University of Southern California Dissertations and Theses

Electron transfer capability and metabolic processes of the genus Shewanella with applications to the optimization of microbial fuel cells

(USC Thesis Other)

Electron transfer capability and metabolic processes of the genus Shewanella with applications to the optimization of microbial fuel cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

ELECTRON TRANSFER CAPABILITY AND METABOLIC PROCESSES OF
THE GENUS SHEWANELLA WITH APPLICATIONS TO THE OPTIMIZATION OF
MICROBIAL FUEL CELLS

by

Orianna Bretschger

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(ENGINEERING)

August 2008

Copyright 2008 Orianna Bretschger
ii
Epigraph

We stand now where two roads diverge. But unlike the roads in Robert Frost's
familiar poem, they are not equally fair. The road we have long been traveling is
deceptively easy, a smooth superhighway on which we progress with great speed, but at
its end lies disaster. The other fork of the road, the one less traveled by, offers our last,
our only chance to reach a destination that assures the preservation of the earth.
-Rachel Carson
iii
Dedication

For my teachers, starting with my parents whose love and support taught me the most.

iv
Acknowledgements

I have many people to thank and acknowledge for their help and advice over the
last four years. Many thanks are due to In Seop Chang and Byung Hong Kim for
teaching me how to work with microbial fuel cells at the Korea Institute of Science and
Technology. Thanks also to the University of Southern California Glass shop and
Machine shop for building the microbial fuel cells utilized in theses studies. Thanks to
several people at Pacific Northwest National Laboratory, University of Wisconsin-
Milwaukee, Oak Ridge National Laboratory and the University of Oklahoma for
generating the MR-1 mutants used in several studies. Thanks are also due to Alicia
Thompson and John Curulli at the University of Southern California Center for Electron
Microscopy for teaching me how to collect scanning electron microscopy images.
Thanks to all my friends and colleagues in the Nealson Lab that have shared their
knowledge and helped me over the years, most especially Anna Obraztsova, Andrea C. M.
Cheung, Lewis Hsu, Carie M. Frantz, and Prithviraj Chellamuthu. Last but not least,
many thanks to John P. McCrow for lovingly tolerating my absence, as well as my
sometimes intolerable presence, for many years.

v
Table of Contents
Epigraph ii
Dedication iii
Acknowledgements iv
Table List viii
Figure List ix
Abstract 1
Chapter 1: Introduction 2
1.1 Background 3
1.2 Justification 12
Chapter 2: Current production and metal oxide reduction by Shewanella
oneidensis MR-1 wild type and mutants 15
2.1 Materials and Methods 18
2.1.1 Culture and growth conditions 18
2.1.2 Mutagenesis 21
2.1.3 MFC Experiments 22
2.1.4 Manganese(IV) oxide 24
2.1.5 Solid-phase iron(III) oxide 25
2.1.6 Soluble iron(III) 26
2.1.7 Scanning Electron Microscopy (SEM) sample preparation 26
2.2 Results 26
2.2.1 MFC Experiments 26
2.2.2 Manganese(IV) oxide 31
2.2.3 Solid-phase Fe(III)-oxide 34
2.2.4 Soluble iron(III) 36
2.3 Discussion 43
Chapter 3: Evaluation of the power producing abilities of different Shewanella
strains 52
vi
3.1 Initial strain evaluations (KIST) 54
3.1.1 KIST methods 54
3.1.2 KIST results and discussion 56
3.2 Expanded strain evaluations (USC) 61
3.2.1 USC methods 63
3.2.2 USC results 66
3.3 Electrochemical evaluations of strain performance 87
3.3.1 ElS 87
3.3.2 Potentiodynamic polarization 91
3.3.3 Cyclic voltammetry 94
Chapter 4: Microbial fuel cells as a tool for studying microbial physiology 96
4.1 Microbial fuel cell evaluation of Shewanella oneidensis MR1 grown in
bioreactors with different dissolved oxygen tensions 97
4.1.1 Materials and Methods 99
4.1.2 Results and Discussion 100
4.2 Shewanella oneidensis MR-1 power production using different
reductants 102
4.3 Shewanella oneidensis MR-1 power production with an ex-situ grown
biofilm 104
4.4 Shewanella oneidensis MR-1 wild-type and mutants acting as cathode
catalysts in an MFC with different oxidants 106
4.4.1 Materials and Methods 109
4.4.2 Results and Discussion 111
4.4.3 Conclusions 119
Chapter 5: Conclusions 120
5.1 Evaluations of Shewanella oneidensis MR-1 and mutants 120
5.2 Shewanella strain evaluations 125
5.3 MFCs as tools for studying microbial physiology 134
5.4 Future Work 136
Bibliography 139
Appendix A Electrode and membrane pretreatment procedures 157
vii
Graphite Felt Electrodes 157
Electrode pretreatment 157
Electrode Cleaning 158
Cathode electroplating 158
Nafion membranes 159
Appendix B SEM preparation of biological samples 160
Appendix C Organic acid sample preparation for HPLC analysis 161
viii
Table List

Table 1. Summary of the evaluated set of MR-1 cytochrome mutants,
complements, and select transport protein mutants. 19

Table 2. pH values for ∆mtrA and WT at zero hours and ninety-six hours of
exposure to multiple forms of soluble ferric iron. 42

Table 3. Cell densities after inoculation and after 96 hours of exposure to
different forms of soluble iron(III). 42

Table 4. Shewanella strains and a description of the environment they were
isolated from. 62

Table 5. Open-circuit potentials (vs. Ag/AgCl) recorded for the anode and
cathode of one of three MFCs and the average open-circuit cell voltage
found for all three MFCs. 70

Table 6. Current densities at I
max
for each strain evaluated in three different
MFC systems. Highest values are shown in dark grey, the next highest
values are highlighted in light grey, and the lowest values are shown in
white. 86
ix
Figure List

Figure 1. General fuel cell diagram 2

Figure 2. Dual compartment microbial fuel cell diagram. Bacteria are
inoculated into the anode compartment, attach to the graphite felt
electrode and begin transferring electrons. Electrons are conducted
through the anode electrode and across the external circuit to the
cathode electrode. The cathode electrode is graphite felt that has
been electroplated with platinum, the catalyst driving the reduction
of oxygen to water. The anode and cathode compartments are
physically separated by a proton-conductive membrane that
facilitates the transfer of protons from the anode to the cathode,
completing the cell reaction. The cell voltage, V, is measured across
an external load of resistance, R, and current is calculated as I = V/R. 22

Figure 3. Current density values for MR-1 WT, cytochrome deletions and
protein secretion mutants. Averages and standard deviations were
obtained using the peak current density values corresponding to each
lactate injection for triplicate experiments. Maximum current
density was determined to be the highest level of current density that
remained constant for at least three hours. Averages and standard
deviations of the maximum current density were calculated using
these data values (between one-hundred to two-hundred data points
were utilized). 27

Figure 4. Current density values for MR-1 WT (black) and selected
cytochrome mutants (grey) with their complementations (white).
Averages and standard deviations were obtained using the peak
current density values corresponding to each lactate injection for
triplicate experiments. Maximum current density was determined to
be the highest level of current density that remained constant for at
least three hours. Averages and standard deviations of the maximum
current density were calculated using these data values (between
one-hundred to two-hundred data points were utilized). 28

Figure 5. SEM image of graphite anode fibres used during the MFC
evaluations of MR-1. 30

Figure 6. SEM images of graphite anode fibers used during the MFC
evaluations of the MR-1 ∆pilD mutant. 30

Figure 7. SEM images of graphite anode fibers used during the MFC
evaluations of the MR-1 ∆mtrC/∆omcA mutant. 31
x

Figure 8. Average percentage of Mn(IV)-oxide reduced after twenty-four
hours of exposure to MR-1 wild-type (black), cytochrome deletion
and proteins secretion mutants (grey). 32

Figure 9. Average percentage of Mn(IV)-oxide reduced after twenty-four
hours of exposure to MR-1 wild-type (black), selected cytochrome
deletion mutants (grey) and their respective complements (white). 33

Figure 10. Average Fe(II) concentration resulting from solid Fe(III)-oxide
(HFOM) reduction after twenty-four hours of exposure to MR-1 WT
(black), cytochrome deletions, and protein secretion mutants (grey).
Averages and standard deviations were calculated based on the
measured Fe(II) concentrations from triplicate experiments. 34

Figure 11. Average Fe(II) concentration resulting from solid Fe(III)-oxide
(HFOM) reduction after twenty-four hours of exposure to MR-1 WT
(black), select cytochrome deletions (grey), and their respective
complementations (white). Averages and standard deviations were
calculated based on the measured Fe(II) concentrations from
triplicate experiments. 36

Figure 12. Average Fe(II) concentration resulting from soluble Fe(III)-NTA
reduction after twenty-four hours of exposure to MR-1 WT (black),
cytochrome deletions, and protein secretion mutants (grey).
Averages and standard deviations were calculated based on the
measured Fe(II) concentrations from duplicate experiments. 37

Figure 13. Average Fe(II) concentration resulting from soluble Fe(III)-NTA
reduction after twenty-four hours of exposure to MR-1 WT (black),
selected cytochrome deletions (grey), and their respective
complementations (white). Averages and standard deviations were
calculated based on the measured Fe(II) concentrations from
duplicate experiments. 38

Figure 14. Average Fe(II) concentration resulting from soluble Fe(III) reduction
after a) twenty-four hours and b) forty-eight hours of exposure to
MR-1 WT (black) and mtrA mutant (grey). Averages and standard
deviations were calculated based on the measured Fe(II)
concentrations from triplicate experiments. Starting pH values for
Fe(III)-citrate, Fe(III)-NTA(Na+) and Fe(III)-NTA(H+) were 5.6,
7.1 and 4.1, respectively. 40

xi
Figure 15. Average Fe(II) concentrations resulting from soluble Fe(III)
reduction over a period of ninety-six hours. Averages and standard
deviations were calculated based on the measured Fe(II)
concentrations from triplicate experiments. Starting pH values for
Fe(III)-citrate, Fe(III)-NTA(Na+) and Fe(III)-NTA(H+) were 5.6,
7.1 and 4.1, respectively. 41

Figure 16. Average Fe(II) concentration resulting from the reduction of a) 2.5
mM Fe(III)-citrate, and b) 20 mM Fe(III)-NTA(H+). Averages were
calculated based on the measured Fe(II) concentrations from
duplicate experiments. Starting pH values for Fe(III)-citrate and
Fe(III)-NTA(H+) were 6.1 and 4.1, respectively. 43

Figure 17. Average Fe(II) concentration and cell counts resulting from HFOM
reduction after twenty-four hours of exposure to mtrA mutants.
Averages and standard deviations were calculated based on the
measured Fe(II) concentrations from triplicate experiments. 47

Figure 18. Average Fe(II) concentration and cell counts resulting from HFOM
reduction after twenty-four hours of exposure to mtrC/omcA mutants.
Averages and standard deviations were calculated based on the
measured Fe(II) concentrations from triplicate experiments. 47

Figure 19. Average Fe(II) concentration and cell counts resulting from HFOM
reduction after twenty-four hours of exposure to MR1 wild-type.
Averages and standard deviations were calculated based on the
measured Fe(II) concentrations from triplicate experiments. 48

Figure 20. Average Fe(II) concentration and cell counts resulting from HFOM
reduction after twenty-four hours of exposure to MR1 wild-type
membrane fractions. Averages and standard deviations were
calculated based on the measured Fe(II) concentrations from
triplicate experiments. 49

Figure 21. Current densities for each MFC and Shewanella strain for a) the first
lactate feeding and b) the second lactate feeding. Average charge
densities and coulombic efficiencies are shown in c) and d),
respectively. Averages and standard deviations were calculated
using the total charge and measured organic acid concentrations
corresponding to both feedings, in three MFCs, for each strain. 57

xii
Figure 22. Power densities for a) System 1, and b) System 2; and current
densities for c) System 1, and d) System 2. Power density averages
and standard deviations were calculated using the maximum power
generated during potential sweep experiments for three MFCs.
Current densities were calculated as the average stable current for
ten hours of operation across an applied load determined from the
potential sweep experiments for three MFCs. System 1 parameters
included 100 mM phosphate buffer, 20 mM lactate, and Nafion 424
membranes. System 2 parameters included a 50 mM PIPES buffer,
5 mM lactate, and Nafion 117 membranes. All other MFC
components including electrode size, compartment volume, and gas
diffusion rates were the same. 68

Figure 23. a) Charge density, and b) coulombic efficiency for eight Shewanella
strains evaluated in triplicate using a 50 mM PIPES buffer and 5 mM
lactate anolyte with a Nafion 117 membrane. Charge density
averages and standard deviations for each strain were calculated
based on the total charge measured for each lactate feed in each
MFC. Average coulombic efficiencies and corresponding deviations
were calculated as the total charge from each MFC divided by the
total theoretical charge available from the sum of each lactate feed to
the corresponding MFC. Anolyte concentrations for acetate, formate
and pyruvate were determined using high pressure liquid
chromatography. 71

Figure 24. S. oneidensis MR-1 I-t and HPLC profiles for each MFC tested. 73

Figure 25. S. putrefaciens W3-18-1, I-t and HPLC profiles for each MFC tested. 74

