on one side at a load of 0.5 mg Pt cm 2.
Eight MFC units (two for each shear condition) were inoculated
and operated with real domestic wastewater (300
400 mg COD L 1) collected from the primary clarier of the Ulu
Pandan Reclamation Plant, Singapore in a continuous-ow mode
for 2 months. Prior to feeding, the efuent was ltered with a
screen of 200-lm pore size. Feeding was performed with peristaltic pumps (Masterex 07523-70, Spectra-Teknik Pte Ltd., Singapore) in an upow mode at four different ow rates, i.e.,
1.3 mL min 1 (hydraulic retention time (HRT) of 20 min),
12 mL min 1 (HRT of 2 min), 12 mL min 1 (HRT of 2 min) and
24 mL min 1 (HRT of 1 min). For one of the two MFC sets which
were operated at a ow rate of 12 mL min 1, nitrogen sparging
was applied twice a week. Nitrogen sparging was carried out by
stopping the feed to the MFCs and aerating the MFCs with nitrogen
at a ow rate of 300 mL min 1 for 10 min. The voltage (V) across a
5 O resistor connected between the anode and cathode of the MFC
was continuously measured with a digital multimeter connected to
a PC by a data acquisition system (M3500A, Array Electronic,
Taiwan).
2.2. MFC inhibition by Cu(II)
After an enrichment period of 2 months, the MFCs was subjected to Cu(II) toxicity assessment. The toxic substance used in
this study was Cu(II) solution, as 10,000 mg L 1 Cu2+ in 5% nitric
acid (SigmaAldrich Pte. Ltd., Singapore). The effect of Cu(II) on
the performance of the MFCs was carried out as follows: (1) The
feed tank of the MFCs was lled to 10 L with domestic wastewater;
(2) After stabilization for half an hour, 5 or 7 mL of the above Cu(II)
solution was spiked into the feed tank to make a toxic feed containing either 5 or 7 ppm of Cu(II), respectively; (3) The MFCs were
exposed to the toxicants for 4 h (i.e., feeding for 4 h); and (4) Subsequently, fresh domestic wastewater without Cu(II) toxicant was
fed into the MFCs. For each testing condition, MFCs in duplicate
were used.

The inhibition ratio (I) induced by Cu(II) was calculated by I

(%) = 100  (Vnor Vtox)/Vnor according to Kim et al. (2007), where
Vnor (V) is the voltage generated by the microbial fuel cell in the absence of Cu(II) and Vtox (V) is the voltage output directly following
the introduction of Cu(II). Biolm density (D, g VSS L 1) was determined as D = m/L  104, where m (mg cm 2) is the VSS content of
the biolm collected from the 1 cm2 carbon cloth and L is the thickness of the biolm (lm).
3. Results and discussion
3.1. Response of MFCs enriched under different shear rates to Cu (II)
toxicity
After steady state was achieved, toxic incidents were created by
spiking Cu(II) either at 5 or 7 ppm to the wastewater. An immediate decrease in voltage was observed after exposure to Cu(II)
(Fig. 1). The inhibition ratios of the MFCs by Cu(II) at 5 and
7 ppm under enrichment ow rate of 12 and 24 mL min 1 were
similarly low, with inhibition extended beyond 2 h of Cu(II) exposure. Inhibition ratios at 5 ppm of Cu(II) after 4 h were around 30%
for both 12 and 24 mL min 1, and around 50% and 60% at 7 ppm of
Cu(II) for 12 and 24 mL min 1, respectively. However, the degree of
inhibition was inversely proportional to the shear rate, and the
highest inhibition to the toxic event was observed at a ow rate
of 1.3 mL min 1, which after 2 h of Cu(II) exposure led to a 60%
inhibition ratio at 5 ppm of Cu(II) and 85% inhibition ratio at
7 ppm of Cu(II) after 4 h. Intermittent nitrogen sparging increased
the extent of inhibition after 2 h of Cu(II) exposure from 30% to 40%
at 5 ppm of Cu(II) and from 40% to 85% at 7 ppm of Cu(II), showing
a stronger effect at high Cu(II) concentration.
The result showed that an MFC had been successfully developed
as a toxicity biomonitoring system, giving a quick response to a
Cu(II) toxic event. The fast drop of the performance of the MFC
may appear contradictory to the ndings of Patil et al. (2010), in
which no effect on the biolm of the MFC was observed even at
6 ppm of Cu(II). This apparent contradiction, however, proved from
the ndings in this study that the biolm was a key factor determining the sensitivity of the MFC to toxicity, and the shear rate
could be an effective control parameter.
3.2. Effect of the shear rate on the biolm structure

2.3. Biolm analysis

At the end of each experimental period, the anode carbon cloth
was removed from the MFCs, cut into 16 pieces of 1 cm2 squares
and collected in 50 mM phosphate buffer solutions. Different
pieces were used for volatile suspended solids (VSS), EPS and
SEM analysis. VSS and EPS were recovered by centrifugation

The operating conditions applied to each set of MFCs and their

effect on the biolm characteristics are summarized in Table 1. The
VSS averaged 3.68 mg VSS cm 2 of electrode at all ow rates,
showing insignicant effect of the shear rate; however, the biolm
thickness was impacted. At a ow rate of 1.3 mL min 1 (low shear
stress), the thickness of the biolm was about 300 lm. When the

709

Y. Shen et al. / Bioresource Technology 136 (2013) 707710

(b)

100

100

-1

Q = 1.3 mL min
-1
Q = 12 mL min
-1
Q = 24 mL min
-1
Q = 12 mL min + N2

80

Inhibition Ratio (%)

Inhibition Ratio (%)

(a)

60
40
20

80
60
40

-1

Q = 1.3 mL min
-1
Q = 12 mL min
-1
Q = 24 mL min
-1
Q = 12 mL min + N2

20

0
0

Time (h)

Time (h)

Fig. 1. Inhibition ratio of MFCs enriched under different ow rates (Q) exposed to Cu(II) at a concentration of (a) 5 ppm; and (b) 7 ppm. The arrow indicates the beginning of
the toxic event.

Table 1
Volatile suspended solids (VSS), thickness, density and EPS content (protein and carbohydrate) of the anodic biolm of a microbial fuel cell enriched under different shear rates.
Flow rate (mL/
min)

Nitrogen
sparging

VSS (mg/cm2 of
electrode)

Thickness
(lm)

Density (g VSS/L of
biolm)

Protein content (mg/g of

VSS)

Carbohydrate content (mg/g of

VSS)

1.3
12
24
12

No
No
No
Yes

3.67 0.31
3.43 0.54
3.83 0.59
3.79 0.11

295.5 0.7
202.5 2.1
101 5.6
200 15.5

124
170
380
190

2.37 0.48
6.19 3.12
6.48 1.87
2.35 0.31

1.70 0.16
2.87 0.63
2.66 0.55
1.37 0.03

ow was increased to 12 mL min 1, the thickness further decreased to about 200 lm (33% reduction) and nally dropped to
about 100 lm (67% reduction) under the highest ow rate of
24 mL min 1. As a result, the biolm density increased with ow
rate up to 380 g VSS L 1 of biolm at the highest ow rate of
24 mL min 1, indicating that high shear rates resulted in stronger
aggregation and a denser biolm. Denser colonization of the anode
under high shear rate was further conrmed by SEM (Figure not
shown). The electrode surface enriched under high shear rate
was covered much more densely with bacterial cells and appeared
less porous in contrast with that enriched under low shear rate.
Intermittent nitrogen sparging on the other hand did not signicantly affect the biom thickness or density (Table 1).
Our results support the hypothesis that increasing ow rate resulted in increasing shear rate that modied the biolm characteristics, making the developed biolm denser and more compact
with reduced porosity (i.e., less porous). The thickness of the biolm also decreased up to 65% with increasing ow rate. This observation is in good agreement with the works of Kwok et al. (1998)
and Celmer et al. (2008), but contradicted with the conclusions
of Pham et al. (2008) and Rochex et al. (2008). The latter two studies showed that the biolm thickness increases with shear stress
due to increased mass transfer and biomass production. However,
the increased shear rates in our study caused the detachment of
the top layer of the biolm, which has also been shown by Coufort
et al. (2007) whereby about 60% of the biolm mass was fragile and
easily detached. In our study, nitrogen sparging did not further impact on the biolm structure, probably because its intermittent
nature (twice a week) made it insignicant as compared to the effect of the ow rate.
The relationship between the modied biolm structure and
the improved sensitivity of the electrochemically active biolm enriched under low shear rates can be explained by the one-dimensional mass transport model used by Hu et al. (2007) that
describes the spatial distribution of Cu(II) in biolms. According
to this model, the retardation factor (Rf) for diffusive transport of

Cu(II) can be described as Rf = 1 + (D/U)  Kd, where D is the density (g L 1), U is the porosity and Kd is the metal-biomass partition
coefcient. Therefore, low cell density along with the increased
porosity observed under low shear rates are expected to increase
the diffusivity of Cu(II) in the biolm and hence, enhance the toxic
effect, which explains the results observed in this study.
Another interesting nding from the present study is that the
thick biolm obtained under low shear rates did not hinder the
mass transfer of Cu(II). In some cases such as the study conducted
by Hu et al. (2007), Cu(II) could only penetrate the rst 150 lm of
the biolm instead of the whole thickness of the biolm after exposure for 2 h at a concentration of 0.2 mM. However, in our study,
with the fact that EAB only persist up to tens of micrometers away
from the anode and inhabit the inner layer of the biolm (Marcus
et al., 2007), the highest inhibition observed with the MFCs operated under the lowest ow rate (Fig. 1) that had the highest biolm
thickness of nearly 300 lm, suggesting that Cu(II) penetrated
much deeper into the biolm in our study. And it can be concluded
that biolm density and porosity have a stronger impact on the
MFC sensitivity than its actual thickness.
3.3. Effect of the shear rate on the EPS content of the biolm
High shear rates led to overproduction of EPS (Table 1). When
the ow rate was increased from 1.3 to 12 mL min 1, the protein
and the carbohydrate content of the biolm increased from 2.37
to 6.19 mg g 1 of VSS (161% increase) and from 1.70 to 2.87 mg g 1
of VSS (69% increase), respectively. However, increasing the ow
rate to 24 mL min 1 did not further affect the EPS content. Scattered nitrogen sparging, on the other hand, signicantly decreased
the protein and the carbohydrate content from 6.19 to 2.35 mg g 1
of VSS (62% reduction) and from 2.87 to 1.37 mg g 1 of VSS (52%
reduction), respectively.
Our results suggested that high ow rates hence high shear
rates led to EPS overproduction (see Table 1) which resulted in
a loss of sensitivity of the MFC sensor. Stoodley et al. (2002)

710

Y. Shen et al. / Bioresource Technology 136 (2013) 707710

explained that the overproduction of EPS under high shear rates

aimed at protecting the biolm from the physical environment.
However, intermittent nitrogen sparging produced the opposite effect as it signicantly decreased the EPS production by 5060%. In
addition, intermittent decrease or increase of the shear rate might
be benecial to reduce the EPS content and a probable mechanism
is the breakdown of the EPS matrix and simultaneous release of
EPS into the bulk liquid as explained by Henriques and Love
(2007). Reduced levels of EPS were associated with improved sensitivity of the MFC sensor and this can be related to the nature of
the polysaccharides and proteins that constituted the EPS and
which functional groups (e.g., carboxylic acids and amino acids)
can bind positively charged metal ions (Fang et al., 2002). Hence,
reduced EPS content improved the sensitivity of the MFC sensor
under low shear rates and scattered nitrogen sparging.
4. Conclusions
The MFC toxicity sensor developed here allowed the fast
monitoring of the Cu(II) toxicity in the wastewater. It can also be
postulated that biolm density, porosity and EPS content of the
biolm affect the sensitivity of MFC sensors to heavy metals.
Therefore, ow rate and nitrogen sparging seem to be two operational parameters that can be easily used to control the characteristics of the biolm developed on the anode of MFCs. To enhance
the sensitivity of MFC as a toxicity sensor for heavy metals, MFCs
should be operated under low ow rate and intermittent nitrogen
sparging.
Acknowledgements
This study was supported by a grant from the Environment &
Water and Industry Development Council, Singapore (MEWR
651/06/159).
References
Altas, L., 2009. Inhibitory effect of heavy metals on methane-producing anaerobic
granular sludge. Journal of Hazardous Materials 162 (23), 15511556.
APHA, 2005. Standard Methods for Examination of Water and Wastewater, 21th ed.,
Washington, DC.
Celmer, D., Oleszkiewicz, J.A., Cicek, N., 2008. Impact of shear force on the biolm
structure and performance of a membrane biolm reactor for tertiary

hydrogen-driven denitrication of municipal wastewater. Water Research 42

(12), 30573065.
Coufort, C., Derlon, N., Ochoa-Chaves, J., Line, A., Paul, E., 2007. Cohesion and
detachment in biolm systems for different electron acceptor and donors.
Water Science and Technology 55 (89), 421428.
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric
method for determination of sugars and related substances. Analytical
Chemistry 28 (3), 350356.
Fang, H.H.P., Xu, L.C., Chan, K.Y., 2002. Effects of toxic metals and chemicals on
biolm and biocorrosion. Water Research 36 (19), 47094716.
Frolund, B., Griebe, T., Nielsen, P.H., 1995. Enzymatic-activity in the activatedsludge oc matrix. Applied Microbiology and Biotechnology 43 (4), 755
761.
Henriques, I.D.S., Love, N.G., 2007. The role of extracellular polymeric substances in
the toxicity response of activated sludge bacteria to chemical toxins. Water
Research 41 (18), 41774185.
Hu, Z.Q., Jin, J., Abruna, H.D., Houston, P.L., Hay, A.G., Ghiorse, W.C., Shuler, M.L.,
Hidalgo, G., Lion, L.W., 2007. Spatial distributions of copper in microbial
biolms by scanning electrochemical microscopy. Environmental Science and
Technology 41 (3), 936941.
Kim, M., Hyun, M.S., Gadd, G.M., Kim, H.J., 2007. A novel biomonitoring system
using microbial fuel cells. Journal of Environmental Monitoring 9 (12), 1323
1328.
Kwok, W.K., Picioreanu, C., Ong, S.L., van Loosdrecht, M.C.M., Ng, W.J., Heijnen, J.J.,
1998. Inuence of biomass production and detachment forces on biolm
structures in a biolm airlift suspension reactor. Biotechnology and
Bioengineering 58 (4), 400407.
Liu, H., Logan, B., 2004. Electricity generation using an air-cathode single chamber
microbial fuel cell (MFC) in the presence and absence of a proton exchange
membrane. Environmental Science and Technology 38 (14), 40404046.
Liu, Y., Tay, J.H., 2002. The essential role of hydrodynamic shear force in the
formation of biolm and granular sludge. Water Research 36 (7), 16531665.
Lovley, D.R., 2008. The microbe electric: conversion of organic matter to electricity.
Current Opinion in Biotechnology 19 (6), 564571.
Marcus, A.K., Torres, C.I., Rittmann, B.E., 2007. Conduction-based modeling of the
biolm anode of a microbial fuel cell. Biotechnology and Bioengineering 98 (6),
11711182.
Patil, S., Harnisch, F., Schroder, U., 2010. Toxicity response of electroactive microbial
biolms a decisive feature for potential biosensor and power source
applications. ChemPhysChem 11 (13), 28342837.
Pham, H.T., Boon, N., Aelterman, P., Clauwaert, P., De Schamphelaire, L., van
Oostveldt, P., Verbeken, K., Rabaey, K., Verstraete, W., 2008. High shear
enrichment improves the performance of the anodophilic microbial
consortium in a microbial fuel cell. Microbial Biotechnology 1 (6), 487496.
Rochex, A., Godon, J.J., Bernet, N., Escudie, R., 2008. Role of shear stress on
composition, diversity and dynamics of biolm bacterial communities. Water
Research 42 (20), 49154922.
Stoodley, P., Cargo, R., Rupp, C.J., Wilson, S., Klapper, I., 2002. Biolm material
properties as related to shear-induced deformation and detachment
phenomena. Journal of Industrial Microbiology and Biotechnology 29 (6),
361367.
Xing, D.F., Cheng, S.A., Logan, B.E., Regan, J.M., 2010. Isolation of the exoelectrogenic
denitrifying bacterium Comamonas denitricans based on dilution to extinction.
Applied Microbiology and Biotechnology 85 (5), 15751587.