Electricity Generation by Geobacter sulfurreducens Attached to Gold
Hanno Richter,*,†Kevin McCarthy,*,‡Kelly P. Nevin,†Jessica P. Johnson,†
Vincent M. Rotello,*,†and Derek R. Lovley†
Departments of Microbiology and Chemistry, UniVersity of Massachusetts, Amherst, Massachusetts 01003
ReceiVed NoVember 6, 2007. In Final Form: January 8, 2008
The versatility of gold for electrode manufacture suggests that it could be an ideal material for some microbial fuel
cell applications. However, previous studies have suggested that microorganisms that readily transfer electrons to
graphite do not transfer electrons to gold. Investigations with Geobacter sulfurreducens demonstrated that it could
with the development of G. sulfurreducens biofilms up to 40 µm thick. No current was produced if pilA, the gene
for the structural protein of the conductive pili of G. sulfurreducens, was deleted. The finding that gold is a suitable
anode material for microbial fuel cells offers expanded possibilities for the construction of microbial fuel cells and
the electrochemical analysis of microbe-electrode interactions.
Microorganisms that produce electricity by oxidizing organic
compounds with electron transfer to electrodes may be useful
agents for current generation from waste organic matter and
renewable biomass, as well as for sensors.1-4Graphite has
be preferable, either because they enhance electron transfer
between the microorganisms and the anode material or because
they are better adapted to specific applications. For example,
incorporation of manganese, iron, quinones, or neutral red in
Gold is a potentially attractive anode material for some
microbial fuel cell applications because it is highly conductive
manufacture. However, previous studies with Shewanella pu-
trefaciens suggested that bare gold is a poor electrode material
for the anode of microbial fuel cells. Current production with
gold electrodes was low and increased 100-fold when the gold
surface was coated with a surface-associated monolayer (SAM)
of 11-mercapto-undecanoic acid,7even though the SAM would
that the gold surface was either toxic to the cells or otherwise
poorly suited to interact with electron-transfer cell components.
capabilities.9Graphite contains functional groups, such as
electron acceptor for anaerobic respiration in sedimentary
environments.10,11Although gold is highly conductive, it does
not contain such functional groups which conceivably are
important in the interaction between electron transport compo-
nents and electrodes.
of microorganisms that might be employed in microbial fuel
cells or in biosensors based on microbe-electrode interactions.
over Shewanella species. Shewanella species only incompletely
oxidize a limited range of organic acids to acetate, which is
fuel remain in the acetate.3,12In contrast, Geobacter species can
completely oxidize organic compounds to carbon dioxide with
recovery of >90% of the electrons available in the fuels as
electricity.13-15Shewanella species appear to transfer electrons
the anode surface and transfer electrons to the anode via one or
more redox active proteins.14,16,17Geobacter species, or close
relatives, are the primary organisms that colonize the surface of
anodes harvesting electricity from aquatic sediments13,18,19and
* Towhomcorrespondenceshouldbeaddressed.Phone: (413)577-4669
(H.R.); (413) 545-2439 (V.M.R.). Fax: (413) 577-4660 (H.R.). E-mail:
†Department of Microbiology.
‡Department of Chemistry.
(1) Shukla, A. K.; Suresh, P.; Berchmans, S.; Rahjendran, A. Curr. Science
2004, 87, 455-468.
(2) Rabaey, K.; Verstraete, W. Trends Biotechnol. 2005, 23, 291-8.
(3) Lovley, D. R. Nat. ReV. Microbiol. 2006, 4, 497-508.
(4) Seop, C. I.; Moon, H.; Bretschger, O.; Jang, J. K.; Park, H. I.; Nealson,
K. H.; Kim, B. H. J. Microbiol. Biotechnol. 2006, 16, 163-177.
(5) Lowy, D. A.; Tender, L. M.; Zeikus, J. G.; Park, D. H.; Lovley, D. R.
Biosens. Bioelectron. 2006, 21, 2058-63.
(6) Park, D. H.; Zeikus, J. G. Appl. Microbiol. Biotechnol. 2002, 59, 58-61.
(7) Crittenden, S. R.; Sund, C. J.; Sumner, J. J. Langmuir 2006, 22, 9473-6.
(8) Boldt, F. M.; Baltes, N.; Borgwarth, K.; Heinze, J. Surf. Sci. 2005, 597,
(9) Chen, X.; Ferrigno, R.; Yang, J.; Whitesides, G. M. Langmuir 2002, 18,
(10) Lovley, D. R.; Coates, J. D.; Blunt-Harris, E. L.; Phillips, E. J. P.;
Woodward, J. C. Nature (Lett.) 1996, 382.
(11) Lovley, D. R.; Holmes, D. E.; Nevin, K. P. AdV. Microb. Physiol. 2004,
(12) Lanthier,M.;Gregory,K.B.;Lovley,D.R. FEMSMicrobiol.Lett.2008,
(13) Bond, D. R.; Holmes, D. E.; Tender, L. M.; Lovley, D. R. Science 2002,
(15) Nevin, K. P.; Covalla, S. F.; Johnson, J. P.; Woodard, T. L.; Jia, H.;
Zhang, M.; Lovley, D. R. EnViron. Microbiol. 2008, submitted for publication.
(16) Reguera, G.; Nevin, K. P.; Nicoll, J. S.; Covalla, S. F.; Woodard, T. L.;
Lovley, D. R. Appl. EnViron. Microbiol. 2006, 72, 7345-8.
(17) Holmes, D. E.; Chaudhuri, S. K.; Nevin, K. P.; Mehta, T.; Methe, B. A.;
2006, 8, 1805-15.
(18) Holmes, D. E.; Bond, D. R.; O’Neil, R. A.; Reimers, C. E.; Tender, L.
R.; Lovley, D. R. Microb. Ecol. 2004, 48, 178-90.
(19) Tender, L. M.; Reimers, C. E.; Stecher, H. A., 3rd; Holmes, D. E.; Bond,
D. R.; Lowy, D. A.; Pilobello, K.; Fertig, S. J.; Lovley, D. R. Nat. Biotechnol.
2002, 20, 821-5.
10.1021/la703469y CCC: $40.75© xxxx American Chemical Society
Published on Web 02/28/2008PAGE EST: 4
swine waste.20Studies with the most highly studied Geobacter
a graphite anode, it can produce current densities that are the
could use gold as an anode. The results demonstrate that G.
sulfurreducens interacts electrochemically with gold almost as
effectively as with graphite.
Gold Electrodes. Gold electrodes were manufactured using the
template stripping method,21resulting in ultraflat gold surfaces.
Silicon wafers (diameter 4 in., thickness 500-550 µm, resistivity:
1-10 Ωcm, crystal orientation: <1 0 0>, type: P, dopant: boron)
were purchased from Silicon Quest International, Santa Clara, CA.
Glass slides (4 × 4 × 0.2 cm3) were obtained from the local glass
shop and EPO-TEK 377 from Epoxy Technologies, MA. Gold was
of 100 nm, at a rate of 0.2 nm/min.
electrode separated by a Nafion 117 membrane were constructed
from commerically available methanol fuel cells as described
previously,15with the following changes: the anode/working
at a potential of +300 mV vs a Ag/AgCl reference electrode, which
was inserted into the working electrode chamber. The counter
electrode was a piece of graphite cloth (type GC-14, Electrolytica,
Inc., Amherst, NY), 2.4 × 2.4 cm2, supported by a stainless steel
each ministack chamber was flushed with 1 L of sterile deionized
water for over 30 min to remove the ethanol.
Organisms, Media, and Growth Conditions. Geobacter sul-
furreducens, strain PCA (ATCC 51573)22and the mutant in which
and cultured with 10 mM acetate as the electron donor and 40 mM
fumarate as the electron acceptor in pressure tubes under strict
anaerobic conditions, as previously described.24Growth conditions
exceptions: freshwater medium25contained 0.06 g/L NaH2PO4×
H2O. During recirculation mode, the dilution rate was 8.6 h-1and
the initial fumarate concentration was 10 mM. After a switch to
flowthrough mode, when current was produced and the working
electrode chamber was continuously supplied with medium contain-
ing 10 mM acetate, but no fumarate, the dilution rate was 0.86 h-1.
dried. Protein was quantified using the bicinchoninic acid assay
(Sigma) as previously described.15Current was measured with a
Power Lab 8SP unit connected to a Macintosh Powerbook G4, and
in phosphate buffer (50 mM, pH 7.2) with 2% glutaraldehyde and
0.5 mM sodium azide, cut into pieces (1 × 1 cm2) under Pi-buffer
without glutaraldehyde, and rinsed for 5 min in glutaraldehyde-free
phosphate buffer. Samples were incubated for 30 min in phosphate
buffer with 1% osmiumtetroxide, and rinsed in phosphate buffer,
then in deionized water to remove salts. To dehydrate, the electrode
70%, 80%, 95%, 100%, 100%, and 100%), 5 min each; the last
(4A, 8-12 mesh, JT Baker,). The sample was CO2-critical point
dried from the ethanol transitional solvent. Electrode pieces were
mounted on aluminum specimen mounts with Duco cement. The
gold layer was electrically connected to the stage using colloidal
was sputter-coated onto the samples in a Polaron E-5100 sputter
coater (3 min at 2.2 kV, 5 mA) with argon at 0.06 Torr, using a
5400 scanning electron microscope (SEM) operated at 5 kV.
Confocal Microscopy. Biofilms on gold electrodes were fluo-
rescently stained with the LIVE/DEAD BacLight bacterial viability
kit (L7012, Molecular Probes, Inc., Eugene, OR) and examined
with the following exceptions. After disassembly of the ministacks
the electrodes were not rinsed before staining, as the biofilm could
easily detach. After staining (15 min) and soaking in fresh water
medium (5 min), the electrodes were placed upside down on glass
cover slips with a droplet of ProLong Antifade agent (P7481;
Molecular Probes, Inc., Eugene, OR).15A smaller glass cover slip
served as a spacer between one side of the gold electrode and the
glass coverslip, to create a cavity between electrode and cover slip,
where the biofilm was not disturbed and could be examined.
there was a lag period, followed by a rapid increase in current
which is associated with growth of the cells on the anode.
Assuming that cell growth is exponential, a doubling time of
(12-24 h).14Current stabilized at 0.4-0.7 mA after ca. 6-10
days. This maximum current is comparable to the maximum
current previously reported with carbon fiber anodes under the
same conditions.15The inability to produce more current is
(20) Gregory, K. B.; Sullivan, S. A.; Lovley, D. R. Presented at ASM General
(21) Blackstock, J. J.; Li, Z.; Freeman, M. R.; Stewart, D. R. Surf. Sci. 2003,
(22) Caccavo, F., Jr.; Lonergan, D. J.; Lovley, D. R.; Davis, M.; Stolz, J. F.;
McInerney, M. J. Appl. EnViron. Microbiol. 1994, 60, 3752-9.
(23) Reguera, G.; McCarthy, K. D.; Mehta, T.; Nicoll, J. S.; Tuominen, M.
T.; Lovley, D. R. Nature 2005, 435, 1098-101.
(24) Coppi, M. V.; Leang, C.; Sandler, S. J.; Lovley, D. R. Appl. EnViron.
Microbiol. 2001, 67, 3180-7.
(25) Lovley, D. R.; Phillips, E. J. P. Appl. EnViron. Microbiol. 1988, 54,
(26) Butler, J. E.; Kaufmann, F.; Coppi, M. V.; Nunez, C.; Lovley, D. R. J.
Bacteriol. 2004, 186, 4042-5.
(27) Chaudhuri, S. K.; Lovley, D. R. Nat. Biotechnol. 2003, 21, 1229-32.
(28) Niessen, J.; Schroeder, U.; Scholz, F. Electrochem. Commun. 2004, 6,
(29) Richter, H.; Lanthier, M.; Nevin, K. P.; Lovley, D. R. Appl. EnViron.
Microbiol. 2007, 73, 5347-5353.
Figure 1. Current generation in mA/m2electrode surface area by
G. sulfurreducens growing on gold electrodes poised at a potential
of +300 mV vs Ag/AgCl reference electrode. Four replicates A, B,
C, and D, are shown. (1) Inoculation. (2) Switch from recirculating
to flowthrough mode.
B LangmuirRichter et al.
than 70% of the 10 mM acetate flowing into the fuel cell was
recovered in the effluent.
with a biofilm that was visible with the naked eye and was red,
due to the high abundance of c-type cytochromes in G.
sulfurreducens. Coverage was heterogeneous and the biofilm
readily detached from the gold surface, in contrast to G.
sulfurreducens biofilms on the surface of graphite stick14,16or
graphite fiber15anodes to which the cells tightly adhere.
mA was 1.68 ( 0.38 mg. When graphite fiber anodes with
comparable geometric dimensions to the gold anodes were
substituted for the gold anodes, current production increased
per gram of cell protein was comparable for the two anode
materials. Thus, the increased current with the graphite fiber
anodes could be attributed to more biomass on these surfaces.
dimension, the highly textured carbon fiber anodes provide
much more surface for microbial attachment than the flat gold
with cells stacking in on top of each other (Figure 2). Zones
where the biofilm had begun to peel off the gold surface were
polysaccharide binding the cells together or to the electrode. In
some instances one or two long filaments extended from the
for extracellular electron transfer and biofilm formation16,23,30
than the 3-5 nm of the pili, (2) G. sulfurreducens pili emanate
from the side of the cell rather than the ends as seen here, and
(3) there are typically many pili emanating from each cell, but
only one or two filaments. It is more likely that these filaments
are flagella or strands of extracellular material.
However, pili were required for optimal current production
with gold anodes. A mutant in which the gene for PilA, the
structural pilin protein, had been deleted and thus did not make
pili,23did not produce current with gold anodes, and cells could
Confocal Laser Scanning Microscopy. The procedure for
The treatment with ethanol might wash away constituents, and
dehydration might affect structure and thickness.31Therefore,
after 10 days of growth covered most of the gold anode surface
and included pillars up to 12 µm high (Figure 3A). Most of cells
in the biofilm stained green, indicating that most of the cells
were metabolically active. With longer incubation, the biofilm
was a system of thin channels in the bottom layer of the biofilm
stained red, suggesting that they had compromised membranes
and might not be metabolically active.
These results demonstrate that gold electrodes are a suitable
electron acceptor for G. sulfurreducens, functioning nearly as
well as graphite. This is significant because although gold is
highly conductive, it was not clear that microorganisms could
use a gold anode as an electron acceptor. Gold does not contain
the functional groups, such as quinones, that are present on
Furthermore, recent studies with the electricity-producing
anode material for microbial fuel cells.7
Gold offers several potential advantages over graphite as an
anode material for some applications, as well as basic investiga-
tions of microbe-electrode interactions. For example, gold can
readily be deposited on a variety of materials and in a diversity
of configurations down to the nanoscale. This may be beneficial
for applications such as microbially based sensors and small-
scale microbial fuel cells. Furthermore, gold offers a highly
defined, conductive surface that may be ideal for evaluation of
the electrochemical properties of microorganisms growing on
their mechanisms for extracellular electron transfer to anodes
appear to be different.12Current-producing cells of G. sulfurre-
many of the cells in S. oneidensis fuel cells are planktonic.12
Cells of G. sulfurreducens closely associated with the anode
transfer through the biofilm on anodes may proceed via the
electrically conductive pili of G. sulfurreducens.16In contrast,
soluble electron shuttles are important for electron transfer to
anodes in Shewanella species.12,32It is beyond the scope of this
would not function with a gold anode, but one possibility is that
the electron shuttling molecule(s) are chemically unstable when
interacting with a gold surface.
The finding that, per gram of cell protein, G. sulfurreducens
is nearly as effective in transferring electrons to gold anodes as
transfer. This might be surprising because some c-type cyto-
(31) Fratesi, S. E.; Lynch, F. L.; Kirkland, B. L.; Brown, L. R. J. Sediment.
Res. 2004, 74, 858-867.
General Meeting, Toronto, Canada, 2007.
(33) Armstrong, F. A. Struct. Bond. 1990, 72, 137-221.
(35) Reed, D. E.; Hawkridge, F. M. Anal. Chem. 1987, 59, 2334-2339.
Table 1. Comparison of Current, Specific Electron Transfer
(ET) Rate in µmol of Electrons Per Minute and Gram Cell
Protein Attached, and Current Density Per Geometric Surface
Area of G. sulfurreducens on Working Electrodes Composed of
Gold or Graphitea
2.030 (0.024) 240 (3) 3147 (38)
aValues are means of triplicate experiments; standard deviations are
given in brackets.
Geobacter Gold ElectrodesLangmuir C
cytochromes are important in electron transfer to anodes by G.
sulfurreducens.17However, not all electrochemically active
biological molecules lose their function in contact with gold, as
cytochromes have yet been purified so their interaction with
gold surfaces cannot yet be readily investigated.
The finding that the mutant in which the gene for pilA was
deleted did not produce electricity suggests that the pili are
essential for electricity production with gold anodes. This result
contrasts with previous results with graphite anodes on which
able to attach to the anode surface, but did not form multiple
of pili.16The total lack of pilA-deficient cells on the gold anode
suggests that the pili are also required for attachment to the gold
anode surface. The reasons for this are not clear.
transfer electrons to gold anodes expands the range of materials
for this electron transfer to gold are warranted.
of Science (BER), U.S. Department of Energy, Cooperative
Agreement No. DE-FC02-02ER63446 and the Office of Naval
Research, Award No. N00014-06-1-0802. Confocal laser scan-
ning microscopy was supported by the National Science
Foundation Grant No. BBS8714235 to the University of
and Sudhanshu Srivastava from the Department of Chemistry at
University of Massachusetts, Amherst for their help fabricating
the gold electrodes.
(36) Hinne, C.; Parsons, R.; Niki, K. J. Electroanal. Chem. 1983, 147, 329-
(37) Barazzouk, S.; Kamat, P. V.; Hotchandani, S. J. Phys. Chem. B 2005,
Figure 2. SEM images of G. sulfurreducens growing on a gold electrode. Magnification: 75-20 000×. (A) Biofilm attached to the surface,
partially peeling off. (B) Closeup of Figure 3A where the biofilm was attached to the electrode surface. (C and D) Closeups of Figure 3A:
the edge of the biofilm.
Figure 3. Confocal laser scanning microscope images of G.
sulfurreducens biofilms on gold electrodes. Magnification: 250×.
The large images are top views on the biofilm. The smaller images
are orthogonal cross sections with the gold attached side of the
biofilm on the left, the outer surface on the right. The vertical blue
from gold electrode C in Figure 1 after 10 days. (B) Biofilm from
gold electrode D in Figure 1 after 18 days.
D Langmuir PAGE EST: 4 Richter et al.