Figure 26. S. amazonensis SB2B I-t and HPLC profiles for each MFC tested. 75

Figure 27. S. ANA-3 I-t and HPLC profiles for each MFC tested. 76

Figure 28. S. oneidensis MR-4 I-t and HPLC profiles for each MFC tested. 77

Figure 29. S. oneidensis MR-7 I-t and HPLC profiles for each MFC tested. 78

Figure 30. S. putrefaciens I-t and HPLC profiles for each MFC tested. 79

Figure 31. S. loihica PV-4 I-t and HPLC profiles for each MFC tested. 80

Figure 32. Scanning electron microscopy images of graphite fiber anode
electrodes exposed to S. oneidensis MR1, S. putrefaciens W3-18-1
and S. amazonensis SB2B, respectively, in System 1 and System 2. 82

xiii
Figure 33. Scanning electron microscopy images of graphite fiber anode
electrodes exposed to S. ANA-3, S. oneidensis MR-4 and S.
oneidensis MR-7, respectively, in System 1 and System 2. 83

Figure 34. Scanning electron microscopy images of graphite fiber anode
electrodes exposed to S. putrefaciens CN-32 and S. loihica PV-4,
respectively, in System 1 and System 2. 83

Figure 35. Electrochemical impedance spectroscopy data for the anode and
cathode of MFCs operated with sterile anodes or an anode bacterial
catalyst including a) S. oneidensis MR-1, c) S. putrefaciens W3-18-1
and e) S. loihica PV-4. Figures b), d) and f) are the sterile cathode
spectra recorded under different anode conditions for the
corresponding bacterial catalyst. All spectra were collected at the
open-circuit potentials (vs. Ag/AgCl) of the anode and cathode,
respectively. 88

Figure 36. Equivalent circuit for impedance spectra showing inductance
behavior and low frequencies 90

Figure 37. Potentiodynamic scans of the anode and cathode for sterile anode
conditions and after a) S. oneidensis MR-1, b) S. putrefaciens W3-
18-1, and c) S. loihica PV-4 had been inoculated to the MFC anode.
All scans were performed at a rate of 0.167 mV/sec vs Ag/AgCl. 93

Figure 38. Cyclic voltammograms for S. oneidensis MR-1, S. putrefaciens W3-
18-1, and S. loihica PV-4 collected at a scan rate of 25 mV/sec. MR-
1 and W3-18-1 scans were collected in a range of -750 mV to
+700mV and PV-4 scans were collected from -750 mV to 850 mV
vs. Ag/AgCl. 95

Figure 39. Average stable current densities from six MFCs inoculated with a
5% or 50% DOT grown MR-1 culture. Data for both cultures are
shown for two lactate feeds, and the averages and standard
deviations were calculated using seven hours of data corresponding
to the stable current densities for each feeding. 101

xiv
Figure 40. I-t curves for duplicate MFCs featuring MR-1 grown at 50% DOT
(OD
600
= 0.15) and fed lactate as the initial carbon sources and a)
acetate or b) succinate. Figures c) and d), respectively show I-t
curves for duplicate MFCs featuring MR-1 grown at 5% DOT
(OD
600
= 0.8) and fed lactate as the initial carbon source and either
formate or pyruvate as the final carbon source. The MFCs fed
pyruvate, d), were also operating with a lower dissolved oxygen
tension at the cathodes relative to those MFCs fed other carbon
sources, a), b) and c), respectively. 103

Figure 41. I-t curves for the first thirty hours of operation of a) an ex-situ grown,
anode biofilm; and b) an in-situ grown anode biofilm. Scanning
electron microscopy images of c) ex-situ grown, anode biofilm; and
d) in-situ grown anode biofilm. Each MFC was operated using a 10
ohm resistor and only lactate was provided in the MFC anode
compartment as a carbon source and electron donor. The in-situ
grown biofilm was achieved by exposing the anode electrode to
approximately 2 x 10
9
cells/mL of planktonic MR-1 and a total of 6
mM of lactate. 105

Figure 42. Average current densities obtained for Shewanella oneidensis MR-1
WT at the anode with lactate as an electron donor, and WT (black)
or mutants (grey) at the cathode with oxygen, Fe(III)-citrate, or
fumarate as the oxidant. Controls (white) were run using WT as the
anode catalyst with lactate as the electron donor, and only the
oxidant at the cathode, with no bacteria. Averages and standard
deviations were calculated from triplicate experiments, using the
data associated with the observed stable current density over twelve
hours of operation. 112

Figure 43. a) Charge densities, and b) coulombic efficiencies for each evaluated
cathode catalyst and cathode oxidant. Averages and standard
deviations were calculated using the total experimental and
theoretical charge yielded from triplicate MFC evaluations. 113

Figure 44. Scanning electron microscopy images of graphite electrode fibers
exposed to WT as the anode catalyst and WT, ∆cyoA or ∆mtrB as
the cathode catalyst. Oxygen served as the oxidant at the cathode
and lactate as the reductant at the anode. 114

Figure 45. Scanning electron microscopy images of graphite electrode fibers
exposed to WT as the anode catalyst and WT, ∆cymA, ∆mtrB or
∆fccA as the cathode catalyst. Fumarate served as the oxidant at the
cathode and lactate as the reductant at the anode. 117

xv
Figure 46. Coulombic efficiency for each strain during the second lactate
feeding only. 130

Figure 47. Growth curves over a forty-eight hour period for S. oneidensis MR-1,
S. putrefaciens W3-18-1, S. amazonensis SB2B, S. ANA-3, S.
oneidensis MR-4, S. oneidensis MR-7, S. putrefaciens CN-32 and S.
loihica PV-4. All strains were grown aerobically at 30°C at an
agitation rate of 150 rpm using PIPES buffered minimal media with
18 mM lactate as the electron donor. 132

1
Abstract
The mechanisms that bacteria employ to transfer electrons to their surrounding
environments are diverse and not well understood. This study provides original data that
begin to elucidate specific mechanisms involved with electron transfer to microbial fuel
cell (MFC) electrodes, Fe(III)- and Mn(IV)-oxides using various strains and species of
the genus Shewanella. Additionally, MFC were used to study the carbon and energy
metabolisms of several different Shewanella strains, identify efficiencies, and study the
physiology of Shewanella oneidensis MR-1. These data have implications toward the
optimization of bioremediation technologies, MFC development and to the fundamental
understanding of how bacteria interact with and affect their environments.

2
Chapter 1: Introduction

An ever-increasing global energy demand and the deleterious environmental
effects of burning fossil fuels have inspired a worldwide exploration of many alternative
energy technologies including fuel cells. A fuel cell is an electrochemical system that
couples an oxidation and reduction reaction via an external electrical circuit and internal
proton flow. For example, in a dual-compartment fuel cell, fuel is oxidized at the anode
electrode resulting in electrons, protons and an oxidized product. The oxidation products
diffuse from the system while the electrons travel externally across a circuit, making
electricity, and the protons flow internally through a proton selective membrane. The
electrons and protons recombine with an oxidant at the cathode electrode and produce a
reduced oxidant. Both the oxidation and reduction reactions are facilitated by a catalyst
at the anode and cathode electrodes, respectively (Figure 1).

Figure 1. General fuel cell diagram

3
Fuel cells offer the ability to produce electricity in a non-polluting manner and, if
a microbial fuel cell (MFC) is considered, may also facilitate the removal of organic, or
inorganic, waste from water supplies while producing a by-product of electricity (Roller,
Bennetto et al. 1984; Allen and Bennetto 1993; Liu, Ramnarayanan et al. 2004; Min and
Logan 2004; Min, Kim et al. 2005; Bullen, Arnot et al. 2006; Lane 2006; Logan,
Hamelers et al. 2006). MFC systems may also be employed to produce hydrogen or
other high energy fuels for chemical fuel cells.
A MFC is an electrochemical system, in which living microorganisms are utilized
as catalysts to drive the oxidation (Cohen 1931; Roller, Bennetto et al. 1984; Allen and
Bennetto 1993; Kim, Kim et al. 1999; Bond, Holmes et al. 2002; Rabaey, Lissens et al.
2003; Liu and Logan 2004) and/or reduction reactions (He and Angenent 2006;
Clauwaert, van der Ha et al. 2007; Rozendal, Jeremiasse et al. 2008). There are several
advantages associated with the use of microbial catalysts including: 1) microorganisms
are extremely inexpensive relative to metallic catalysts; 2) microorganisms are capable of
self repair and maintenance; 3) microorganisms can be metabolically versatile allowing
for the use of many different energy sources, i.e. fuel sources. For these reasons, there
has been mounting interest throughout the last century, and continuing today, for the
development of MFC technology.
1.1 Background

The idea that microorganisms could facilitate an electrical signal through their
metabolic processes was introduced in the year 1911 by M.C. Potter (Potter 1911).
Potter used a glass electrochemical cell with two platinum wires as electrodes and a

4
parchment dialysis bag as a membrane. A glucose solution, or other biological media,
was used as the anolyte and catholyte to investigate if electrical signals could be observed
when microorganisms were introduced into either side of the dialysis bag. Potter
discovered that electrical signals produced from his systems were unique and dependent
on the type of microorganism. Potter utilized a number microorganisms including
Bacillus coli communis (now known as Escherichia coli), Bacillus flourescens (now
classified as Pseudomonas flourescens), Bacillus violaceus, Sarcina lutea (now known as
Mirococcus variens migula), and Saccharomyces cerevisiae (yeast). Yeast served as the
model organism during most of the evaluations. Potter measured the electromotive force
(EMF), or cell voltage, generated by his system by connecting the electrodes to a
condenser (capacitor) of one micro-farad, which would discharge through a
galvanometer by means of a Morse key. The resulting measurement on the
galvanometric scale corresponded to a given cell voltage. Potter observed an EMF when
the yeast began fermenting glucose and found that the cell voltage varied with glucose
concentration, yeast concentration, and temperature. Additionally, he found that different
electrode materials produced different electrical signals when exposed to the same yeast
culture. In this publication, Potter was the first to propose that enzymes, specifically
invertase and diastase display a unique electrical signal and should be studied further
in terms of their electrochemical activity. Although Potter was able to demonstrate that
the decomposition of glucose by microorganisms yielded an electrical signal, and that
harnessing the signal in series would produce 1.25 mA of current, this body of work did
not immediately inspire further investigations.

5
During the Thirty-second Annual Meeting of American Bacteriologists in 1931,
Dr. Barnett Cohen presented an abstract entitled The Bacterial Culture as an
Electrochemical Half-Cell (Cohen 1931). In this abstract Cohen described unique cell
voltages attributed to different bacterial cultures in an electrochemical half-cell, and
reported the use of potassium ferricyanide (K
3
[Fe(CN)
6
]) and benzoquinone (C
6
H
4
O
2
) to
increase the cell voltage by changing the redox potential of the electrolyte at the anode.
This report appears to be the first use of chemical oxidizing agents, or mediators, in a
microbial fuel cell. However, Cohen did not specify how these compounds interacted
between the bacterial cultures and the noble metal electrodes.
In 1934 von Wolzogen Kühr introduced the concept of microbially induced
corrosion (von Wolzogen Kuhr and van der Vlugt 1934). The authors concluded that the
microbially driven process of sulfate reduction by organisms like Desulfovibrio, under
anaerobic conditions, was responsible for the accelerated corrosion of cast iron water
pipes buried several meters underground. The idea that microbially induced corrosion
processes could be related to other electrochemical systems, including batteries and fuel
cells, was followed up by scientists in the early 1960s.
Dr. Frederick D. Sisler published an article in 1961 that reported the use of an
unspecified Desulfovibrio strain as a biocatalyst to oxidize calcium lactate to electrons
and protons in anaerobic artificial seawater media (Sisler 1961). Sisler used a potassium
chloride salt bridge to link the anode compartment, containing Desulfovibrio, with the
cathode compartment containing sterile, oxygenated seawater as the electrolyte. The
cathode reduction reaction of oxygen was believed to be abiotic. Sisler was able to

6
produce an open-circuit cell voltage of 0.5 V and reported a maximum current in excess
of one milliampere. Here the author postulated that these biochemical fuel cells could
use any number of microorganisms to drive the catalytic reactions and that these systems
could be deployed in remote areas to act as sustainable energy sources. In a 1962
publication by Sisler, he outlines the importance of this research to the United States
Navy and briefly discusses the Biological Electrical Energy Production (BEEP) project
that was conducted by the Advanced Concepts Division, Bureau of Ships (Sisler 1962).
Sisler reported that BEEP studied many variations of biological fuel cells including
systems with different electrodes, bacterial strains, yeast, algae, and enzymes. The
current densities that resulted from these studies were observed to be between 0.10 and
10.0 A/ft
2
.
It was also in 1962 that MFCs made their debut in Science magazine with an
article published by two researchers at the Mobile Oil Company, J. B. Davis and H. F.
Yarbrough (Davis and Yarbrough 1962). In this publication the two authors tested the
use of hydrocarbons, specifically ethane, as a fuel source in a microbial fuel cell that
featured the ethane-oxidizing microbe, Nocardia, as the biocatalyst at the anode. The
authors also investigated the use of Escherichia coli and the enzyme, glucose oxidase,
with glucose as the fuel at the anode of their fuel cell system.
Davis and Yarbrough found that Nocardia would produce an open-cell voltage
(OCV) of 192 mV, but when the cell was operated with a 1 k-Ohm resistor as the load, no
measurable current was observed. When the authors explored the enzymatic glucose
oxidase cell with glucose as the fuel, they observed an OCV of 198 mV, but no

7
measurable current increase. However, the addition of methylene blue, a chemical redox
and hydrogen acceptor, to the anode did produce a small current when the MFC was
operated under a load. Davis and Yarbrough also added methylene blue to cultures of
Escherichia coli and glucose, and observed a 6-fold increase in open-circuit cell voltage
relative to the cultures without methylene blue. The authors additionally reported that the
methylene blue was rapidly decolorized, indicating that the redox indicator had been fully
reduced.
To further explore the impact of chemical redox compounds, the authors
introduced potassium ferricyanide into the cathode compartment. This chemical addition
increased the current production of the E. coli system (with methylene blue) by a factor
of ten. Although Davis and Yarbrough suggested that the added chemical redox
indicators play a role as hydrogen acceptors for the organisms tested, they did not
speculate about the mechanisms of biological electron transfer.
The subject of biological electron transfer, as opposed to hydrogen transfer, was
introduced by Yahiro, Lee and Kimble in 1964 (Yahiro, Lee et al. 1964) during their
explorations of enzymatic electron transfer in biofuel cells. These authors employed the
enzyme glucose oxidase with glucose as a fuel and found that the enzyme, which
catalyzed the oxidation of glucose to gluconic acid, was also responsible for electron
transfer to the electrode via iron molecules present in the enzyme.
Through the mid-1960s enzymatic fuel cells and microbial fuel cells were
explored theoretically (Young, Hadjipetrou et al. 1966), and practically using different
enzyme systems and live cultures of yeast and bacteria (Davis and Yarbrough 1962;

8
Rohrback 1962; Brake, Momyer et al. 1963; Davis 1963; Konikoff 1963; Reynolds and
Konikoff 1963; Shaw 1963; Berk and Canfield 1964; May, Blanchard et al. 1964; Yahiro,
Lee et al. 1964; Fischer, Landes et al. 1965; van Hees 1965; Lewis 1966). Most of these
authors were concerned with developing microbial fuel cells for military power
production, hydrogen production, or for waste management on manned space missions;
however it was suggested by Cohn during the first symposium on biochemical fuel cells,
that MFCs could be utilized further. He postulated that MFCs could be employed for
agricultural uses, desalination of water, heating, and lighting Cohn also went further
to express that Economical biocells could thus be of far-reaching importance to
underdeveloped areas (Cohn 1963). Even though it was recognized in the 1960s that
the development of MFCs held promise for many applications, another ten years passed
before a new group of scientists began fervently exploring the practical uses of MFC
technology.
In the late 1970s and early 1980s, MFC research rekindled primarily in Europe
and Asia with investigations that triggered the modern study of MFC biological
mechanisms, design and pilot applications.
During this period, hydrogen was receiving much attention as a fuel resource, and
the use of MFCs for hydrogen production was exuberantly revisited. Beginning in 1976,
several reports about immobilized cells of Clostridium butyricum and Rhodospirillum
rubrum, in MFCs, were published by Japanese and European researchers (Karube,
Matsunaga et al. 1976; Karube, Matsunaga et al. 1977; Suzuki, Karube et al. 1980;
Karube, Suzuki et al. 1981; Karube, Matsuoka et al. 1984; von Felten, Zurrer et al. 1985)

9
exploring the conversion of carbohydrates to hydrogen. Additionally, Suzuki, Karube
and Matsunaga presented work at the 1978 Biotechnology and Bioengineering
Symposium that addressed using MFCs for wastewater treatment (Suzuki, Karube et al.
1978). Also beginning in 1976, MFC electrodes were explored as tools for detecting
microbial populations in polluted waters, and used as sensors for organic contaminants
and glucose concentrations (Rao, Richter et al. 1976; Wilkins 1978; Matsunaga, Karube
et al. 1979; Junter, Lemeland et al. 1980; Matsunaga, Karube et al. 1980; Wilkins, Grana
et al. 1980; Nishikawa, Sakai et al. 1982; Turner and Ramsay 1983; Yang and Yang
1988).
While hydrogen production and wastewater treatment were being explored in
Europe and Asia, another group of scientists were working in South America to
understand the electrochemical reactions that occur in MFC systems. Several
publications by Disalvo, Videla and de Mele investigated the electrochemical oxidation
of glucose and the relationship between biochemical and electrochemical reactions in
MFCs using Saccharomyces cerevisiae and Micrococcus cerificans cultures (Videla and
Arvía 1975; Disalvo and Videla 1979; Disalvo 1980; Disalvo 1980; Disalvo and Videla
1981; de Mele, Videla et al. 1982).
By the early 1980s a group of scientists from the United Kingdom began
scrutinizing the biochemical and electrochemical reactions that were related to MFC
systems employing chemical redox indicators such as methylene blue, thionene, and
potassium ferricyanide. It was proven by Bennetto, Delaney and Allen that these redox
indicators act as electron shuttles that are reduced by microorganisms and oxidized by the

10
MFC electrodes thereby transporting the electrons produced via biological metabolism to
the electrodes in a fuel cell (Bennetto, Stirling et al. 1983; Roller, Bennetto et al. 1984;
Bennetto, Delaney et al. 1985; Akiba, Bennetto et al. 1987; Allen and Bennetto 1992).
Bennetto was also responsible for promoting the use of MFCs as teaching tools for
understanding microbial physiology and microbial electrochemistry. He created
commercially available MFC systems for the classroom and heavily advocated the
development of MFC technology for scientific exploration and practical applications
(Bennetto 1987; Bennetto 1990; Bennetto 1990).
Until the late 1980s and early 1990s, MFC study was confined to using only a
handful of different organisms to catalyze oxidation and reduction reactions. The focus
of most investigations was glucose oxidation using, glucose oxidase enzymes,
Saccharomyces cervisiae, Escherichia coli, and a few other fermentative organisms.
However these confines were broken with the discovery and use of dissimilatory metal
reducing (DMRB) and electrochemically active bacteria in MFC systems. These unique
organisms can transfer electrons directly to solid substrates by way of outer membrane
proteins (cytochromes) that are up-regulated under anaerobic conditions (Myers and
Nealson 1988; Myers and Myers 1992; Beliaev and Saffarini 1998; Beliaev, Saffarini et
al. 2001; Wan, VerBerkmoes et al. 2004). The majority of organisms that have been
found to employ this type of electron transfer are DMRB and electrochemically active
strains have been found in genera such as Clostridium (Park, Kim et al. 2001), Geobacter
(Bond and Lovley 2003), Aeromonas (Pham, Jung et al. 2003), Rhodoferax (Chaudhuri

11
and Lovley 2003), Desulfobulbus (Holmes, Bond et al. 2004) and Shewanella (Kim, Kim
et al. 1999; Chang, Moon et al. 2006; Lovley 2006).
With the discovery of DMRB came a new way to employ and study MFC systems.
Presently, there are several groups focused on applying MFCs for wastewater treatment
and they use these systems to identify new electrochemically active organisms, and
communities of organisms, that naturally exist in pristine or contaminated waters
(Reimers, Tender et al. 2001; Bond, Holmes et al. 2002; Tender, Reimers et al. 2002; Lee,
Phung et al. 2003; Rabaey, Lissens et al. 2003; Back, Kim et al. 2004; Choi, Jung et al.
2004; Kim 2004; Kim, Park et al. 2004; Liu, Ramnarayanan et al. 2004; Min and Logan
2004; Moon, Chang et al. 2004; Phung, Lee et al. 2004; Rabaey, Boon et al. 2004; Liu,
Cheng et al. 2005; Min, Kim et al. 2005; Moon 2005; Rabaey and Verstraete 2005;
Chang, Moon et al. 2006; Pham, Rabaey et al. 2006; Freguia, Rabaey et al. 2007; Kim,
Jung et al. 2007; Rabaey, Rodriguez et al. 2007; Chen, Choi et al. 2008; Rabaey, Read et
al. 2008). These same enrichment and community analysis techniques are being
employed for MFC hydrogen production (Liu, Grot et al. 2005; Oh and Logan 2005;
Logan, Hamelers et al. 2006) and bioremediation (Gregory and Lovley 2005).
Additionally, with the understanding that certain bacteria can physically attach to MFC
electrodes and catalyze reactions, many new design concepts have been introduced to
increase electrode surface area, enhance proton conductivity, and otherwise enhance
microbial activity at electrode surfaces (Park and Zeikus 2002; Park and Zeikus 2003;
Schröder, Nießen et al. 2003; Liu and Logan 2004; Niessen, Schröder et al. 2004;

12
Aelterman, Rabaey et al. 2006; Cheng, Liu et al. 2006; Ringeisen, Henderson et al. 2006;
Biffinger, Pietron et al. 2007; Biffinger, Ray et al. 2007).
1.2 Justification

The identification of new electrochemically active bacteria and the improvement
of MFC designs for specific applications are extraordinarily important pursuits. However,
if the mechanisms behind bacterial electron transfer are not well understood, the
development of MFC technology will again face a road block. A well designed
electrochemical system will still be limited by catalytic kinetics and efficiency.
Therefore, it is of fundamental importance to understand how microbial catalysts function
in MFC environments; and how to effectively employ the diversity of microbial
metabolism and different microbial strategies for electron transfer to solid substrates
within their environments. This later topic has been widely studied, and hotly debated,
throughout the biological community since the discovery of DMRB; and there is still
much that is unknown.
For these reasons, this study focuses on understanding the mechanism(s)
employed by several strains in the genus Shewanella to adapt and thrive in MFC
environments. First, Shewanella oneidensis MR-1, a DMRB strain, was chosen as a
model organism because it demonstrates great metabolic flexibility (Myers and Nealson
1988), has its genome sequenced (Heidelberg, Paulsen et al. 2002), and is genetically
accessible with regard to both generating and complementing mutants (Gorby, Yanina et
al. 2006).

13
Strain MR-1 was initially studied because of its ability to reduce metal oxides
including manganese and iron oxides (Myers and Nealson 1988). The mechanisms
responsible for metal oxide reduction are not fully elucidated, but it is clear that a number
of c-type cytochromes and affiliated proteins are essential (Beliaev, Klingeman et al.
2005; Gorby, Yanina et al. 2006). For example, several multi-heme c-type cytochromes
are found on the outer-membrane during conditions of electron acceptor limitation, and
are involved with extra-cellular electron transfer to iron oxides (Myers and Myers 1992;
Beliaev, Saffarini et al. 2001; Gorby, Yanina et al. 2006). Given that these multi-heme c-
type cytochromes are essential for dissimilatory metal reduction, the question thus arose
as to whether they are also involved in the transfer of electrons to other solid substrates
including electrodes.
Second, the entire genome of nineteen Shewanella strains have now been
sequenced and it is known that many of these strains feature similar genetic machinery
and produce several outer-membrane multi-heme c-type cytochromes during conditions
of electron acceptor limitation (Nealson and Scott 2006; Hau and Gralnick 2007).
However, even though the production of several c-type cytochromes is a seemingly
shared trait, the respiratory capacity of each strain is unique. This leads to the question of
how different physiological constraints within one genus may impact relative power
production in an MFC.
Third, given the diversity of S. oneidensis MR-1 carbon and energy metabolisms,
many different experiments were conducted to determine how this diversity would affect
power production in an MFC.

14
Several different electron donors were evaluated in MFCs using MR-1 as the
anode catalyst. Additionally, different MR-1 growth conditions were employed to
investigate if oxygen concentrations during pre-growth outside of the MFC, would effect
the current production.
Finally, MFC systems were utilized to demonstrate the effective splitting of
MR-1 metabolism. MR-1 is known to reduce solid substrates; however, the ability of
MR-1 to oxidize solid substrates is not well documented. To investigate these
phenomena, MR-1 was simultaneously introduced to the anode and cathode electrodes of
a MFC that did not feature any metallic catalysts. Many terminal electron acceptors were
utilized at the cathode, while lactate was provided as an electron donor at the anode, and
the limiting effects that energy metabolism has on power production in an MFC were
observed.
These combined results yield a greater understanding of how microbial
metabolism and bacterial electron transfer mechanisms impact MFC power production
and will hopefully lead to improved optimization of MFC systems for specific
applications.

15
Chapter 2: Current production and metal oxide reduction by
Shewanella oneidensis MR-1 wild type and mutants

The ability of certain bacteria to reduce metal oxides has been explored in-depth
over the last twenty years. Microbes capable of such activity, called dissimilatory metal-
reducing bacteria (DMRB) are of great interest to the science and engineering community
due to their roles in various geobiological phenomena, and for possible use in
bioremediation and biotechnology applications. Organisms that have been a focus of
interest in recent years include DMRB in genera such as Clostridium (Park, Kim et al.
2001), Geobacter (Bond and Lovley 2003; Holmes, Chaudhuri et al. 2006), Aeromonas
(Pham, Jung et al. 2003), Rhodoferax (Chaudhuri and Lovley 2003), Desulfobulbus
(Holmes, Bond et al. 2004) and Shewanella (Kim, Kim et al. 1999; Chang, Moon et al.
2006). All of these DMRB have also been shown to produce current in microbial fuel
cell (MFC) systems (Bond, Holmes et al. 2002; Logan, Hamelers et al. 2006). This study
focuses on the electron transfer mechanisms employed by Shewanella oneidensis MR-1,
a strain that was chosen as a model organism based on its metabolic versatility (Myers
and Nealson 1988), sequenced and annotated genome (Heidelberg, Paulsen et al. 2002),
and its genetic accessibility with regard to both generation and complementation of
mutants (Gorby, Yanina et al. 2006).
MR-1 and other members of the genus Shewanella were originally studied
because of their ability to couple the oxidation of organic compounds and H
2
to the
reduction of manganese and iron oxides (Myers and Nealson 1988; Lovley, Phillips et al.
1989; Arnold, Hoffmann et al. 1990; Myers and Nealson 1990; Luu and Ramsay 2003;

16
Lies, Hernandez et al. 2005; Ruebush, Brantley et al. 2006; Weber, Achenbach et al.
2006). The mechanisms responsible for metal oxide reduction are not fully understood,
but it is clear that a number of genes are involved (Nealson and Saffarini 1994; Beliaev
and Saffarini 1998; Myers and Myers 2000; Hernandez and Newman 2001; Myers and
Myers 2001; DiChristina, Moore et al. 2002; Myers and Myers 2002; Pitts, Dobbin et al.
2003; Yost, Hauser et al. 2003; Meyer, Tsapin et al. 2004; Beliaev, Klingeman et al.
2005; Gorby, Yanina et al. 2006). These include the mtrA, mtrB, mtrC (also known as
omcB), omcA and cymA genes. mtrA encodes a periplasmic decaheme c-type cytochrome
involved in Mn(IV) and Fe(III) reduction (Beliaev and Saffarini 1998; Beliaev, Saffarini
et al. 2001; Pitts, Dobbin et al. 2003; Beliaev, Klingeman et al. 2005; Bencheikh-Latmani,
Williams et al. 2005). mtrB encodes an outer membrane protein of 679 amino acids and
may play a role in the binding of metals during reduction (Beliaev and Saffarini 1998)
and/or may be required for the proper localization and insertion of outer membrane (OM)
cytochromes involved in direct electron transfer (Myers and Myers 2002). mtrC and
omcA encode OM decaheme c-type cytochromes that have been hypothesized to serve as
terminal Mn(IV) (Beliaev, Saffarini et al. 2001; Myers and Myers 2001; Myers and
Myers 2003; Myers and Myers 2003; Myers and Myers 2004) and Fe(III) reductases
(Beliaev, Saffarini et al. 2001; Gorby, Yanina et al. 2006), and to be involved in electron
transfer to electrodes in microbial fuel cells (MFCs) (Gorby, Yanina et al. 2006). cymA
encodes a cytoplasmic membrane-bound, tetraheme c-type cytochrome, that is involved
in mediating electron flow from the cytoplasm to several terminal reductases (including
OM c-type cytochromes) during anaerobic respiration (Myers and Myers 1997; Myers
and Myers 2000; Schwalb, Chapman et al. 2002).

17
The cymA and mtr genes described above represent only a small fraction of the
MR-1 genome that may be involved in energy metabolism. Forty-two possible c-type
cytochrome genes have been identified by sequence homology in the MR-1 genome
(Heidelberg, Paulsen et al. 2002; Meyer, Tsapin et al. 2004; Kolker, Picone et al. 2005)
and only a few of these have been characterized. In addition, some regulatory and
protein secretion genes, as well as hydrogenase and genes associated with type IV pilin
production have been implicated in energy metabolism (Thompson, Beliaev et al. 2002;
Saffarini, Schultz et al. 2003; Wan, VerBerkmoes et al. 2004; Gorby, Yanina et al. 2006).
For example, Thompson et al. (Thompson, Beliaev et al. 2002) suggested that the fur
(ferric uptake regulator) gene product is a global regulator that positively controls genes
involved in electron transport. Additionally, Wan et al. (Wan, VerBerkmoes et al. 2004)
proposed that omcA is a direct target of fur modulon activation. Saffarini et al. (Saffarini,
Schultz et al. 2003) identified crp that encodes the cyclic AMP (cAMP) receptor protein
(CRP) as a global regulator of anaerobic respiration including Fe(III) and Mn(IV)
reduction. An analysis of MR-1s twin arginine translocation (TAT) machinery suggests
that this system plays a key role in the secretion of a number of redox proteins to the
periplasm, some of which are either directly or indirectly involved in metal reduction
(Fredrickson, Beliaev et al. 2004).
The diverse energy metabolism of MR-1 is clearly governed by a number of
elements; however, the evidence thus far offers only a glimpse into the larger picture of
energy metabolism. Questions still remain as to which gene products are utilized for
electron transfer to solid substrates, and whether these gene products are universally

18
utilized for different types of metal oxide reduction and electron transfer to MFC
electrodes.
The mechanism(s) of electron transfer to solids by MR-1 was investigated using
the wild type (WT) strain and mutants deficient in c-type cytochromes and protein
secretion systems. The evaluation was based on the ability of each mutant to: 1) produce
current in an MFC system (Kim, Park et al. 2002; Chang, Moon et al. 2006); 2) reduce
Mn(IV)-oxides; and, 3) reduce solid phase iron(III)-oxides. In addition, two hydrogenase
mutants (hyaB::pDS3.1 and hydA::pDS3.1), three regulatory mutants (tatC::pDS3.1,
fur::pKNOCK and crp::Tn5), an outer-membrane protein mutant (ompW::pDS3.1), and a
quorum sensing mutant (luxS::pDS3.1) were evaluated for current production in an MFC.
These data represent the first identification of gene products specifically related to current
production in strain MR-1, and strongly suggest that strain MR-1 features multiple
pathways for electron transfer to different solid substrates.
2.1 Materials and Methods
2.1.1 Culture and growth conditions

A list of the mutants used in this study is provided in Table 1.

19
Table 1. Summary of the evaluated set of MR-1 cytochrome mutants, complements, and select
transport protein mutants.
Functional
Category
ORF
targeted
Gene product deleted
SO0479 Octaheme cytochrome c; involved with the sulfur cycle
SO0610 Ubiquinol-cytochrome c reductase (petC)
SO0714 Periplasmic monoheme cytochrome c4; involved in sulfite oxidation
SO0716 Periplasmic monoheme cytochrome c (sorB); involved in sulfite oxidation
SO0717 Periplasmic monoheme cytochrome c4; involved in sulfite oxidation
SO0845 Diheme cytochrome c (napB); involved in nitrate reduction
SO0939 Split-soret diheme cytochrome c
SO0970 Fumarate reductase tetraheme cytochrome c (fccA)
SO1413 Split tetraheme flavocytochrome c
SO1421 Fumarate reductase flavoprotein subunit (ifcA-1)
SO1427 Periplasmic decaheme cytochrome c, involved in DMSO reduction (dmsC)
SO1659 OmcA-like decaheme cytochrome c
SO1776 Outer membrane protein precursor (mtrB), involved in metal oxide reduction
SO1777 Periplasmic decaheme cytochrome c (mtrA), involved in metal oxide reduction
SO1778/SO1779 Decaheme cytochrome c complex (mtrC/omcA), involved in metal oxide reduction
SO1780 Outer membrane decaheme cytochrome c (mtrF)
SO1782 Periplasmic decaheme cytochrome c (mtrD)
SO2098 Quinone-reactive Ni/Fe-hydrogenase, large subunit (hyaB)
SO2361 Cytochrome c oxidase, cbb3-type, subunit III (ccoP)
SO2363 Cytochrome c oxidase, cbb3-type, subunit II (ccoO)
SO2727 Small tetraheme cytochrome c (cctA)
SO2930
Cytochrome c with carbohydrate binding domain; involved with the sulphur
cycle
SO2931 Cytochrome c lipoprotein; involved with the sulphur cycle
SO3300 Split tetraheme flavocytochrome c
SO3420 Monoheme cytochrome c'
SO3920 Periplasmic Fe-hydrogenase, small subunit (hydA)
SO3980 Cytochrome c552 nitrite reductase (nrfA); involved with nitrite reduction
SO4047 SoxA-like diheme c; involved with the sulphur cycle
SO4048 Diheme c4
SO4142 Monoheme cytochrome c
SO4144 Octaheme cytochrome c; involved in tetrathionate reduction
SO4360 MtrA-like decaheme cytochrome c
SO4484 Monoheme cytochrome c (Shp)
SO4485 Diheme cytochrome c
SO4572 Triheme cytochrome c
SO4591
Tetraheme cytochrome c (cymA); involved in anaerobic respiration (except
TMAO)
SO4606 Diheme cytochrome c oxidase, subunit II (CyoA); involved in aerobic respiration
Energy Metabolism:
Electron Transport
SO4666 Cytochrome c (cytcB)
SO0166 General secretion protein (gspG)
SO0169 General secretion protein (gspD)
Protein Secretion
SO0414 Type 4 prepilin-like proteins leader peptide processing enzyme (pilD)
Protein Transport SO4204 Secretion-independent periplasmic translocation protein (tatC)
Quorum Sensing SO1101 Autoinducer-2 production protein (luxS)
Unknown SO1673 Outer-membrane protein (ompw); putative
SO1937 Ferric uptake regulation protein (fur) Regulatory: DNA
interactions SO0624 Catabolic gene activator (crp)

20
The WT, cytochrome mutant, and protein secretion mutant cultures were grown
aerobically in batch culture using a defined medium containing: 18 mM lactate as the sole
carbon source, 50 mM PIPES, 7.5 mM NaOH, 28 mM NH
4
Cl, 1.3 mM KCl, 4.3 mM
NaH
2
PO
4
·H
2
O, 100 mM NaCl, and 10 mL/L each of vitamin solution (Kieft, Fredrickson
et al. 1999), amino acid solution and trace mineral stock solutions. The amino acid
solution (pH 7.0) contained: 2 µg/L L-glutamic acid, 2 µg/L L-arginine, 2 µg/L DL-serine.
The trace mineral solution (pH 7.0) contained: 78.49 µM C
6
H
9
NO
3
, 121.71 µM
MgSO
4
·7H
2
O, 29.58 µM MnSO
4
·H
2
O, 171.12 µM NaCl, 3.60 µM FeSO
4
·7H
2
O, 6.80 µM
CaCl
2
·2H
2
O, 4.20 µM CoCl
2
·6H
2
O, 9.54 µM ZnCl
2
, 0.40 µM CuSO
4
·5H
2
O, 0.21 µM
AlK(SO
4
)
2
·12H
2
O, 1.62 µM H
3
BO
3
, 1.03 µM Na
2
MoO
4
·2H
2
O, 1.01 µM NiCl
2
·6H
2
O, and
0.76 µM Na
2
WO
4
·2H
2
O. Cultures were grown at 30°C and agitated at rate of 140 rpm
until the late stationary phase was achieved.
Colleagues at the Korea Institute of Science and Technology (KIST) employed a
different growth strategy for MR-1 insertional mutants including hyaB::pDS3.1,
hydA::pDS3.1, luxS::pDS3.1, fur::pKNOCK, crp::Tn5, ompW::pDS3.1 and tatC::pDS3.1.
These mutants (and WT) were grown anaerobically using a phosphate-buffered basal
media (PBBM) containing lactate (30 mM) as an electron donor, fumarate (60 mM) as an
electron acceptor and 1 g/L yeast extract (pH 7.0) (Kim, Hyun et al. 2005). The cells
were washed twice in a phosphate buffer (50 mM PO
4
, 100 mM NaCl, pH 7.0), and
injected into the MFC anode compartment.
The final OD
600
of every culture (for each growth method) was measured and
used to calculate an experimental dilution to OD
600
0.4. The appropriate volume of cells

21
was then injected into each experimental setup such that approximately 2 x 10
9
cells/mL
were present for every evaluation.
2.1.2 Mutagenesis

Several S. oneidensis MR-1 deletion mutants were constructed by colleagues from
the Pacific Northwest National Laboratory (PNNL) using allele replacement using a two-
step homologous recombination method as described by Marshall et al (Marshall, Beliaev
et al. 2006) and Gorby et al. (Gorby, Yanina et al. 2006). The remaining deletion
mutants were constructed at Oak Ridge National Laboratory by a cre-lox recombination
method described by Marx and Lidstrom (Marx and Lidstrom 2002). Additionally,
complemented strains of the in-frame deletion mutants mtrA, mtrB, mtrC and omcA were
constructed at PNNL using the cloning vector pBBR1MCS-5. The genes mtrA, mtrB and
mtrC were constitutively expressed under the lacZ promoter while the omcA expression
was controlled by its natural promoter.
Mutants deficient in type IV prepilin peptidase (PilD; SO0166) and type II
secretion system proteins (GspG and GspD) were generated at the University of
Wisconsin-Milwaukee using minihimarRB-1 transposon mutagenesis as described by
Bouhenni et al. (Bouhenni, Gehrke et al. 2005) and Beliaev and Saffarini (Beliaev and
Saffarini 1998; Saffarini, Schultz et al. 2003; Gorby, Yanina et al. 2006). The ferric
uptake regulation mutant, SO1937::pKNOCK (fur::pKNOCK), was created at PNNL by
integrating the suicide plasmid pKNOCK-K
r
into the chromosomal fur locus as described
by Thompson (Thompson, Beliaev et al. 2002) and Yost (Yost, Hauser et al. 2003). The
cAMP receptor protein insertional mutant, SO0624::Tn5 (crp::Tn5), was constructed at

22
the University of Wisconsin-Milwaukee using a Tn5 insertion in crp according to
Saffarini et al. (Saffarini, Schultz et al. 2003). The suicide plasmid pDS3.1 (Wan,
VerBerkmoes et al. 2004; Marshall, Beliaev et al. 2006) was used at PNNL to construct
the insertional mutants SO2098::pDS3.1 (hyaB::pDS3.1), SO3920::pDS3.1
(hydA::pDS3.1), SO1101::pDS3.1 (luxS::pDS3.1), SO1673::pDS3.1 (ompW::pDS3.1)
and SO4204::pDS3.1 (tatC::pDS3.1).
2.1.3 MFC Experiments

Current production was observed using dual compartment MFCs of the type
shown in Figure 2.

Figure 2. Dual compartment microbial fuel cell diagram. Bacteria are inoculated into the anode
compartment, attach to the graphite felt electrode and begin transferring electrons. Electrons are
conducted through the anode electrode and across the external circuit to the cathode electrode. The
cathode electrode is graphite felt that has been electroplated with platinum, the catalyst driving the
reduction of oxygen to water. The anode and cathode compartments are physically separated by a
proton-conductive membrane that facilitates the transfer of protons from the anode to the cathode,
completing the cell reaction. The cell voltage, V, is measured across an external load of resistance, R,
and current is calculated as I = V/R.

23
The MFCs were assembled using proton-exchange membranes (Nafion
®
424,
DuPont) and electrodes constructed from graphite felt (GF-S6-06, Electrolytica) bonded
to platinum wire (0.3 mm, Alfa-Aesar) (Gorby, Yanina et al. 2006). The cathode
electrodes were electroplated with a platinum catalyst over the entire surface area at a
loading of 0.15 mg/cm
2
to drive the oxygen reduction reaction, while the anode
electrodes were exposed to MR-1 (the catalyst for lactate oxidation). A sodium-PIPES
buffer (100 mM NaCl, 50 mM PIPES, pH 7.0) was used in both anode and cathode
compartments as the diluting solution for bacteria. Each compartment had a working
volume of 25 mL. Anaerobic conditions were maintained at the anode by continuously
flushing the compartment with sterile filtered N
2
gas at a rate of 20 mL/min. Aerobic
conditions were maintained at the cathode by continuously flushing the compartment
with air at a rate of 40 mL/min. Each bacterial strain was evaluated in triplicate and cell
voltage, (V), was recorded every five minutes across a 10 ohm resistor, (R), by a high-
impedance, digital multi-meter (Model 2700, Keithley Instruments). Baseline cell
voltages were collected under two conditions: 1) buffer only at the anode; and, 2) buffer
and bacteria at the anode, but no additional lactate. Each baseline voltage was monitored
for one hour prior to changing the anodic condition. Lactate (2 mM) was added to each
MFC system after the bacteria were added and a baseline voltage was observed. Two
more lactate additions occurred thereafter, depending on when the cell voltage dropped to
baseline levels (approximately every 24 hours). MFCs were operated in batch after the
addition of bacteria (no solution exchanges) over a period of three days. A lactate
concentration of 2 mM was chosen based on previous experiments demonstrating that a

24
higher concentration of lactate (i.e., 20 mM) yields higher charge, but the maximum
current density remained the same.
Current (I) was calculated according to I = V/R and maximum current densities
were calculated using the maximum current values that remained constant for
approximately four hours (corresponding to each lactate feeding), divided by the total
apparent surface area of the electrode (20 cm
2
).
2.1.4 Manganese(IV) oxide

Manganese(IV)-oxide reduction data was collected in collaboration with Anna
Obraztsova, Carter Sturm, and Prithviraj Chellamuthu. Mn(IV)-oxide was investigated
using autoclaved, anaerobic, δ-manganese oxide (δMnO
2
) that was prepared according to
Burdige and Nealson (Burdige and Nealson 1985). Duplicate manganese oxide reduction
experiments were conducted in anaerobic serum bottles using 20 mM lactate, 300 µM
δMnO
2
, sodium-PIPES buffer (described above) and an appropriate volume of bacterial
cell suspension. Experimental controls were prepared with: 1) cells and δMnO
2
, but
without lactate; and 2) lactate and δMnO
2
, but without cells. A colorimetric assay with
Leucoberbelin Blue (LBB) (Krumbein and Altmann 1973) was used to quantify the
amount of Mn(IV) present in each bottle immediately after inoculation and after twenty-
four and forty-eight hours of exposure to bacteria. Briefly, a 0.1 mL volume of
subsample was added to 0.9 mL of LBB solution and incubated for 15 minutes in the
dark. Color changes in the LBB solution were measured at a fixed wavelength of 690 nm.
Mn(IV) concentrations were determined based on a standard curve generated from
known concentrations of KMnO
4
. The percentage of Mn(IV) depletion was calculated

25
based on measured Mn(IV) concentrations after twenty-four and forty-eight hours of
exposure.
2.1.5 Solid-phase iron(III) oxide

Solid ferric oxide reduction was investigated using autoclaved anaerobic hydrous
ferric oxide (HFO), prepared according to the protocol of Cornell and Schwertmann
(Cornell and Schwertmann 1996). This form of HFO is not stabilized by silica and
becomes more crystalline with time; therefore, all strains were evaluated within a period
of three days to ensure comparable results between strains. X-ray diffraction spectra,
provided by Everett Salas, showed that the solid phase iron was a combination of goethite,
hematite and nano-particles of HFO that we refer to as the HFO mixture or HFOM.
Triplicate HFOM reduction experiments were performed in sterile, anaerobic serum
bottles. Each bottle contained 20 mM of HFOM as the sole electron acceptor, 20 mM of
lactate as the sole electron donor and the sodium-PIPES buffer (described above).
Experimental controls were prepared with: 1) cells and HFOM, but without lactate; and
2) lactate and HFOM, but without cells. Ferrozine, which develops a purple color in the
presence of Fe(II) was used to monitor the production of Fe(II) using a colorimetric
ferrozine assay (Stookey 1970). Briefly, a subsample from each bottle was acidified and
added to the ferrozine solution. Color changes in the ferrozine solution were measured at
a fixed wavelength of 562 nm. Fe(II) concentrations were determined based on a
standard curve generated from known concentrations of FeCl
2
.

26
2.1.6 Soluble iron(III)

Soluble ferric reduction was investigated in collaboration with Carter Sturm using
the iron(III) chelate, ferric nitrilotriacetic acid (Fe(III)-NTA) prepared using the acid
form of nitrilotriacetic acid. Fe(III)-NTA reduction experiments were performed in
duplicate. Anaerobic Fe(III)-NTA (20mM) was added to sterile, anaerobic serum bottles
in addition to 20mM of lactate, sodium-PIPES buffer and appropriate volumes of bacteria.
Experimental controls were prepared with: 1) cells and Fe(III)-NTA, but without lactate;
and 2) lactate and Fe(III)-NTA, but no cells. Fe(II) production was monitored using the
standard ferrozine assay (Stookey 1970).
2.1.7 Scanning Electron Microscopy (SEM) sample preparation

Select MFC electrodes were removed at the end of the experiment and
subsamples were fixed in 2.5% gluteraldehyde solution and ethanol dehydrated according
to the protocol outlined by Gorby et al. (Gorby, Yanina et al. 2006).
2.2 Results
2.2.1 MFC Experiments

The average current density obtained for S. oneidensis MR-1 WT and mutants is
shown in Figure 3.

27
0
5
10
15
20
25
30
35
40
45
50
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1778/1779 (∆mtrC/∆omcA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
current density (µA/cm
2
)
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
Control (no bacteria)
0
5
10
15
20
25
30
35
40
45
50
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1778/1779 (∆mtrC/∆omcA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
current density (µA/cm
2
)
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
Control (no bacteria)

Figure 3. Current density values for MR-1 WT, cytochrome deletions and protein secretion mutants.
Averages and standard deviations were obtained using the peak current density values
corresponding to each lactate injection for triplicate experiments. Maximum current density was
determined to be the highest level of current density that remained constant for at least three hours.
Averages and standard deviations of the maximum current density were calculated using these data
values (between one-hundred to two-hundred data points were utilized).
It is apparent from Figure 3 that only a select few of the deletion mutants were
significantly diminished in their abilities to produce current in an MFC system relative to
the WT. Specifically, the mutants lacking mtrA, mtrB, mtrC/omcA and cymA generated
less than 20% of the current produced by the WT. For example, ∆mtrA produced 0.35 ±
0.05 µA/cm
2
as compared to 13.8 ± 1.37 µA/cm
2
(382 ± 76.7 µW/m
2
or 3.03 x 10
4
± 6.08

28
x 10
3
µW/m
3
given the entire volume of the non-conductive chamber)

produced by WT.
However, the complemented strains of the mtrA, mtrB and mtrC/omcA deletion mutants
resulted in full recovery or enhanced current producing abilities; i.e., the complemented
mtrA and mtrB mutants produced 13.9 ± 4.23 µA/cm
2
and 16.9 ± 2.27 µA/cm
2
,
respectively.

Additionally, a 35% increase in current production (relative to the WT) was
observed when the mtrC cytochrome is over-expressed in the WT strain (21.3 ± 5.02
µA/cm
2
) (Figure 4).
0
5
10
15
20
25
30
35
40
45
50
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
current density (µA/cm
2
)
0
5
10
15
20
25
30
35
40
45
50
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
current density (µA/cm
2
)

Figure 4. Current density values for MR-1 WT (black) and selected cytochrome mutants (grey) with
their complementations (white). Averages and standard deviations were obtained using the peak
current density values corresponding to each lactate injection for triplicate experiments. Maximum
current density was determined to be the highest level of current density that remained constant for
at least three hours. Averages and standard deviations of the maximum current density were
calculated using these data values (between one-hundred to two-hundred data points were utilized).
Several cytochrome deletion mutants also showed higher current values than the
WT strain (at least 20% higher) (Figure 2a). These are ∆SO0714, ∆SO0716 (∆sorB) and

29
∆SO0717 (monoheme cytochromes predicted to be involved in sulfite oxidation),
∆SO0845 (diheme c-type cytochrome predicted to be involved in nitrate reduction,
∆napB), ∆SO1427 (periplasmic decaheme c-type cytochrome predicted to be involved in
DMSO reduction, ∆dmsC), ∆SO2930 (c-type cytochrome with carbohydrate binding
domain predicted to be involved with the sulphur cycle), and ∆SO3980 (cytochrome
reductase predicted to be involved with nitrite reduction, ∆nrfA).
The cytochrome mutants were not the only group deficient in current production
relative to the WT strain (Figure 3). The mutants deficient in type IV prepilin peptidase
or type II secretion (∆pilD, ∆gspD and ∆gspG) and three other mutants, fur::pKNOCK,
crp::Tn5 and tatC::pDS3.1 mutants were all severely limited in current producing
abilities, showing less than 20% of WT values (data not shown).
A select number of MFC electrodes were examined by scanning electron
microscopy (SEM) to evaluate the distribution of cells on the electrode surfaces after
current data were collected. It is apparent from these images that the electrode exposed
to the WT strain (Figure 5) featured much greater surface coverage (with intact cells and
what appears to be a developing biofilm) than ∆pilD (Figure 6) or ∆mtrC/∆omcA (Figure
7). Additional SEM images can be found in Appendix A.

30

Figure 5. SEM image of graphite anode fibers used during the MFC evaluations of MR-1

Figure 6. SEM images of graphite anode fibers used during the MFC evaluations of the MR-1 ∆pilD
mutant.

31

Figure 7. SEM images of graphite anode fibers used during the MFC evaluations of the MR-1
∆mtrC/∆omcA mutant.
2.2.2 Manganese(IV) oxide

The average percentages of Mn(IV)-oxide reduced by the MR-1 WT, cytochrome
and protein secretion mutants, and cytochrome complemented strains were investigated.
As with current production, ∆mtrA, ∆mtrB and ∆cymA were limited in their ability to
reduce Mn(IV)-oxide relative to the WT strain (Figure 8).

32
0
10
20
30
40
50
60
70
80
90
100
∆SO1778/1779 (∆omcA/∆mtrC)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
%[Mn(IV)] reduced after 24 hrs
Control (no bacteria)
0
10
20
30
40
50
60
70
80
90
100
∆SO1778/1779 (∆omcA/∆mtrC)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
%[Mn(IV)] reduced after 24 hrs
Control (no bacteria)

Figure 8. Average percentage of Mn(IV)-oxide reduced after twenty-four hours of exposure to MR-1
wild-type (black), cytochrome deletion and proteins secretion mutants (grey).
For example, results after twenty-four hours show no Mn(IV)-oxide reduction by
∆mtrA or ∆mtrB (Figure 8). However, after forty-eight hours ∆mtrB reduced 17.3 ±
2.39% of the starting concentration of Mn(IV)-oxide, while WT reduced 36.7 ± 0.39%.
The ∆gspD, ∆gspG and ∆pilD were also affected in their ability to reduce Mn(IV)-oxide,
however not as significantly as ∆mtrA and ∆mtrB. Unlike the results obtained for current
production, the ∆mtrC/∆omcA double mutant was able to reduce Mn(IV)-oxides after

33
twenty-four hours (29.9 ± 0.66%) and then after forty-eight hours the reduction exceeded
that of WT (43.8 ± 0.30%). Several strains also exhibited higher Mn(IV)-oxide reduction
than WT including all of the cytochrome mutants involved with sulphur redox reactions
(∆SO0479, ∆SO0714, ∆sorB, ∆SO0717, ∆SO2930, ∆SO2931 and ∆SO4047), nitrate
reduction (∆napB and ∆nrfA), DMSO reduction (∆dmsC), tetrathionate reduction
(∆SO4144) and many mutants that have unknown functions in MR-1s energy
metabolism (∆cctA, ∆SO3420, ∆SO4048, ∆SO4360, ∆shp, ∆SO4485 and ∆cytcB). The
∆mtrA and ∆mtrB complemented mutants were able to reduce Mn(IV)-oxides as well as
the WT. However, ∆mtrC/∆omcA showed equal ability to reduce Mn(IV)-oxide relative
to the WT regardless of complementation (Figure 9). The deletion of the mtrC/omcA
gene did not appear to have an effect on Mn(IV)-oxide reduction in MR-1.
0
10
20
30
40
50
60
70
80
90
100
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
%[Mn(IV)] reduced after 24 hrs
0
10
20
30
40
50
60
70
80
90
100
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
%[Mn(IV)] reduced after 24 hrs

Figure 9. Average percentage of Mn(IV)-oxide reduced after twenty-four hours of exposure to MR-1
wild-type (black), selected cytochrome deletion mutants (grey) and their respective complements
(white).

34
2.2.3 Solid-phase Fe(III)-oxide

The average concentrations of Fe(II) resulting from the reduction of HFOM by
the MR-1 WT, mutants are shown in Figure 10.
0
100
200
300
400
500
600
700
800
900
1000
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown Function
Anaerobic respiration (except TMAO)
[Fe(II)] after 24 hrs (µM)
Control (no bacteria)
∆SO1778/1779 (∆omcA/∆mtrC)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
0
100
200
300
400
500
600
700
800
900
1000
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown Function
Anaerobic respiration (except TMAO)
[Fe(II)] after 24 hrs (µM)
Control (no bacteria)
∆SO1778/1779 (∆omcA/∆mtrC)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control

Figure 10. Average Fe(II) concentration resulting from solid Fe(III)-oxide (HFOM) reduction after
twenty-four hours of exposure to MR-1 WT (black), cytochrome deletions, and protein secretion
mutants (grey). Averages and standard deviations were calculated based on the measured Fe(II)
concentrations from triplicate experiments.

35
These data show that ∆mtrA, ∆mtrB, ∆omcA/∆mtrC, ∆cymA, ∆SO4144,
∆SO4572 (a triheme c-type cytochrome), ∆gspG, ∆gspD and ∆pilD were limited in
HFOM reduction relative to the WT. For example, ∆mtrC/∆omcA produced 68.4 ± 5.08
µM and WT produced 170 ± 32.7 µM of Fe(II) after twenty-four hours. Data collected
after forty-eight hours showed a slight increase in Fe(II) production from ∆mtrC/∆omcA
(98.7 ± 22.7 µM), however this amount was much less than WT (245 ± 12.6 µM).
Other mutants, such as ∆napB (345 ± 68.0 µM), ∆ifcA-1 (380 ± 42.8 µM), ∆mtrF
(368 ± 0.00 µM), ∆mtrD (362 ± 22.7 µM) and ∆SO3420 (492 ± 25.2 µM) had even
greater capacity to reduce HFOM than the WT. For example, the concentration of Fe(II)
produced by ∆ifcA-1 and ∆mtrF was two- to three-times higher than WT after twenty-
four hours. The ∆napB mutant was also enhanced in its ability to produce current in an
MFC, but decreased in Mn(IV) reducing ability. The ∆ifcA-1 mutant is deficient in the
ability to produce a tetraheme c-type flavocytochrome, ∆mtrF is lacking the ability to
code an outer-membrane decaheme c-type cytochrome, ∆mtrD is lacking ability to code a
periplasmic decaheme c-type cytochrome, and ∆SO3420 is deficient in the ability to
produce a monoheme cytochrome c.
The ∆mtrA, ∆mtrB and ∆mtrC/∆omcA complemented strains made a full recovery
of HFOM reducing ability relative to the WT strain, shown in Figure 11.

36
0
100
200
300
400
500
600
700
800
900
1000
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
[Fe(II)] after 24 hrs (µM)
0
100
200
300
400
500
600
700
800
900
1000
Wildtype
WT + mtrC
∆mtrB
∆mtrB complement
∆mtrA
∆mtrA complement
∆mtrC complement
∆omcA complement
∆mtrC/∆omcA
Control
mtrB deletion
mutant and
complement
Wild type
Over-expressed mtrC in WT
mtrA deletion
mutant and
complement
mtrC/omcA
deletion mutant
and complements
Control (no bacteria)
[Fe(II)] after 24 hrs (µM)

Figure 11. Average Fe(II) concentration resulting from solid Fe(III)-oxide (HFOM) reduction after
twenty-four hours of exposure to MR-1 WT (black), select cytochrome deletions (grey), and their
respective complementations (white). Averages and standard deviations were calculated based on
the measured Fe(II) concentrations from triplicate experiments.
2.2.4 Soluble iron(III)

The average concentration of Fe(II) resulting from the reduction of soluble
Fe(III)-NTA by the MR-1 WT and deletion mutants is shown in Figure 12.

37
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[Fe(II)] after 24 hrs (mM)
∆SO1778/1779 (∆mtrC/∆omcA)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
Control (no bacteria)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[Fe(II)] after 24 hrs (mM)
∆SO1778/1779 (∆mtrC/∆omcA)
Wildtype
∆SO0479
∆SO0610 (∆petC)
∆SO0714
∆SO0716 (∆sorB)
∆SO0717
∆SO0845 (∆napB)
∆SO0939
∆SO0970 (∆fccA)
∆SO1413
∆SO1421 (∆ifcA-1)
∆SO1427 (∆dmsC)
∆SO1659
∆SO1776 (∆mtrB)
∆SO1777 (∆mtrA)
∆SO1780 (∆mtrF)
∆SO1782 (∆mtrD)
∆SO2361 (∆ccoP)
∆SO2363 (∆ccoO)
∆SO2727 (∆cctA)
∆SO2930
∆SO2931
∆SO3300
∆SO3420
∆SO3980 (∆nrfA)
∆SO4047
∆SO4048
∆SO4142
∆SO4144
∆SO4360
∆SO4484 (∆shp)
∆SO4485
∆SO4572
∆SO4591 (∆cymA)
∆SO4606 (∆cyoA)
∆SO4666 (∆cytcB)
∆SO0166 (∆gspG)
∆SO0169 (∆gspD)
∆SO0414 (∆pilD)
control
Metal oxide reduction
Sulfur redox
Nitrate reduction
Fumarate reduction
DMSO reduction
Tetrathionate reduction
Ubiquinol reduction
Oxidase
Protein secretion
Unknown function
Anaerobic respiration (except TMAO)
Control (no bacteria)

Figure 12. Average Fe(II) concentration resulting from soluble Fe(III)-NTA reduction after twenty-
four hours of exposure to MR-1 WT (black), cytochrome deletions, and protein secretion mutants
(grey). Averages and standard deviations were calculated based on the measured Fe(II)
concentrations from duplicate experiments.
It has long been known that the rate of Fe(II) reduction is much greater for soluble
Fe(III)-NTA experiments than for the solid HFOM experiments (Arnold, Olson et al.
1986). This data set also shows much more variability in the number of mutants that are
inhibited and/or enhanced in Fe(III)-NTA reduction relative to the solid Fe(III)-oxide and
current production experiments. For example, those mutants that were limited in their

38
ability to reduce Fe(III)-NTA relative to WT (by about half) include ∆SO0479, ∆SO0939,
∆SO1413, ∆mtrC/∆omcA, ∆mtrf, ∆ccoO, ∆cctA, ∆SO3300, ∆nrfA, ∆SO4360, ∆SO4585,
∆SO4572, ∆gspG and ∆pilD. Those mutants that were enhanced in Fe(III)-NTA
reducing capabilities (between 2.5 and 7-times) include: ∆SO0714, ∆sorB, ∆SO0717,
∆napB, ∆fccA, ∆ifcA-1, ∆dmsC, ∆SO1659, ∆mtrB, ∆mtrD, ∆ccoP, ∆SO2930, ∆SO2931,
∆cyoA and ∆cyctcB (see Table 1 for a description of the predicted function of each
mutant). It is of note that while ∆mtrA and ∆mtrB were unaffected in Fe(III)-NTA
reduction ability, their respective complements showed a large increase (up to 6-times
higher) in their ability to reduce soluble iron oxide relative to the WT (Figure 13).
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[Fe(II)] after 24 hrs (mM)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[Fe(II)] after 24 hrs (mM)

Figure 13. Average Fe(II) concentration resulting from soluble Fe(III)-NTA reduction after twenty-
four hours of exposure to MR-1 WT (black), selected cytochrome deletions (grey), and their
respective complementations (white). Averages and standard deviations were calculated based on
the measured Fe(II) concentrations from duplicate experiments.

39
It is also interesting to note that the Fe(III)-NTA reduction experiments were
unintentionally carried out using the acid form of NTA, which was close to pH 4.0. This
low pH may have negatively affected or exaggerated the performance of the WT and
mutants in terms of soluble Fe(III) reduction.
To better understand these effects, an additional study was conducted in
collaboration with Carter Sturm and Prithviraj Chellamuthu using select mutants and the
WT with three different forms of soluble ferric iron including Fe(III)-citrate (20 mM, pH
5.6), the sodium form of NTA, Fe(III)-NTA(Na
+
) (20 mM, pH 7.1), and the acid form of
NTA, Fe(III)-NTA(H
+
) (20 mM, pH 4.2). The same methods as described above were
employed to evaluate the soluble Fe(III)-reduction capabilities for each mutant and WT.
Initial pH values were determined after the appropriate volumes of soluble ferric iron,
buffer, and lactate were added to anaerobic bottles at the beginning to the experiment.
Final pH values and cell densities, for each bacterial culture and soluble form of ferric
iron, were measured after four days of exposure.
Experiments performed using ∆mtrA showed that the mtrA mutant consistently
reduced less soluble ferric iron than WT when exposed to Fe(III)-citrate and Fe(III)-
NTA(Na
+
); however, after forty-eight hours of exposure to Fe(III)-citrate, ∆mtrA was
very near WT levels of reduction. A different relationship was observed for the Fe(III)-
NTA(H
+
), in that ∆mtrA reduced as much or more of the Fe(III)-NTA(H
+
) (Figure 14).

40
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
a) 24 hrs b) 48 hrs
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
[Fe(II)] after 48 hrs (mM)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
0
2
4
6
8
10
12
14
16
18
20
Wildtype
∆mtrA
Wildtype
∆mtrA
Wildtype
∆mtrA
[Fe(II)] after 24 hrs (mM)
Fe(III)-citrate Fe(III)-NTA(Na
+
) Fe(III)-NTA(H
+
)
a) 24 hrs b) 48 hrs

Figure 14. Average Fe(II) concentration resulting from soluble Fe(III) reduction after a) twenty-four
hours and b) forty-eight hours of exposure to MR-1 WT (black) and mtrA mutant (grey). Averages
and standard deviations were calculated based on the measured Fe(II) concentrations from triplicate
experiments. Starting pH values for Fe(III)-citrate, Fe(III)-NTA(Na+) and Fe(III)-NTA(H+) were 5.6,
7.1 and 4.1, respectively.

The observed trend between ∆mtrA and WT remained consistent, but while the
Fe(III)-citrate and Fe(III)-NTA(H
+
) cultures showed little activity beyond forty-eight
hours, the Fe(III)-NTA(Na
+
) cultures continued to reduce. After ninety-six hours, the
Fe(II) concentrations for ∆mtrA and WT were very similar in Fe(III)-NTA(Na
+
) (Figure
15).

41
0
2
4
6
8
10
12
14
16
18
20
0 102030405060 708090 100
[Fe(II)] produced (mM)
MR1 WT
∆mtrA
Fe(III)-citrate
Fe(III)-NTA(Na
+
)
Fe(III)-NTA(H
+
)
Time (hrs)
0
2
4
6
8
10
12
14
16
18
20
0 102030405060 708090 100
[Fe(II)] produced (mM)
MR1 WT
∆mtrA
Fe(III)-citrate
Fe(III)-NTA(Na
+
)
Fe(III)-NTA(H
+
)
MR1 WT
∆mtrA
Fe(III)-citrate
Fe(III)-NTA(Na
+
)
Fe(III)-NTA(H
+
)
Time (hrs)

Figure 15. Average Fe(II) concentrations resulting from soluble Fe(III) reduction over a period of
ninety-six hours. Averages and standard deviations were calculated based on the measured Fe(II)
concentrations from triplicate experiments. Starting pH values for Fe(III)-citrate, Fe(III)-NTA(Na+)
and Fe(III)-NTA(H+) were 5.6, 7.1 and 4.1, respectively.
The pH values shifted uniquely for each form of soluble ferric iron media over the
ninety-six hour period. All of the different ferric iron media became more basic after
exposure to both WT and ∆mtrA cultures. The pH of the Fe(III)-citrate media for both
WT and ∆mtrA increased from 5.6 to 6.3; and the pH of the Fe(III)-NTA(Na
+
) media for
WT and ∆mtrA increased from 7.1 to 7.3, and 7.2 to 7.4, respectively. The pH increased
for the Fe(III)-NTA(H
+
) cultures as well, however ∆mtrA was more effective at
increasing the pH than WT. The WT culture changed the Fe(III)-NTA(H
+
) media from
pH 4.1 to 4.6, while ∆mtrA changed the Fe(III)-NTA(H
+
) media from 4.1 to 5.4. A
summary of these results is presented in Table 2.

42
Table 2. pH values for ∆mtrA and WT at zero hours and ninety-six hours of exposure to multiple
forms of soluble ferric iron.
∆mtrA (pH) WT (pH)
0 hrs 96 hrs 0 hrs 96 hrs
Fe(III)-citrate 5.6 6.3 5.6 6.3
Fe(III)-NTA(Na
+
) 7.2 7.4 7.1 7.3
Fe(III)-NTA(H
+
) 4.1 5.4 4.1 4.6

In addition to changes in pH values, the final cell counts for both Fe(III)-citrate
and Fe(III)-NTA(Na
+
) showed differences between the WT and ∆mtrA cultures. In both
ferric iron media ∆mtrA had cell densities of at least one order of magnitude higher than
WT after four days of exposure. The cell counts for Fe(III)-NTA(H
+
) indicate that the
cells could not survive multiple days at such a low pH (Table 3).
Table 3. Cell densities after inoculation and after 96 hours of exposure to different forms of soluble
iron(III).
∆mtrA (cells/mL) WT (cells/mL)

0 hrs 96 hrs 0 hrs 96 hrs
Fe(III)-citrate 1 x 10
9
8 x 10
5
1 x 10
9
2 x 10
3
Fe(III)-NTA(Na
+
) 1 x 10
9
3 x 10
6
1 x 10
9
1 x 10
5
Fe(III)-NTA(H
+
) 1 x 10
9
0 1 x 10
9
0

Early experiments conducted with a higher pH (pH 6.1) and lower concentration
(2.5 mM) Fe(III)-citrate were compared to the results from the Fe(III)-NTA(H
+
) (20mM,
pH 4.2) experiments using ∆mtrA, ∆mtrC/∆omcA, ∆mtrF, ∆ifcA-1, ∆dmsC, and WT.
The mutants ∆ifcA-1 and ∆dmsC were able to reduce nearly half of the concentration of
Fe(III)-NTA(H
+
) after twenty-four hours of exposure, but were very limited in their

43
ability to reduce Fe(III)-citrate in the same time period. Conversely, WT and
∆mtrC/∆omcA were able to reduce nearly all of the Fe(III)-citrate after twenty-four hours
of exposure; and ∆mtrA was able to reduce most of the Fe(III)-citrate after seventy-two
hours of exposure. The mutant ∆mtrF demonstrated very little ability to reduce either the
Fe(III)-citrate or the Fe(III)-NTA(H
+
) (Figure 16).
0
0.5
1
1.5
2
2.5
3
3.5
4
0 10203040506070 80
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
0.5
1
1.5
2
2.5
3
3.5
4
0 10203040506070 80
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
1
2
3
4
5
6
7
8
9
10
0 1020304050607080
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
1
2
3
4
5
6
7
8
9
10
0 1020304050607080
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
a) b)
0
0.5
1
1.5
2
2.5
3
3.5
4
0 10203040506070 80
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
0.5
1
1.5
2
2.5
3
3.5
4
0 10203040506070 80
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
1
2
3
4
5
6
7
8
9
10
0 1020304050607080
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
0
1
2
3
4
5
6
7
8
9
10
0 1020304050607080
[Fe(II)] produced (mM)
Time (hrs)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
∆mtrF
MR1 WT
∆mtrC/omcA
∆mtrA
∆ifcA-1
∆dmsC
Control (no bacteria)
a) b)

Figure 16. Average Fe(II) concentration resulting from the reduction of a) 2.5 mM Fe(III)-citrate,
and b) 20 mM Fe(III)-NTA(H+). Averages were calculated based on the measured Fe(II)
concentrations from duplicate experiments. Starting pH values for Fe(III)-citrate and Fe(III)-
NTA(H+) were 6.1 and 4.1, respectively.

2.3 Discussion

Results from these investigations do not substantiate a working hypothesis that
MR-1 utilizes the same set of gene products to transfer electrons to different solid
substrates. From these data it is evident that the electron transport system in MR-1 is
complex with much functional redundancy. There are some gene products that
participate in electron transfer to several solid substrates including MFC electrodes,
Mn(IV)-oxides and solid Fe(III)-oxides, however, the overall picture of how specific
gene products participate in electron transfer to these substrates is not yet clear. For

44
example, ∆mtrA, ∆mtrB, ∆omcA/∆mtrC, ∆cymA, ∆gspG and ∆pilD were all limited in
their abilities to produce current and reduce HFOM relative to the WT strain. However,
the ∆mtrC/∆omcA mutant was not limited in Mn(IV)-oxide reduction, implying that the
∆mtrC/∆omcA genes are perhaps not directly involved with solid phase Mn(IV)-oxide
reduction. The trends in the solid substrate data (Figure 3, Figure 8, and Figure 10)
suggest that ∆mtrC/∆omcA is not as limited in electron transfer abilities to solids as the
∆mtrA and ∆mtrB mutants. This is especially apparent in the Mn(IV)-oxide reduction
data, which show that after forty-eight hours, the ∆mtrC/∆omcA mutant has reduced 43.8
± 0.30% of the starting concentration of Mn(IV), as compared to 1.65 ± 2.20% for ∆mtrA,
17.3 ± 2.93% for ∆mtrB, and 39.1 ± 0.57% for the WT strain.
Additionally, the ∆mtrC/∆omcA mutant shows higher concentrations of Fe(II)
(approximately two-times higher) from the reduction of HFOM after twenty-four (Figure
10) and forty-eight hours, relative to ∆mtrA and ∆mtrB. Additionally, higher values of
current production (approximately five-times higher) were observed for ∆mtrC/∆omcA
relative to ∆mtrA and ∆mtrB (Figure 3) during the experimental duration. Higher
Mn(IV) and Fe(III) reduction rates of an mtrC mutant, relative to mutants lacking only
mtrA or mtrB, have also been reported by Beliaev et al. (Beliaev, Saffarini et al. 2001).
The data presented herein suggest that the omcA and mtrC gene products are not
the only outer-membrane cytochromes serving as terminal reductase(s) for Mn(IV) or
Fe(III). Furthermore, the solid metal oxide reduction data taken together with the current
production data, (which demonstrated similar limitation patterns relative to the mtr and

45
omc mutants), indicate that MR-1 features more than one pathway for electron transfer to
solids.
The ∆cymA and ∆gspD mutants show the same patterns as ∆mtrB and ∆mtrA; i.e.,
limited in current production, HFOM reduction and Mn(IV)-oxide reduction. These
results are in agreement with Myers and Myers (Myers and Myers 1997) who suggest
that the cymA cytochrome is a necessary component in the electron transport chain that is
common to fumarate, nitrate, Fe(III) and Mn(IV) oxides.
Interestingly, the two mutants deficient in type II protein secretion, ∆gspD and
∆gspG yielded different results relative to each other. Relative to ∆gspD, the ∆gspG
mutant showed limited ability to reduce HFOM, Mn(IV)-oxide, and to produce current in
an MFC.
Another surprising result revealed by this data set was the number of cytochrome
deletion mutants that were able to exceed the WT levels of current production, HFOM,
and Mn(IV)-oxide reduction. There are many possibilities for why this might occur
including the re-direction of electron flow to alternate terminal reductase(s) in the
occurrence of a road-block to a desired pathway or reductase. It may also be the case
that the cytochromes themselves are negative regulators of the transcription and/or
translation of factors involved with these processes. Furthermore, an alternative
possibility is that these mutants are enhanced in their abilities to form biofilms and
therefore yield higher current production and metal oxide reduction rates due to the
presence of more bacteria at the surface.

46
The production of current in an MFC is directly linked to the ability of the
bacteria to oxidize a substrate and transfer electrons resulting from this oxidation to the
anode electrode. The current production results for each evaluated mutant showed that
only five cytochrome deletion mutants, out of the thirty-six tested, were severely limited
in current producing ability relative to the WT. These five mutants produced less than
one-quarter of the current density demonstrated by the WT and include ∆mtrA, ∆mtrB,
the complex ∆mtrC/∆omcA, and ∆cymA (Figure 3). Both type II protein secretion
mutants and the pilD mutant were also severely limited in current production as
previously discussed. Figure 6, Figure 7 and Figure 5 show three different MFC
electrodes that were exposed to: a) ∆pilD, b) ∆mtrC/∆omcA, and 3) WT for three days
and periodic lactate additions. SEM images of the electrodes after MFC evaluation
showed the ∆pilD cells looking stressed, and apparently limited in their ability to
successfully colonize the electrode (Figure 6) relative to the MR-1 WT (Figure 5). The
∆mtrC/∆omcA mutant was able to form a monolayer of cells on the electrode surface and
the cells appear to be in good condition (Figure 7). These images suggest that two
mechanisms may play a role in current production: 1) the ability to produce certain OM
proteins for electron transport; and, 2) the ability to colonize the electrode surface.
Presently, work is being done to determine which mechanism contributes most
significantly to these results.
The decrease in current production and Fe(III)-oxide reduction by the deletion
mutants could be attributed to a loss of cell viability, i.e. the mutant cells are dead and
therefore cannot transfer electrons during either process. However, additional

47
experiments showed that the mtrA (Figure 17) and mtrC/omcA mutants (Figure 18),
which were limited in their abilities to reduce Fe(III)-oxide and produce current, showed
no loss in viability relative to MR-1 as judged by colony forming units (CFU) during the
24 hour time-course experiment (Figure 19).
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
∆mtrA cell counts
∆mtrA Fe(II) production
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
∆mtrA cell counts
∆mtrA Fe(II) production

Figure 17. Average Fe(II) concentration and cell counts resulting from HFOM reduction after
twenty-four hours of exposure to mtrA mutants. Averages and standard deviations were calculated
based on the measured Fe(II) concentrations from triplicate experiments.
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
∆mtrC/omcA cell counts
∆mtrC/omcA Fe(II) production
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
∆mtrC/omcA cell counts
∆mtrC/omcA Fe(II) production

Figure 18. Average Fe(II) concentration and cell counts resulting from HFOM reduction after
twenty-four hours of exposure to mtrC/omcA mutants. Averages and standard deviations were
calculated based on the measured Fe(II) concentrations from triplicate experiments.

48
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
MR1 wild-type cell counts
MR1 wild-type Fe(II) production
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
MR1 wild-type cell counts
MR1 wild-type Fe(II) production

Figure 19. Average Fe(II) concentration and cell counts resulting from HFOM reduction after
twenty-four hours of exposure to MR1 wild-type. Averages and standard deviations were calculated
based on the measured Fe(II) concentrations from triplicate experiments.
A set of HFOM reduction experiments were also executed using the membrane
fractions of MR1 wild-type cells (Figure 20). Membrane fractions were extracted by
French press and inoculated into anaerobic bottles containing lactate and HFOM in the
concentrations reported above. The resulting cell counts and reduction data verify that
cell viability was critical for the reduction of solid phase Fe(III)-oxide.

49
0
10
20
30
40
50
60
70
0246 8 10 12 14 16 18 20
1.0E+00
1.0E+01
1.0E+02
1.0E+03
1.0E+04
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
MR1 wild-type membrane
fraction cell counts
MR1 wild-type membrane
fraction Fe(II) production
0
10
20
30
40
50
60
70
0246 8 10 12 14 16 18 20
1.0E+00
1.0E+01
1.0E+02
1.0E+03
1.0E+04
1.0E+05
1.0E+06
1.0E+07
1.0E+08
1.0E+09
1.0E+10
Time (hrs)
Fe(II) produced (µM)
Cell counts (cells/mL)
MR1 wild-type membrane
fraction cell counts
MR1 wild-type membrane
fraction Fe(II) production

Figure 20. Average Fe(II) concentration and cell counts resulting from HFOM reduction after
twenty-four hours of exposure to MR1 wild-type membrane fractions. Averages and standard
deviations were calculated based on the measured Fe(II) concentrations from triplicate experiments.
The results from this experiment also correlate with the surface attachment ability
of MR-1 and its mutants. For example, ∆mtrA (and the other deletion mutants showing
limited abilities to reduce Fe(III)-oxide and produce current) may have been unable to
facilitate electron transfer because of an inability to produce OM proteins that interact
with solid substrates. These results further indicate that electron transfer to solid
substrates by MR-1 is directly related to physical interactions between the substrate
surface and cells (or conductive cell appendages (Gorby, Yanina et al. 2006)).
The regulatory fur::pKNOCK and crp::Tn5 mutants were limited to below 20% of
the current density values obtained for the MR-1 WT. These genes were previously
implicated as influencing the regulation of metal oxide reduction and are herein
implicated as contributors to current production. The tatC::pDS3.1 mutant was also
limited in current producing abilities.

50
Notably, a putative OM protein ∆SO1673 (ompW::pDS3.1) was also moderately
affected in current density, showing only about 60% of the WT current density values.
The results from the soluble Fe(III)-NTA data show a very different pattern of
reduction by these mutants, relative to the solid-oxide and MFC data. It is expected that
cytochromes and other proteins might play unique roles in the reduction of a soluble
compounds relative to how they would be employed to interact with solid compounds.
Soluble compounds like Fe(III)-NTA can permeate the cell membrane and interact with
the organisms electron transport chain within the confines of the cell membrane. In
contrast, the reduction of solid compounds requires either direct physical contact by the
organism or the production of an extra-cellular electron shuttle to mediate electron
transfer. These data corroborate the conclusions of Ruebush et al. (Ruebush, Brantley et
al. 2006) indicating that MR-1 utilizes different mechanisms for solid and soluble Fe(III)
reduction. However, further exploration of the abilities of ∆mtrA, ∆mtrC/∆omcA, ∆mtrF,
∆ifcA-1, ∆dmsC, and WT to reduce different forms of soluble ferric iron led to a very
complex picture of pH relationships and perhaps unique affinity for different soluble iron
compounds.
The ability of WT and ∆mtrA to increase the pH of every form of soluble ferric
media tested may play a role in increased Fe(III)-reduction over time. Interestingly,
∆mtrA demonstrated the ability to shift the Fe(III)-NTA(H
+
) media from pH 4.1 to pH
5.4 in four days, while WT was only able to increase the pH from 4.1 to 4.6 during the
same period. The mtrA mutant also achieved higher cell densities than WT at the end of
a four day exposure to Fe(III)-citrate (pH 5.6) and Fe(III)-NTA(Na
+
) (pH 7.2). It should

51
be noted that cell viability at pH 4.1, the pH of the Fe(III)-NTA media during each
mutant experiment, was very poor. Both WT and ∆mtrA cultures were completely
nonviable after four days of exposure to these acidic conditions. This is likely the reason
that WT showed such low values of Fe(II) production after twenty-four hours of exposure
to the Fe(III)-NTA(H
+
) media.
Unexpectedly, the pH 4.1 media had a very positive effect on several of the
mutants including ∆ifcA-1 and ∆dmsC, which had previously demonstrated very low
Fe(III) reduction abilities with Fe(III)-citrate at a pH of 6.1. These results now introduce
a new question about how various cell mutations in MR1 may affect the organisms pH
tolerance and to what degree does abiotic reduction play a role.
As a whole, this is the first set of current densities obtained from microbial fuel
cells that can be directly related via specific mutants to microbial physiology.
Additionally, this is the first data set associating specific MR-1 gene products with
current production in a MFC. Data from these MR-1 mutant evaluations also indicate
that the gene products associated with current production are not necessarily identical
with those involved in Mn(IV)- and Fe(III)-oxide, or Fe(III)-NTA reduction. Instead,
these data suggest that several different pathways, and perhaps mechanisms, may be
employed for current production and metal oxide reduction. Ultimately, the different
patterns of metal oxide reduction and current production indicate a very complicated
picture of electron flow via MR-1 cytochromes.

52
Chapter 3: Evaluation of the power producing abilities of different
Shewanella strains

The knowledge about which genes are involved in electron transfer to a MFC
anode by Shewanella oneidensis MR-1, leads to the question of how many other strains
of Shewanella also possess these genes; and are they, like MR-1, able to produce power
in a MFC? Several Shewanella strains have now been genetically sequenced including S.
oneidensis MR-4 and MR-7, S. putrefaciens CN-32 and SP200, S. ANA-3 and W3-18-1,
S. loihica PV-4, S. frigidimarina NCIMB400, S. denitrificans OS217, S. baltica OS155,
OS185, OS195 and OS223, S. woodyi, S. benthica, S. sediminis, S. amazonensis SB2B, S.
pealeana and S. halifaxens (Fredrickson, Romine et al. 2008). Fredrickson et al.
(Fredrickson, Romine et al. 2008) found that every sequenced strain, except S.
denitrificans OS217, features the genes mtrA, mtrB and mtrC, which are known to be
involved in MFC power production. However, several strains also feature unique sets of
genes that neighbor mtrC, and encode several other c-type cytochromes and/or outer-
membrane proteins and precursors (Kolker, Picone et al. 2005; Elias, Monroe et al. 2006;
Hartshorne, Jepson et al. 2007; Fredrickson, Romine et al. 2008). These unique sets of
genes are not conserved among all the strains, and they may impact power production in
an MFC.
In addition to known genetic differences among the Shewanella strains, there are
also differences in physiology (Venkateswaran, Moser et al. 1999; Fredrickson, Beliaev
et al. 2004; Nealson and Scott 2006). Each strain has unique preferences for, among
other things, forming biofilms, carbon sources, electron donors, electron acceptors, pH,

53
salinity, and temperature tolerances (Obraztsova and Wang 2008). All of these factors
may also impact how a strain could perform in a MFC system.
A subset of sequenced Shewanella strains were evaluated as anode catalysts in the
MFC to gain a better understanding about the genetic and physiological differences that
may play a role in power production. As a preliminary investigation, S. oneidensis MR-1
(Myers and Nealson 1988), S. putrefaciens SP200 (Obuekwe and Westlake 1982), S.
putrefaciens CN-32 (Fredrickson, Zachara et al. 1998), S. amazonensis SB2B
(Venkateswaran, Dollhopf et al. 1998), S. baltica OS155 (Brettar, Moore et al. 2001), S.
loihica PV-4 (Gao, Obraztova et al. 2006) and S. denitrificans OS217 (Brettar, Moore et
al. 2001) were evaluated in terms of current and charge densities while the MFCs were
operated with a load of 10 Ohm. The outcome of these experiments indicated that some
strains of Shewanella could produce equivalent current densities relative to the model
organism MR-1; and that MFC system differences may lead to altered performance
trends. These initial experiments led to the expansion of MFC testing using additional
Shewanella strains and different MFC conditions.
The expanded investigation of power performance by the genus Shewanella
included stain evaluations of S. oneidensis MR-1, S. putrefaciens CN-32, S. amazonensis
SB2B, S. putrefaciens W3-18-1 (Murray, Lies et al. 2001), S. ANA-3 (Saltikov, Cifuentes
et al. 2003), S. loihica PV-4 and S. oneidensis MR-4 and MR-7 (Nealson, Myers et al.
1991). All of these strains were tested in triplicate using MFCs assembled with either a
Nafion 117 (0.178 mm thickness) or 424 membrane (fiber reinforced, 0.178 mm
thickness) and featuring either a PIPES (50 mM) or sodium phosphate (100 mM) buffer
to evaluate how small system changes may affect overall performance.

54
The comprehensive results for the eleven evaluated strains of Shewanella are the
first set of data comparing the power performances of several different bacterial strains
within one genus, using the same MFC system. Additionally, these results represent the
first MFC evaluations that include electrochemical testing, operational data at maximum
current and maximum power, periodic electrolyte analyses, and electron microscopy of
the MFC anodes under different MFC conditions.
3.1 Initial strain evaluations (KIST)

The first MFC evaluations of different Shewanella strains were conducted at the
Korea Institute of Science and Technology (KIST) in collaboration with Dr.s Byung
Hong Kim and In Seop Chang. Samples of S. oneidensis MR-1, S. putrefaciens SP200, S.
putrefaciens CN-32, S. amazonensis SB2B, S. baltica OS155, S. frigidimarina
NCIMB400, S. loihica PV-4 and S. denitrificans OS217 were shipped from the Nealson
Lab to KIST and evaluated using MFC systems similar to that shown in Figure 2
(Chapter 2).
3.1.1 KIST methods

Each strain was grown anaerobically using a phosphate-buffered basal media
(PBBM) containing lactate (30 mM) as an electron donor, fumarate (60 mM) as an
electron acceptor and 1 g/L yeast extract (pH 7.0) (Kim, Hyun et al. 2005). The cells
were washed twice in a sodium phosphate buffer (50 mM PO
4
, 100 mM NaCl, pH 7.0),
and injected into the MFC anode compartment. The final OD
660
of every culture was
measured and used to calculate an experimental dilution of OD
660
0.4. The appropriate

55
volume of cells was injected into the anode compartment of each MFC so that
approximately the same number of bacteria was present for each experiment; and
phosphate buffer was used as the diluting electrolyte.
Three dual-compartment MFCs were used to test each strain. The MFCs were
assembled using bare graphite felt anodes and platinum coated graphite felt cathodes
(0.07 mg/cm
2
platinum loading), both using platinum wire leads and an apparent surface
area of 17.9 cm
2
. Pretreated Nafion® 424 ion exchange membranes were used to
separate the anode and cathode compartments and facilitate proton transfer between them.
Each sterile MFC compartment was injected with 30 mL of sodium phosphate
buffer and the anode compartment was continuously flushed with sterile filtered nitrogen
for one hour. After this one hour period, air was introduced to the cathode compartment
and data were recorded for an additional hour to establish a baseline voltage without
bacteria at the anode. The anode and cathode compartments were continuously purged
with nitrogen and air, respectively, at a rate of 10 mL/min throughout each experiment.
All MFCs were operated using a 10 Ohm resistor (R) as the load. The voltage (V)
differences across the load were recorded using a digital multimeter (Keithley
Instruments, 2700) every 5 minutes over a two-day period.
The observed voltages over the two-day period were used to calculate current (I)
according to Ohms Law (V = IR), and current versus time (t) profiles were generated.
The I-t profiles were utilized to determine total experimental charge (C
e
) by integrating
the area under each I-t curve with respect to time. Coulombic efficiencies were
calculated as C
e
/C
t
, where C
t
is the theoretical coulombic yield. The theoretical

56
coulombic yield was calculated as
M
V S Fb
C
t
) (∆ = , where F is Faradays constant (96485
C/mol), b is the number of moles of electrons produced per mole of substrate oxidized,
∆S is the substrate concentration utilized at a given sampling time (g/L), V is the liquid
volume of the anode compartment (0.03 L), and M is the substrate molecular weight
(g/mole) (Liu and Logan 2004).
Bacteria were injected into the MFC anode after the baseline voltage was
recorded and allowed to achieve a stable voltage before lactate (1.5 mM) was added to
the MFC anode as the electron donor. Two lactate feedings were conducted for each
strain evaluation and electrolyte samples were taken from the MFC anode after every
lactate feeding and when the voltage difference reached baseline values after feedings.
Electrolyte samples were analyzed via high pressure liquid chromatography
(HPLC) using a mobile phase of 2.5 mM sulfuric acid at a flow rate of 0.80 ml/min. A
UV detector set at 210 nm was used to generate chromatograms for each MFC sample.
Organic acid concentrations were calculated using the peak areas of each peak that
corresponded to a known retention time of a specific organic acid. Calculations were
based on a set of standards previously recorded by Dr. In Seop Chang for the specific
HPLC machine.
3.1.2 KIST results and discussion

The current and charge densities, and the coulombic efficiencies for each
evaluated strain are shown in Figure 21 a, b, c, and d, respectively. These results show
that S. oneidensis MR-1, S. putrefaciens CN-32, S. putrefaciens SP200 and S.

57
amazonensis SB2B were able to produce the highest current densities for each lactate
feeding; and that MR-1 consistently performed better than the other strains in MFCs 1
and 3.
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
a) Current density (µA/cm
2
), Feed 1 b) Current density (µA/cm2), Feed 2
Charge density (Coulomb/cm
2
)
0.00
0.05
0.10
0.15
0.20
0.25
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
c) Average charge density (C/cm
2
)
0.0
5.0
10.0
15.0
20.0
25.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Coulombic efficiency (%)
d) Average coulombic efficiency (%)
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155 MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-1
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-2
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-3
MFC-1
MFC-2
MFC-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155 MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Current density (µA/cm
2
)
a) Current density (µA/cm
2
), Feed 1 b) Current density (µA/cm2), Feed 2
Charge density (Coulomb/cm
2
)
0.00
0.05
0.10
0.15
0.20
0.25
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Charge density (Coulomb/cm
2
)
0.00
0.05
0.10
0.15
0.20
0.25
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
0.00
0.05
0.10
0.15
0.20
0.25
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155 MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
c) Average charge density (C/cm
2
)
0.0
5.0
10.0
15.0
20.0
25.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Coulombic efficiency (%)
0.0
5.0
10.0
15.0
20.0
25.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
0.0
5.0
10.0
15.0
20.0
25.0
S. oneidensis MR-1 S. putrefaciens CN32 S. putrefaciens SP200 S. dentrificans OS217 S. amazonensis SB2B S. species PV4 S. baltica OS155
MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155 MR-1 CN-32 SP200 OS217 SB2B PV-4 OS155
Coulombic efficiency (%)
d) Average coulombic efficiency (%)

Figure 21. Current densities for each MFC and Shewanella strain for a) the first lactate feeding and
b) the second lactate feeding. Average charge densities and coulombic efficiencies are shown in c)
and d), respectively. Averages and standard deviations were calculated using the total charge and
measured organic acid concentrations corresponding to both feedings, in three MFCs, for each strain.
Interestingly, the current density results from MFC 2 were consistently higher
than those for MFCs 1 and 3. Additionally, the relative trends shown for each bacterial
strain in MFC 2 were slightly different than that of MFC1 and MFC3. All of the MFCs
had been previously used in other studies at KIST and it is unknown whether MFC 2 was
treated differently than the others. These data indicate that there were system differences

58
between MFC 2 and MFCs 1 and 3, perhaps higher platinum loading at the cathode,
higher flow rates of air to the cathode, closer electrode spacing, or differences in how the
membranes were conditioned prior to assembly. All of these parameters may affect the
rate limiting steps in an MFC and could therefore lead to different performances (Gil,
Chang et al. 2003; Oh, Min et al. 2004; Liu, Cheng et al. 2005; Oh and Logan 2006).
The relative performance trends for each bacterial strain tested in MFC1 and MFC3 were
fairly consistent.
Another consistent trend was apparent between the current densities results for the
two lactate feedings. With the exception of strain OS155, the average of all of the MFC
current densities were higher for the second lactate feeding relative to the first. Roughly
a 20% to 30% increase in current density was observed for the second lactate feeding for
all strains and MFCs.
These results may indicate a relaxation time that is required for each strain to
adapt and thrive at the MFC anode. Alternatively, these results may be indicative of the
stress caused by the bacterial harvesting and washing process prior to MFC inoculation.
Each harvested culture was twice centrifuged into a pellet of cells and washed with
phosphate buffer to remove any excess organic compounds that might have existed in the
growth media. These actions may have stripped the bacteria of their outer-membrane
cytochromes required for electron transfer to surfaces (Myers and Myers 2002; Myers
and Myers 2003; Gorby, Yanina et al. 2006; Bretschger, Obraztsova et al. 2007) and/or
removed any biologically produced mediators (Lies, Hernandez et al. 2005; Marsili,
Baron et al. 2008).

59
While there were statistically significant differences in current density between
several strains, the same was not true for charge density. