APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Dec. 2009, p. 7579–7587
0099-2240/09/$12.00 doi:10.1128/AEM.01760-09
Copyright © 2009, American Society for Microbiology. All Rights Reserved.
Vol. 75, No. 24
Hydrogen Production by Geobacter Species and a Mixed
Consortium in a Microbial Electrolysis Cell䌤
Douglas F. Call, Rachel C. Wagner, and Bruce E. Logan*
Hydrogen Energy Center and Department of Civil and Environmental Engineering,
212 Sackett Building, Pennsylvania State University, University Park, Pennsylvania 16802
Received 24 July 2009/Accepted 2 October 2009
A hydrogen utilizing exoelectrogenic bacterium (Geobacter sulfurreducens) was compared to both a nonhydrogen oxidizer (Geobacter metallireducens) and a mixed consortium in order to compare the hydrogen production rates and hydrogen recoveries of pure and mixed cultures in microbial electrolysis cells (MECs). At an
applied voltage of 0.7 V, both G. sulfurreducens and the mixed culture generated similar current densities (ca.
160 A/m3), resulting in hydrogen production rates of ca. 1.9 m3 H2/m3/day, whereas G. metallireducens exhibited
lower current densities and production rates of 110 ⴞ 7 A/m3 and 1.3 ⴞ 0.1 m3 H2/m3/day, respectively. Before
methane was detected in the mixed-culture MEC, the mixed consortium achieved the highest overall energy
recovery (relative to both electricity and substrate energy inputs) of 82% ⴞ 8% compared to G. sulfurreducens
(77% ⴞ 2%) and G. metallireducens (78% ⴞ 5%), due to the higher coulombic efficiency of the mixed consortium.
At an applied voltage of 0.4 V, methane production increased in the mixed-culture MEC and, as a result, the
hydrogen recovery decreased and the overall energy recovery dropped to 38% ⴞ 16% compared to 80% ⴞ 5%
for G. sulfurreducens and 76% ⴞ 0% for G. metallireducens. Internal hydrogen recycling was confirmed since the
mixed culture generated a stable current density of 31 ⴞ 0 A/m3 when fed hydrogen gas, whereas G. sulfurreducens exhibited a steady decrease in current production. Community analysis suggested that G. sulfurreducens
was predominant in the mixed-culture MEC (72% of clones) despite its relative absence in the mixed-culture
inoculum obtained from a microbial fuel cell reactor (2% of clones). These results demonstrate that Geobacter
species are capable of obtaining similar hydrogen production rates and energy recoveries as mixed cultures in
an MEC and that high coulombic efficiencies in mixed culture MECs can be attributed in part to the recycling
of hydrogen into current.
Electrohydrogenesis is an efficient method for generating
hydrogen gas from organic matter in reactors known as microbial electrolysis cells (MECs) (17, 18, 26). MECs differ from
air-cathode microbial fuel cells (MFCs) in that the cathode
remains anaerobic, and voltage is added in order to generate
hydrogen at the cathode. Under the biological conditions in
MECs, hydrogen evolution is not a thermodynamically favorable reaction. However, combining the hydrogen formation
reaction potential of ⫺0.41 V at the cathode (ECAT) with the
anode potential (EAN) typically obtained in MFCs with an EAN
of ⫺0.30 V (1 g of acetate/liter) results in a minimum required
voltage of only 0.14 V. Applied voltages (EAP) of 0.2 V (0.45
kWh/m3 H2) or larger are needed in practice to produce measurable quantities of hydrogen, but this input is substantially
less than the average of 2.3 V (5.1 kWh/m3 H2) required for
water electrolysis (13).
Recent improvements in designs and materials have substantially improved hydrogen yields, production rates, and energy recoveries (3, 18, 27–29, 33). Hydrogen recoveries using
typical dead-end fermentation end products such as acetate
and butyrate have reached 80 to 100%, whereas other complex
substrates such as glucose and cellulose have yielded recoveries
of ca. 70% (5). Production rates larger than 6 m3 H2/m3/day
have been obtained using MECs (32), which are similar to an
average rate of 2.5 m3 H2/m3/day obtained for hydrogen production by biological fermentation (10). Energy recoveries relative to the electrical energy input as high as 680% have already been shown (5), and overall energy recoveries that
include the energy of the substrate have reached 85% (2, 5).
Hydrogen losses can occur using a mixed culture in an MEC,
reducing hydrogen yields, production rates, and recoveries (3,
11, 16, 32). Hydrogen recoveries can drop significantly at lower
applied voltages in membraneless MECs because of methanogenic consumption of hydrogen (2, 8, 11, 34). Using a membraneless MEC, Call and Logan (2) found that the overall
hydrogen recovery of 90% at an EAP of 0.6 V was reduced to
18% at an EAP of 0.2 V and that methane concentrations
increased from 0.9 to 28% in the product gas. Reducing solution pH can help inhibit methanogens, but a methane concentration of 22% was observed in a membrane free MEC at pH
5.8 (11). When hydrogen is the intended product of an MEC,
methane production is detrimental to the process. However,
biologically produced methane is a renewable energy source,
and membraneless MECs can be used to generate methane
instead of hydrogen, although energy recoveries are lower (8).
Hydrogen can also be consumed by chemolithotrophic bacteria
in mixed-culture MECs. These bacteria may transfer the associated electrons to a suitable electron acceptor, such as carbon
dioxide, and in some cases, the anode. In the latter scenario,
the electrons from hydrogen would be recycled internally,
causing an increase in coulombic efficiency (16). Hydrogen
losses reduce hydrogen and energy recoveries, and alternative
* Corresponding author. Mailing address: Department of Civil and
Environmental Engineering, 212 Sackett Building, Penn State University, University Park, PA 16802. Phone: (814) 863-7908. Fax: (814)
863-7304. E-mail: blogan@psu.edu.
䌤
Published ahead of print on 9 October 2009.
7579
7580
CALL ET AL.
methods for generating methane-free and high hydrogen content gas are needed.
Pure culture MECs are one method to avoid losses to methanogens, but production rates and efficiencies with pure cultures can be low compared to those with mixed cultures. Using
a pure culture of Shewanella oneidensis MR-1 and lactate, Hu
et al. obtained a hydrogen production rate of 0.025 m3 H2/m3/
day at an EAP of 0.6 V (11). However, production rates at this
same applied voltage using mixed cultures have reached 1 to 2
m3 H2/m3/day (2, 5). In MFCs, S. oneidensis has produced low
coulombic efficiencies (⬍10%) (24, 25) and maximum current
densities of ca. 50 mA/m2 (15) with lactate, compared to ca.
9,900 mA/m2 (9) for mixed cultures.
Several Geobacter species are commonly found in mixed
culture MFCs, and tests with pure cultures of Geobacter sulfurreducens have demonstrated power and current densities
close to or equal to those achieved with mixed cultures. In an
air cathode MFC, G. sulfurreducens produced a lower power
density (461 mW/m2, 1.5 A/m2) than a mixed culture (576
mW/m2, 1.3 A/m2) (12). The reduced performance of G. sulfurreducens in the air cathode MFC may have been due to
oxygen intrusion across the cathode. Using an MFC with a
ferricyanide cathode, Nevin et al. (23) reported a power density of 1.9 W/m2 (4.6 A/m2) for G. sulfurreducens compared to
1.6 W/m2 (3.2 A/m2) for a mixed consortium. When the authors placed the G. sulfurreducens MFC in an anaerobic chamber, the coulombic efficiency improved from 55% to ca. 100%,
confirming the importance of strictly anaerobic conditions for
G. sulfurreducens. This suggests that the anaerobic environment of MECs may provide excellent conditions for obtaining
current densities comparable to those of mixed cultures with
pure cultures of Geobacter species, while at the same time
eliminating methane gas production.
In order to investigate the performance of Geobacter species
in MECs, we selected two Geobacter species based on their
differences in hydrogen utilization. G. sulfurreducens was selected because it is capable of producing high current densities
in MFCs, and it can utilize hydrogen. G. metallireducens, which
does not oxidize hydrogen, was examined to determine
whether higher hydrogen recoveries were possible with a bacterium that cannot oxidize hydrogen. Both of these cultures
were compared to a mixed culture under identical conditions
in order to further examine the role of internal hydrogen
recycling in MECs and to show that methane-free gas can be
produced in MECs at rates comparable to those obtained with
mixed cultures.
MATERIALS AND METHODS
Microorganisms and culture media. G. sulfurreducens PCA (ATCC 51573)
and G. metallireducens GS-15 (ATCC 53774) were obtained from the American
Type Culture Collection (ATCC) and cultured using the ATCC media 1957 and
1768, respectively. Sodium acetate (6.8 g/liter) was provided as an electron
donor, and ferric citrate (13.7 g/liter) or sodium fumarate (8.0 g/liter) was
supplied as an electron acceptor for G. metallireducens and G. sulfurreducens,
respectively. Growth medium (75 ml) was added to serum bottles (125-ml capacity), sparged with anaerobic N2/CO2 (80/20 [vol/vol]) gas, sealed with thick
stoppers and aluminum crimp tops, and sterilized by autoclaving. To inoculate
the Geobacter MECs, 40 ml of growth medium containing cells was centrifuged,
the supernatant was removed, and a pellet of cells was resuspended in sterile
MEC medium. The medium used for the MECs (MEC medium) was identical to
the growth medium except that it contained 1 g of sodium acetate/liter and
lacked ferric citrate or fumarate. The conductivity of the MEC medium was 4.0
APPL. ENVIRON. MICROBIOL.
FIG. 1. The MEC single-chamber reactor design used in this study
is shown. A reference electrode is shown extending from the front of
the reactor.
mS/cm, and the initial pH was 6.8. The mixed consortium was obtained from a
previously acclimated MFC operating for several months with the same MEC
medium except that a phosphate buffer (50 mM phosphate-buffered saline;
Na2HPO4, 4.58 g/liter; NaH2PO4 䡠 H2O, 2.45 g/liter [pH ⫽ 7.0]) was used instead
of bicarbonate. After three batches of reproducible voltage generation, the
mixed-culture MFC anode was placed in sterile MEC medium containing glass
beads, vortexed to suspend the cells, and used to inoculate the MECs (10%
[vol/vol]).
MEC construction and operation. Identical single-chamber MECs (no membrane) made from a solid polycarbonate block were drilled to form a cylindrical
interior with a length of 2.5 cm and a diameter of 3 cm (empty bed volume of 18
ml) (Fig. 1). A glass gas collection tube (5 ml) was attached to the top of the cubic
reactor as previously described (2). The anodes were ammonia-treated carbon
cloth (type A; E-TEK), and the cathodes were carbon cloth (type B; E-TEK)
containing 0.5 mg Pt/cm2 catalyst on the side facing the anode (2, 4). Both
electrodes had projected surface areas of 7 cm2. After crimping the collection
tubes shut with thick rubber stoppers, and sterilizing the fully assembled MECs
by autoclaving, an ethanol-sterilized Ag/AgCl reference electrode (⫹200 mV
versus a standard hydrogen electrode [RE-5B; BASi]) was installed in the middle
of the reactor to record the EAN. The MECs were filled with anaerobic MEC
medium and inoculated in an anaerobic glove box (N2/H2 ratio of 95/5 [vol/vol]).
A control lacking an inoculum was also prepared by using the same procedures.
Each MEC was operated in fed-batch mode by draining the contents of the
reactor at the end of each batch, refilling with sterile MEC medium containing
substrate, and sparging the headspace with sterile, anoxic gas (N2/CO2 ratio of
80/20 [vol/vol]). Operating the MECs in this manner produced a rise in current
upon feeding and then a drop in current after the substrate was depleted. Before
and after each batch, the pH of the MEC medium for each culture was near
neutral (ca. 7.0 to 7.5).
Continuous gas production was recorded by using a respirometer (AER-200;
Challenge Environmental), and the produced gas was collected by using gas bags
(0.1 liter; Calibrated Instruments) as described previously (2). A fixed voltage
(EAP) was applied to the MECs by using a power supply (model 3645A; Circuit
Specialists), and the voltage across a resistor (10 ⍀) was recorded by using a
multimeter (model 2700; Keithley Instruments, Inc.) in order to calculate the
current (2). The EAN potential relative to the Ag/AgCl electrode was also
recorded. All anode potentials are reported relative to the Ag/AgCl reference
unless otherwise stated. For the H2 utilization test, the MECs were filled with
MEC medium lacking acetate and continuously sparged with sterile H2/CO2
(80/20 [vol/vol]) gas.
VOL. 75, 2009
Acetate analysis. Acetate was analyzed by using a gas chromatograph (GC;
6890N; Agilent) equipped with a flame ionization detector and a CA-FFAP
fused-silica capillary column (30 m by 0.32 mm by 0.5 m). Samples were filtered
through a 0.2-m-pore-diameter membrane and were acidified by using formic
acid (0.63 M [final concentration]) before analysis. The temperature of the GC
column was started at 60°C, increased at 20°C/min to 120°C, and then at 30°C/
min to a final temperature of 240°C for another 3 min. The temperatures of
injector and detector were both 250°C. Helium was used as the carrier gas at a
constant column flow rate of 3.30 ml/min.
Gas analysis. At the end of each batch, the gas composition in the gas bag was
analyzed by gas chromatography using a gastight syringe (250 l; Hamilton
Samplelock Syringe). The concentrations of H2, N2, and CH4 were analyzed with
one GC (argon carrier gas [model 2610B; SRI Instruments]), and the concentration of CO2 was analyzed with a separate GC (helium carrier gas [model 310,
SRI Instruments]).
Calculations. MEC performance was characterized using calculations as described previously (2, 18). Briefly, the current was normalized to either the liquid
volume of reactor IV (A/m3) or the anode surface area IA (A/m2). The coulombic
efficiency, CE (%), was calculated as CE ⫽ CT/CC, where CT is the total coulombs
calculated by integrating the current over time, and CC is the total charge
consumed based on acetate removal (eight electrons per acetate). To convert the
acetate concentration (g/liter) to coulombs, a chemical oxygen demand (COD) of
1.07 g-COD/liter was used along with Faraday’s constant (F ⫽ 96,500 C/mol).
Cathodic hydrogen recovery, rCAT (%), was calculated as rCAT ⫽ nR/nE, where
nR (mol) is the amount of H2 produced, and nE is the amount of H2 expected
based on the current. Overall hydrogen recovery, rH2 (%), was calculated as
rH2 ⫽ CE rCAT. The H2 production rate, QV (m3 H2/m3/day) was calculated as
QV ⫽ 43.2 IV rCAT/F cg(T), where cg is the concentration of a gas at a temperature T calculated using the ideal gas law, and 43.2 is for unit conversion.
Calculated parameters are based on data collected over a fed-batch cycle after
performance (based on current generation) had been stable for at least two
successive batch cycles at each applied voltage. The average values for each
successive cycle (three or more separate cycles) was then used to calculate the
reported averages and standard deviations.
Energy recovery relative to only the electrical energy input, E (%), was
calculated as E ⫽ WH2/WE, where WH2 is the energy of the produced hydrogen
(for ⌬H calculations, the higher heating value of 285.8 kJ/mol H2 was used; for
⌬G calculations, the Gibb’s free energy of 237.2 kJ/mol H2 was used), and WE is
the electricity input determined as WE ⫽ CT EAP, where EAP is the applied
voltage corrected for the power loss across the resistor used to measure the
current. The overall energy recovery, E⫹S (%), was calculated as E⫹S ⫽
WH2/(WE ⫹ WS), where WS is the energy of the removed acetate (for ⌬H
calculations, 874.3 kJ/mol acetate was used; for ⌬G calculations, 844.1 kJ/mol
acetate was used). When noted, methane was included in the energy recovery for
the mixed-culture MECs (⌬H ⫽ 74.6 kJ/mol CH4; ⌬G ⫽ 50.5 kJ/mol CH4).
CV. In order to further compare electrode reducing rates between G. sulfurreducens and the mixed culture MEC, cyclic voltammetry (CV) was performed by
using a potentiostat (PC 4/750; Gamry Instruments, Inc.) at a scan rate of 10
mV/s, over a potential range of ⫺0.8 to 0.2 V relative to the Ag/AgCl reference
electrode. Prior to the CV scans, the MECs were filled with fresh MEC medium
and operated for at least 1 h at EAP ⫽ 0.7 V. The peak current was defined as
the maximum current produced averaged over a period of 3 s.
Scanning electron microscopy. At the end of the experiments, the anodes of
the G. sulfurreducens and mixed culture MECs were examined to compare
biofilm morphology differences by using a scanning electron microscope (SEM).
The electrode samples were fixed with 1.5% glutaraldehyde for 1 h and then 1%
osmium tetroxide for 30 min. After each fixation step, the samples were washed
in 0.1 M cacodylate buffer three times. The fixed samples were dehydrated with
a graded series of ethanol and dried using a critical point drying process in liquid
CO2 (BAL-TEC CPD030; Bal-Tec). Samples were then sputter coated with
10-nm Au/Pd (BAL-TEC SCD050; Bal-Tec) and examined using a scanning
electron microscope (JSM 5400; JEOL, Ltd.) at an accelerating voltage of 20 kV.
Community analysis. Community and purity analysis was performed at the end
of the experiments for the mixed culture and Geobacter species, respectively, and
were compared to the cultures used for inoculation. The anodes were placed in
sterile MEC medium lacking acetate and vortexed with glass beads to suspend
the cells in solution. DNA was extracted by using a PowerSoil DNA isolation kit
(MO BIO Laboratories) according to the manufacturer’s instructions. For the
Geobacter reactors, a 16S rRNA gene fragment of the extracted DNA was
amplified by PCR using universal bacterial primers 338F (5⬘-ACTCCTACGGG
AGGCAGCAG-3⬘) and 518R (5⬘-ATTACCGCGGCTGCTGG-3⬘), which flank
the V3 region (22). For the mixed community, a 16S rRNA gene fragment of the
extracted DNA was amplified by PCR using universal bacterial primers 27F
HYDROGEN PRODUCTION BY GEOBACTER SPECIES
7581
FIG. 2. Current densities (A), anode potentials (B), and gas production (C) recorded at EAP ⫽ 0.7 V for G. sulfurreducens (GS), mixed
culture (MC), G. metallireducens (GM), and the control (CNTRL; no
inoculum).
(5⬘-AGAGTTTGATCCTGGCTCAG-3⬘) and 1541R (5⬘-AAGGAGGTGATCC
AGCC-3⬘) (35). PCR products were purified by using QIAquick PCR purification kit (Qiagen) and then ligated and cloned by using a TOPO TA cloning kit
(Invitrogen) according to the manufacturer’s instructions. Plasmids were extracted by using the Ultraclean 6-min Plasmid Prep kit (MoBio) according to the
manufacturer’s instructions. Eight plasmids from each Geobacter reactor and
forty-six plasmids from each mixed reactor were sequenced with the T7 primer
on the TOPO plasmid by using an ABI 3730XL DNA sequencer (Applied
Biosystems). The nucleotide collection (nr/nt) of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/BLAST/) was searched by
using the BLASTn algorithm to analyze the sequences. MEGA 4.0.2 (31) was
used to align these sequences and generate a phylogenetic tree using the neighbor-joining method with a bootstrap test (500 replicates) of phylogeny. Sequences derived from the analysis of the mixed reactors were deposited in
GenBank under accession numbers GQ152894 to GQ152986.
RESULTS
MEC performance at EAP ⴝ 0.7 V. Each MEC was operated
in batch mode at EAP ⫽ 0.7 V until stable current generation
was observed. During the first batch, G. sulfurreducens reached
a peak in current production after 1.5 days, compared to 3 days
for the mixed culture and about 8.5 days for G. metallireducens
(Fig. 2A). In the first batch, both G. sulfurreducens and the
mixed culture generated similar peak current densities (IV) of
80 ⫾ 1 A/m3 and 83 ⫾ 4 A/m3 respectively, whereas G. metallireducens produced 31 ⫾ 0 A/m3. After G. metallireducens
reached a peak current density of 120 ⫾ 1 A/m3 (batch 3),
7582
CALL ET AL.
current generation decreased with each subsequent batch,
reaching 106 ⫾ 2 A/m3 by the final batch (data not shown). By
the final batch at EAP ⫽ 0.7 V, G. sulfurreducens generated
160 ⫾ 2 A/m3 compared to 166 ⫾ 0 A/m3 for the mixed culture.
The EAN values for each reactor at EAP ⫽ 0.7 V showed a
sharp decrease in the overpotential for the initial batches
(batches 1 to 3) and then were more constant in subsequent
batches (Fig. 2B). G. sulfurreducens and the mixed culture had
similar anode potentials for each batch, and both reached a
potential of ⫺0.29 ⫾ 0.00 V by the final batch. G. metallireducens operated at a higher anodic overpotential, with a recorded
potential of ⫺0.23 ⫾ 0.00 V by the final batch. For comparison,
the theoretical anode potential under the experimental conditions (pH 6.8, 12 mM NaH3C2O2, 30 mM HCO3⫺, T ⫽ 30°C)
was ⫺0.48 V. Thus, G. sulfurreducens and the mixed culture
had ⫺0.19-V anodic overpotentials, whereas G. metallireducens was around ⫺0.25 V.
G. sulfurreducens and the mixed culture exhibited reproducible gas production (VG) after the first few batches (Fig. 2C),
whereas G. metallireducens reached a peak in volume and then
produced decreasing volumes with each subsequent batch
(data not shown). Both G. sulfurreducens and the mixed culture
produced similar volumes of gas after the first few batches,
with an average of 24 ⫾ 1 ml for G. sulfurreducens (batches 3
to 10) and 26 ⫾ 2 ml (batches 4 to 12) for the mixed culture.
G. metallireducens reached a peak volume of 22 ml at the end
of batch 6, but between batches 7 and 10, G. metallireducens
generated smaller quantities of gas with each subsequent
batch. On average, G. metallireducens produced 21 ⫾ 2 ml.
MEC performance at EAP ⴝ 0.4 V. When the applied voltage
was lowered to EAP ⫽ 0.4 V, the current densities decreased as
expected (Fig. 3A). G. sulfurreducens generated an average of
57 ⫾ 2 A/m3, which was ca. 21% higher than the current
density produced by the mixed culture (47 ⫾ 1 A/m3) and
roughly 84% larger than that of G. metallireducens (31 ⫾ 5
A/m3). Both G. sulfurreducens and the mixed culture exhibited
consistent and reproducible batches of current generation,
whereas G. metallireducens produced a slightly higher average
maximum current density with each batch (batch 1, 22 ⫾ 0
A/m3; batch 2, 27 ⫾ 0 A/m3; batch 3, 35 ⫾ 5 A/m3).
The anode potentials followed a trend similar to that observed at EAP ⫽ 0.7 V, with a slight decrease in the overpotential with each consecutive batch (Fig. 3B). By the fourth
batch the anode potential for G. sulfurreducens was ⫺0.40 ⫾
0.00 V. The mixed culture had a more negative potential of
⫺0.42 ⫾ 0.00 V, indicating that G. sulfurreducens had a slightly
larger overpotential than the mixed culture. At this higher
overpotential, however, G. sulfurreducens was capable of generating a higher current density (Fig. 3A). G. metallireducens
exhibited the lowest anode potential during the first batch
(EAN ⫽ ⫺0.39 ⫾ 0.00 V) but showed a sharp reduction in the
overpotential by the second batch. Combining this reduction in
overpotential with the increase in current density suggests that
the lower applied voltage was more favorable for current generation by G. metallireducens than the higher applied voltage.
Improvements in gas production with each batch at EAP ⫽
0.4 V were also noted for G. metallireducens (Fig. 3C). In the
first batch, G. metallireducens produced 11 ml of gas, but by the
end of the second and third batches, it produced 19 and 22 ml,
respectively. In contrast, the mixed culture showed a decrease
APPL. ENVIRON. MICROBIOL.
FIG. 3. Current densities (A), anode potentials (B), and gas production (C) recorded at EAP ⫽ 0.4 V for G. sulfurreducens (GS), a
mixed culture (MC), G. metallireducens (GM), and the control
(CNTRL; no inoculum).
in the volume of gas collected with each batch, with 17 ml
produced at the end of the first batch and only 11 ml produced
after the last. This decrease was likely due to methanogenic
consumption of the produced hydrogen, as evidenced by an
increase in methane concentration from 11 to 54% for the first
and final batches, respectively. G. sulfurreducens generated reproducible volumes of gas for each batch, with an average of
23 ⫾ 1 ml for all batches.
Hydrogen recoveries and production rates. The performance of each reactor in terms of recoveries, yields, and production rates, depended on the applied voltage and for the
mixed culture, hydrogen losses to methanogens (Table 1). Before methane was detected in the mixed culture in experiments
at EAP ⫽ 0.7 V, coulombic efficiencies averaged CE ⫽ 89% ⫾
4%. After methane appeared, the coulombic efficiency (CE ⫽
93% ⫾ 2%) did not increase significantly (Student t test, P ⬎
0.09). The coulombic efficiencies of both G. sulfurreducens
(86% ⫾ 2%) and G. metallireducens (81% ⫾ 5%) were lower
than those obtained for the mixed culture.
Cathodic hydrogen recoveries (rCAT) in all of the MECs
were above 100% at EAP ⫽ 0.7 V, indicating that more hydrogen was recovered than expected based on the current. Values
greater than 100% have similarly been reported by others (17,
27). At EAP ⫽ 0.4 V, the cathodic hydrogen recovery of the
mixed culture dropped to 43% ⫾ 19% due to methanogenic
VOL. 75, 2009
HYDROGEN PRODUCTION BY GEOBACTER SPECIES
7583
TABLE 1. Coulombic efficiencies, hydrogen recoveries, energy recoveries, acetate removal, and hydrogen production rates for each
MEC at EAP ⫽ 0.7 V and 0.4 Va
Reactor and
culture typeb
CE (%)
rCAT (%)
rH2 (%)
EAP ⫽ 0.7 V
G. sulfurreducens
G. metallireducens
Mixed culture 1
Mixed culture 2
Mixed culture 3
86
81
89
93
92
104
109
108
105
74
EAP ⫽ 0.4 V
G. sulfurreducens
G. metallireducens
Mixed culture
89
82
89
91
90
43
E (%)
E⫹S (%)
⌬H
⌬G
⌬H
⌬G
Acetate
removal (%)
QV
(m3 H2/m3/day)
89
88
96
97
68
228
236
236
230
163
189
196
195
191
135
77
78
82
82
58
66
66
70
70
49
93
84
90
90
90
1.9
1.3
1.8
1.9
1.3
80
74
38
344
338
164
285
235
136
80
76
38
61
53
33
96
96
94
0.6
0.3
0.2
a
⌬H, based on heat of combustion; ⌬G, based on Gibbs free energy.
Mixed culture 1, before the appearance of methane in the collected gas; mixed culture 2, after methane appeared; mixed culture 3, after operating several batches
at EAP ⫽ 0.4 V.
b
conversion of hydrogen to methane. The longer reaction times
at EAP ⫽ 0.4 V of 2 to 3 days, compared to ⬃1 day at EAP ⫽
0.7 V, were presumably more favorable for methane production by slow-growing methanogens. When a final batch was
operated at EAP ⫽ 0.7 V following the batches at EAP ⫽ 0.4 V,
the cathodic hydrogen recovery dropped to 74%, compared to
105% ⫾ 3% originally obtained at this applied voltage, demonstrating the impact of methanogen growth on hydrogen recovery.
Overall, hydrogen recoveries (E⫹S) were highest for the
mixed culture (97% ⫾ 1%) when methane concentrations remained below 1% at EAP ⫽ 0.7 V. After the methane concentrations increased, the hydrogen recovery (rH2) decreased to
38% ⫾ 16% at EAP ⫽ 0.4 V. Both G. sulfurreducens and G.
metallireducens exhibited lower hydrogen recoveries of 89% ⫾
4% and 88% ⫾ 1%, respectively, at EAP ⫽ 0.7 V because of
their lower coulombic efficiencies. The similarity of these recoveries for the two strains indicates that the ability of G.
sulfurreducens to consume hydrogen did not have a significant
impact on hydrogen recovery.
G. sulfurreducens and the mixed culture produced hydrogen at
similar production rates (QV) of ⬃1.9 m3 H2/m3/day at EAP ⫽ 0.7
V, as expected from comparable current densities and cathodic
hydrogen recoveries. When the cathodic hydrogen recovery
dropped for the mixed culture due to hydrogen losses to methanogens, however, the production rate subsequently dropped
to 1.3 m3 H2/m3/day, even though the current density was
similar to that obtained at EAP ⫽ 0.7 V in the initial batches.
G. metallireducens obtained the lowest current density and
subsequently had the lowest production rate of 1.3 ⫾ 0.1 m3
H2/m3/day.
Internal hydrogen recycling. To examine the potential for
hydrogen losses and coulombic efficiency increases due to internal hydrogen recycling, a hydrogen utilization test was performed using each MEC at EAP ⫽ 0.7 V (Fig. 4). During this
test a gas mixture of 80% hydrogen and 20% carbon dioxide
was constantly sparged through MECs containing medium
lacking acetate. After an initial drop in current density during
the first 2 h, G. sulfurreducens showed a slight increase in
current production from 64 ⫾ 2 A/m3 to 66 ⫾ 1 A/m3, followed
by a steady decrease. The mixed culture also showed an initial
drop in current to 27 ⫾ 1 A/m3 with a slight increase to 31 ⫾
0 A/m3, followed by steady current production for the remainder of the test. G. metallireducens, which does not utilize H2,
initially produced current, likely as a result of stored substrate.
However, there was a sharp decrease in current production
toward the zero current value of the control at the end of the
test, indicating that hydrogen did not sustain current generation by this bacterium.
Energy recoveries. Relative to only the electrical energy
input, all MECs at both applied voltages achieved efficiencies
over 100% (Table 1). At EAP ⫽ 0.7 V, each MEC obtained
roughly the same energy efficiency (E ⫽ 230% versus ⌬H; see
Table 1 for efficiencies based on ⌬G), but the mixed culture
MEC showed a sharp drop to E ⫽ 163% after methane
concentrations increased to 9%. At EAP ⫽ 0.4 V, G. sulfurreducens obtained the highest electrical energy recovery of E ⫽
344% ⫾ 12%, while G. metallireducens and the mixed culture
reached E ⫽ 338% ⫾ 15% and E ⫽ 164% ⫾ 72%, respectively. The electrical energy recoveries with the mixed culture
were significantly different from those of the two Geobacter
species (Student t test, P ⬍ 0.02), but there was no significant
difference between the two Geobacter species (Student t test,
P ⬎ 0.58). When the energy of the substrate was included as an
energy input, the mixed culture obtained the highest value of
FIG. 4. Current densities recorded during hydrogen utilization
tests (no acetate provided) at an applied voltage of EAP ⫽ 0.7 V for G.
sulfurreducens (GS), a mixed culture (MC), G. metallireducens (GM),
and the control (CNTRL; no inoculum).
7584
CALL ET AL.
FIG. 5. CV recorded at 10 mV/s from ⫺0.8 V to ⫹ 0.2 V for G.
sulfurreducens (GS), a mixed culture (MC), and the control (CNTRL;
no inoculum).
E⫹S ⫽ 82% ⫾ 1% at EAP ⫽ 0.7 V, but this was appreciably
reduced to E⫹S ⫽ 38% ⫾ 16% at EAP ⫽ 0.4 V. Both
Geobacter species had slightly lower overall energy recoveries
at EAP ⫽ 0.7 V, with G. sulfurreducens and G. metallireducens
reaching E⫹S ⫽ 77% ⫾ 3% and E⫹S ⫽ 78% ⫾ 5%, respectively.
Including the energy of the methane produced in the mixed
culture MEC did not substantially increase overall energy recoveries. At EAP ⫽ 0.4 V, the overall energy recovery was 10%
larger (E⫹S ⫽ 42% ⫾ 14%) with methane and hydrogen than
with just hydrogen (E⫹S ⫽ 38% ⫾ 16%). In the final batch at
EAP ⫽ 0.7 V, the overall energy recovery increased by only 3%
from E⫹S ⫽ 58% to E⫹S ⫽ 60% when methane was accounted for in the energy balance.
Cyclic voltammograms. Since no distinct differences were
evident between the current densities of G. sulfurreducens and
the mixed culture at both applied voltages, CV was performed
to examine the electrode reducing rates at a wider range of
potentials (Fig. 5). CV analyses were conducted after 35 days
of operation for G. sulfurreducens and after 26 days for the
mixed culture. CV analysis of both G. sulfurreducens and the
mixed culture yielded a sigmoidal shape characteristic of catalytic activity and consistent with previous G. sulfurreducens
CV studies (21, 30). During the forward scan, G. sulfurreducens
began producing positive current at E ⫽ ⫺0.55 V, while the
mixed culture initially operated at a higher overpotential at
around E ⫽ ⫺0.47 V. Between E ⫽ ⫺0.55 V and ⫺0.16 V, G.
sulfurreducens operated at lower anodic overpotentials than
the mixed culture, but above E ⫽ ⫺0.16 V the mixed culture
produced higher current. The mixed culture also produced a
higher peak current of 5.9 ⫾ 0.0 mA versus 4.6 ⫾ 0.0 mA for
G. sulfurreducens.
G. sulfurreducens and mixed culture current densities for 0.3
V < EAP < 1.0 V. Since CV analysis indicated that the mixed
consortium was capable of higher current densities at higher
applied voltages, the mixed culture MEC and G. sulfurreducens
were operated at applied voltages from 0.3 V to 1.0 V (Fig. 6).
For EAP ⱖ 0.7 V, G. sulfurreducens generated on average a 2%
higher current density, but for EAP ⱕ 0.6 V, G. sulfurreducens
produced on average a 30% larger current density. For all
applied voltages, the mixed culture current density increased as
a factor of 354 ⫾ 9 A/m3 per volt, whereas the current density
of G. sulfurreducens was slightly lower at 343 ⫾ 2 A/m3 per volt.
At the highest applied voltage (EAP ⫽ 1.0 V), G. sulfurreducens
APPL. ENVIRON. MICROBIOL.
FIG. 6. Current densities recorded as a function of applied voltage
for G. sulfurreducens (GS) and a mixed culture (MC).
generated 263 ⫾ 3 A/m3 (6.8 ⫾ 0.1 A/m2) and the mixed
culture produced 260 ⫾ 24 A/m3 (6.7 ⫾ 0.6 A/m2). At the
lowest applied voltage (EAP ⫽ 0.3 V), G. sulfurreducens produced 25 ⫾ 0 A/m3 (0.6 ⫾ 0.0 A/m2) and the mixed culture
reached 15 ⫾ 0 A/m3 (0.4 ⫾ 0.0 A/m2).
SEM images. SEM imaging revealed different biofilm morphologies of the mixed culture and G. sulfurreducens (Fig. 7).
G. sulfurreducens formed layers of cells stacked on top of each
other and the graphite fibers, whereas the mixed culture
formed tight structures that wrapped around individual graphite fibers. Higher magnification indicated that the mixed culture covered a larger surface area of the graphite fibers compared to G. sulfurreducens which appeared to form looser
biofilms around the fibers.
Community analysis and purity check. The mixed culture
from the acclimated MFCs used as the MEC inoculum contained primarily Firmicutes (26% of clones), Proteobacteria
(30%), and Bacteroides (38%) (Table 2). Actinobacteria was
also present, representing 4.3% of the community, whereas
only one G. sulfurreducens clone (2.1% of the total) was detected. The community profile of the MEC anode measured by
16S rRNA gene cloning was determined. In a BLAST search
result (nearest match) for G. sulfurreducens PCA, there were
33 clones with ⬎97% identity, and for the uncultured bacterium clone e03-d04 there were 2 clones with ⬎97% identity. In
this analysis the following were represented by one clone: Acidovorax sp. strain R-24607, Arcobacter butzleri RM4018,
Pseudomonas sp. strain ZZ5, uncultured Arcobacter sp. strain
clone DS031, uncultured bacterium clone HB99 (96% identity), uncultured bacterium clone HDBW-WB48, uncultured
bacterium clone LTR-R12 (96% identity), uncultured bacterium clone M3B06, uncultured bacterium clone R1B-24, uncultured Deferribacter sp. clone 25IIISN (92% identity), and
Wolinella succinogenes strain ATCC 29543. After the MEC
experiments, the anode community was dominated by G. sulfurreducens (72%). Proteobacteria (including G. sulfurreducens)
composed 87% of the overall population, with 6.5% represented by Bacteroides, 4.3% by Deferribacteres, and 2.2% by
Firmicutes. Actinobacteria was not present in the mixed-culture
MECs.
Sequence analysis from both the G. sulfurreducens and the
G. metallireducens reactors confirmed (98 to 100% identity)
that these reactors remained uncontaminated throughout the
experiments (data not shown).
VOL. 75, 2009
HYDROGEN PRODUCTION BY GEOBACTER SPECIES
7585
FIG. 7. SEM images of the control (A and D), G. sulfurreducens (B and E), and a mixed culture (C and F) anodes at the end of all batches.
Images were taken after day 63 for G. sulfurreducens and day 47 for the mixed culture. The bars in panels A to C represent 100 m. The bars in
panels D to F represent 10 m.
DISCUSSION
These results show that hydrogen losses to methane generation can be avoided in MECs by using pure cultures and that
hydrogen production rates and yields similar to those of mixed
cultures can be obtained with G. sulfurreducens. Although G.
metallireducens is exoelectrogenic, current densities produced
using this bacterium were consistently lower than those of the
mixed culture or G. sulfurreducens.
Evidence for internal hydrogen recycling was confirmed for
both G. sulfurreducens and the mixed culture based on hydrogen utilization tests in the absence of acetate (Fig. 4) and
coulombic efficiencies measured in the presence of acetate.
Some current was always initially produced by enriched anodes
(likely from endogenous respiration), but the magnitude and
stability of these currents varied in the tests. For example, ca.
24 A/m3 was initially produced by G. metallireducens when the
reactor was sparged with hydrogen gas (no acetate). However,
this current was only ca. 20% of that produced in MEC tests
with acetate, and this strain cannot respire using hydrogen
(19). With G. sulfurreducens under the same conditions (hydrogen gas, no acetate), the current density was higher (ca. 66
A/m3), with values initially ca. 40% of that produced in MEC
tests with acetate. This suggests that there was appreciable
hydrogen utilization by G. sulfurreducens in MEC tests that
resulted in current generation. In addition, the coulombic efficiency for G. sulfurreducens (CE ⫽ 86%) was higher than that
of G. metallireducens (CE ⫽ 81%) in acetate-fed MEC tests,
suggesting that the coulombic efficiency may have increased
slightly (from 81 to 86%) as a result of hydrogen utilization
(assuming all other factors, such as cell yields, were constant
between the strains). However, a comparison of the coulombic
efficiencies in these MEC tests to those obtained in other MFC
tests indicates that, overall, the coulombic efficiency values
were not increased to unreasonable values by hydrogen utili-
zation. In MFC tests with G. sulfurreducens using a ferricyanide
catholytes (no oxygen), for example, coulombic efficiencies
were much higher than those obtained here, reaching 100%
(23). Thus, while hydrogen cycling was likely for G. sulfurreducens, the relatively low coulombic efficiency values indicate
that hydrogen recycling in MECs was a relatively small contributor to current production.
In the case of the mixed culture, the high coulombic efficiency (CE ⫽ 93%) in MEC tests and sustained current production in hydrogen utilization tests suggested that hydrogen
cycling contributed more substantially to current production.
TABLE 2. Community profile of the MFC anode measured by
16S rRNA gene cloninga
BLAST search result (nearest match)
No. of clones
with ⬎95%
identity
Uncultured bacterium clone R1B-15 ............................................ 6
Bacteroides sp. strain 22C ............................................................... 3
Uncultured bacterium BB23 .......................................................... 3
Uncultured bacterium clone HDBW-WB48 ................................ 3
Uncultured bacterium clone YWB28............................................ 3
Bacterium strain B4C1-5................................................................. 2
Ruminobacillus xylanolyticum.......................................................... 2
Thermomonas hemolytica strain A50-7-3 ...................................... 2
Uncultured bacterium clone R-1167b ........................................... 2
Uncultured Rhodocyclaceae clone MFC-B162-F11 ..................... 2
a
The following were represented by one clone: Acidovorax sp. strain
R-24607, Azoarcus sp. strain BH72, bacterium enrichment culture clone
PA10, Comamonas sp., Geobacter sulfurreducens PCA, Hydrogenophaga sp.
strain AH-24, Methylobacterium sp. strain F3.2, Niabella sp. strain GR10-1,
Peptococcaceae bacterium strain Y5 (90% identity), Rhodococcus sp. strain
T104, uncultured bacterium clone 1As26, uncultured bacterium clone ATBM1328, uncultured bacterium clone B53, uncultured bacterium clone PL-5B10,
uncultured bacterium clone PL-7B6, uncultured bacterium clone Rhag4-33, uncultured “Candidatus Odyssella sp.” clone 1-K.
b
94% identity.
7586
CALL ET AL.
The mixed culture hydrogen utilization tests produced a current density of ca. 30 A/m3 (no acetate). Although this is only
ca. 19% of the maximum current (160 A/m3) generated with
acetate, this current density was sustained for the mixed culture by hydrogen gas, whereas current density always decreased with the pure cultures in hydrogen gas tests. The
observation of sustained current generation in hydrogen utilization tests, combined with a coulombic efficiency value larger
than those obtained for either Geobacter strain, suggests that
hydrogen cycling increased the coulombic efficiency of the
mixed culture. This hydrogen cycling did not affect overall
energy recoveries, however, as similar values were obtained for
the mixed culture and G. sulfurreducens in MEC tests before
the enrichment of methanogens.
Methanogenesis was observed in mixed culture tests and
increased over time. However, the high coulombic efficiency
and the consumption of hydrogen by the mixed culture suggested that the main route for methane generation was hydrogenotrophic and not acetoclastic methanogenesis. This observation is consistent with most other mixed-culture MEC studies
that have shown hydrogenotrophic methanogenesis predominates in MECs (7, 8, 16, 34). It is also unlikely that methane
generation occurred by electromethanogenesis (6), i.e., directly
from electron release at the cathode. Hydrogen production
was catalyzed here using platinum on the cathode, and there
was insufficient time to acclimate a methanogenic biofilm on
the cathode, and therefore electromethanogenesis could not
have substantially contributed to methane production under
these conditions.
The reduced performance of G. metallireducens compared to
G. sulfurreducens in the MEC was not expected based on iron
reducing rates of these two bacteria. G. metallireducens can
reduce iron at a rate of 887 fmol/cell/day with soluble iron, and
94 fmol/cell/day with insoluble iron, compared to 225 fmol/cell/
day (soluble iron) and 23 fmol/cell/day (insoluble iron) for G.
sulfurreducens (1, 20). The magnitude of the applied voltage
may have also been a factor in the reduced performance of G.
metallireducens compared to G. sulfurreducens. The current
density, anode potential, and volume of produced gas all improved with each successive batch at an EAP of 0.4 V compared
to a decrease in each of these metrics at an EAP of 0.7 V.
Erratic performance of MECs with mixed cultures has also
been observed at higher applied voltages (2).
There was also a noticeable difference in biofilm morphology between G. sulfurreducens and the mixed culture despite
the predominance of G. sulfurreducens in the mixed-culture
MEC community. The mixed-culture biofilms were tighter and
appeared more compact on the surface of the graphite fibers
than those of G. sulfurreducens, which appeared less dense and
comprised of more voids between the cells and fibers. Combining this observation with the higher peak current generated
by the mixed culture during CV analysis suggests that the
mixed culture achieved better contact with the electrode surface than G. sulfurreducens. The higher current density of the
mixed culture is consistent with higher per-biomass electrode
reducing rates of a mixed culture compared to G. sulfurreducens observed in MFCs (12).
There was a large change in the community structure between the bacteria detected in the inoculum taken from an
MFC and that present in the MECs. To our knowledge, this is
APPL. ENVIRON. MICROBIOL.
the first time that changes in bacterial communities have been
studied for a culture transferred from an MFC to an MEC.
Although G. sulfurreducens was detected in only 1 of 46 clones
from the MFC inoculum, in the MEC tests, it accounted for
72% of the community based on 16S rRNA sequence analysis.
This indicates that the conditions in the MEC provided a
better competitive growth environment for Geobacter than the
MFC. The reasons for this include both operational conditions
and the medium. In an MFC oxygen can diffuse through the
cathode, but the MECs were operated under completely
anoxic conditions. Most importantly, a carbonate-rich medium
was used for the MECs, but the original MFC reactor was
operated with a medium containing a phosphate buffer. In
MFC experiments by Ishii et al. (12) using an air-cathode MFC
and a carbonate medium, G. sulfurreducens was also found to
be predominant in the anode biofilm community. We have also
found that G. sulfurreducens grows poorly or not at all in a
phosphate buffer medium (unpublished results), and thus the
use of a carbonate medium is critical for growth of this microorganism.
These results show that pure cultures can produce hydrogen
gas in MECs at rates and recoveries comparable to mixed
cultures. Depending on the species, the current densities may
be higher than mixed cultures, but in all cases, methane-free
gas is produced. Investigations of other pure cultures, in particular species that can derive energy and biomass from sources
other than the substrate are needed in order to increase coulombic efficiencies and hydrogen recoveries. Engineering and
selection of strains adapted to electron transfer to anodes may
also lead to an increase in efficiencies and production rates (14,
36). Furthermore, other strains that can efficiently produce
hydrogen in MECs using complex substrates will help create
new applications such as localized hydrogen production from
diverse and abundant waste streams.
ACKNOWLEDGMENTS
We thank D. W. Jones for help with the analytical measurements;
M. L. Hazen for assistance with the SEM imaging; and the Penn State
Genomics Core Facility, University Park, PA, for DNA sequencing.
This research was funded by the American Society of Engineering
Education National Defense Science and Engineering Graduate Fellowship, the National Science Foundation Graduate Research Fellowship Program, a National Water Research Institute Ronald B. Linsky
Fellowship, and award KUS-I1-003-13 from King Abdullah University
of Science and Technology.
REFERENCES
1. Caccavo, F., Jr., D. J. Lonergan, D. R. Lovley, M. Davis, J. F. Stolz, and M. J.
McInerney. 1994. Geobacter sulfurreducens sp. nov., a hydrogen- and acetate-oxidizing dissimilatory metal-reducing microorganism. Appl. Environ.
Microbiol. 60:3752–3759.
2. Call, D., and B. E. Logan. 2008. Hydrogen production in a single chamber
microbial electrolysis cell lacking a membrane. Environ. Sci. Technol. 42:
3401–3406.
3. Call, D. F., M. D. Merrill, and B. E. Logan. 2009. High surface area stainless
steel brushes as cathodes in microbial electrolysis cells. Environ. Sci. Technol. 43:2179–2183.
4. Cheng, S., and B. E. Logan. 2007. Ammonia treatment of carbon cloth
anodes to enhance power generation of microbial fuel cells. Electrochem.
Commun. 9:492–496.
5. Cheng, S., and B. E. Logan. 2007. Sustainable and efficient biohydrogen
production via electrohydrogenesis. Proc. Natl. Acad. Sci. USA 104:18871–
18873.
6. Cheng, S., D. Xing, D. F. Call, and B. E. Logan. 2009. Direct biological
conversion of electrical current into methane by electromethanogenesis.
Environ. Sci. Technol. 43:3953–3958.
VOL. 75, 2009
7. Clauwaert, P., R. Toledo, D. V. D. Ha, R. Crab, W. Verstraete, H. Hu, K. M.
Udert, and K. Rabaey. 2008. Combining biocatalyzed electrolysis with anaerobic digestion. Water Sci. Technol. 57:575–579.
8. Clauwaert, P., and W. Verstraete. 2009. Methanogenesis in membraneless
microbial electrolysis cells. Appl. Environ. Microbiol. 82:829–836.
9. Fan, Y., H. Hu, and H. Liu. 2007. Sustainable power generation in microbial
fuel cells using bicarbonate buffer and proton transfer mechanisms. Environ.
Sci. Technol. 41:8154–8158.
10. Hawkes, F. R., I. Hussy, G. Kyazze, R. Dinsdale, and D. L. Hawkes. 2007.
Continuous dark fermentative hydrogen production by mesophilic microflora: principles and progress. Int. J. Hydrogen Energy 32:172–184.
11. Hu, H., Y. Fan, and H. Liu. 2008. Hydrogen production using single-chamber
membrane-free microbial electrolysis cells. Water Res. 42:4172–4178.
12. Ishii, S., K. Watanabe, S. Yabuki, B. E. Logan, and Y. Sekiguchi. 2008.
Comparison of electrode reduction activities of Geobacter sulfurreducens and
an enriched consortium in an air-cathode microbial fuel cell. Appl. Environ.
Microbiol. 74:7348–7355.
13. Ivy, J. 2004. Summary of electrolytic hydrogen production: milestone completion report. NREL/MP-560-36734. NREL, Golden, CO.
14. Izallalen, M., R. Mahadevan, A. Burgard, B. Postier, R. Didonato, Jr., J.
Sun, C. H. Schilling, and D. R. Lovley. 2008. Geobacter sulfurreducens strain
engineered for increased rates of respiration. Metab. Eng. 10:267–275.
15. Lanthier, M., K. B. Gregory, and D. R. Lovley. 2008. Growth with high
planktonic biomass in Shewanella oneidensis fuel cells. FEMS Microbiol.
Lett. 278:29–35.
16. Lee, H.-S., C. I. Torres, P. Parameswaran, and B. E. Rittmann. 2009. Fate of
H2 in an upflow single-chamber microbial electrolysis cell using a metalcatalyst-free cathode. Environ. Sci. Technol. 43:7971–7976.
17. Liu, H., S. Grot, and B. E. Logan. 2005. Electrochemically assisted microbial
production of hydrogen from acetate. Environ. Sci. Technol. 39:4317–4320.
18. Logan, B. E., D. Call, S. Cheng, H. V. M. Hamelers, T. H. J. A. Sleutels, A. W.
Jeremiasse, and R. A. Rozendal. 2008. Microbial electrolysis cells for high
yield hydrogen gas production from organic matter. Environ. Sci. Technol.
42:8630–8640.
19. Lovley, D. R., S. J. Giovannoni, D. C. White, J. E. Champine, E. J. P.
Phillips, Y. A. Gorby, and S. Goodwin. 1993. Geobacter metallireducens gen.
nov. sp. nov., a microorganism capable of coupling the complete oxidation of
organic compounds to the reduction of iron and other metals. Arch. Microbiol. 159:336–344.
20. Lovley, D. R., and E. J. P. Phillips. 1988. Novel mode of microbial energy
metabolism: organic carbon oxidation coupled to dissimilatory reduction of
iron or manganese. Appl. Environ. Microbiol. 54:1472–1480.
21. Marsili, E., J. B. Rollefson, D. B. Baron, R. M. Hozalski, and D. R. Bond.
2008. Microbial biofilm voltammetry: direct electrochemical characterization
of catalytic electrode-attached biofilms. Appl. Environ. Microbiol. 74:7329–
7337.
22. Muyzer, G., E. C. de Waal, and A. G. Uitterlinden. 1993. Profiling of complex
HYDROGEN PRODUCTION BY GEOBACTER SPECIES
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
7587
microbial populations by denaturing gradient gel electrophoresis analysis of
polymerase chain reaction-amplified genes coding for 16S rRNA. Appl.
Environ. Microbiol. 59:695–700.
Nevin, K. P., H. Richter, S. F. Covalla, J. P. Johnson, T. L. Woodard, A. L.
Orloff, H. Jia, M. Zhang, and D. R. Lovley. 2008. Power output and columbic
efficiencies from biofilms of Geobacter sulfurreducens comparable to mixed
community microbial fuel cells. Environ. Microb. 10:2505–2514.
Ringeisen, B. R., E. Henderson, P. K. Wu, J. Pietron, R. Ray, B. Little, J. C.
Biffinger, and J. M. Jones-Meehan. 2006. High power density from a miniature microbial fuel cell using Shewanella oneidensis DSP10. Environ. Sci.
Technol. 40:2629–2634.
Ringeisen, B. R., R. Ray, and B. Little. 2007. A miniature microbial fuel cell
operating with an aerobic anode chamber. J. Power Sources 165:591–597.
Rozendal, R. A., H. V. M. Hamelers, G. J. W. Euverink, S. J. Metz, and
C. J. N. Buisman. 2006. Principle and perspectives of hydrogen production
through biocatalyzed electrolysis. Int. J. Hydrogen Energy 31:1632–1640.
Rozendal, R. A., H. V. M. Hamelers, R. J. Molenkamp, and C. J. N. Buisman.
2007. Performance of single chamber biocatalyzed electrolysis with different
types of ion exchange membranes. Water Res. 41:1984–1994.
Rozendal, R. A., A. W. Jeremiasse, H. V. M. Hamelers, and C. J. N. Buisman.
2008. Hydrogen production with a microbial biocathode. Environ. Sci. Technol. 42:629–634.
Selembo, P. A., M. D. Merrill, and B. E. Logan. 2009. The use of stainless
steel and nickel alloys as low-cost cathodes in microbial electrolysis cells. J.
Power Sources 190:271–278.
Srikanth, S., E. Marsili, M. C. Flickinger, and D. R. Bond. 2008. Electrochemical characterization of Geobacter sulfurreducens cells immobilized on
graphite paper electrodes. Biotechnol. Bioeng. 99:1065–1073.
Tamura, K., J. Dudley, M. Nei, and S. Kumar. 2007. MEGA4: molecular
evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol.
24:1596–1599.
Tartakovsky, B., M. F. Manuel, H. Wang, and S. R. Guiot. 2009. High rate
membrane-less microbial electrolysis cell for continuous hydrogen production. Int. J. Hydrogen Energy 34:672–677.
Wagner, R. C., J. M. Regan, S.-E. Oh, Y. Zuo, and B. E. Logan. 2009.
Hydrogen and methane production from swine wastewater using microbial
electrolysis cells. Water Res. 43:1480–1488.
Wang, A., W. Liu, S. Cheng, D. Xing, J. Zhou, and B. E. Logan. 2009. Source
of methane and methods to control its formation in single chamber microbial
electrolysis cells. Int. J. Hydrogen Energy 34:3653–3658.
Winker, S., and C. R. Woese. 1991. A definition of the domains Archaea,
Bacteria, and Eucarya in terms of small subunit rRNA characteristics. Syst.
Appl. Microbiol. 14:305–310.
Yi, H., K. P. Nevin, B.-C. Kim, A. E. Franks, A. Klimes, L. M. Tender, and
D. R. Lovley. 2009. Selection of a variant of Geobacter sulfurreducens with
enhanced capacity for current production in microbial fuel cells. Biosens.
Bioelectron. 24:3498–3503.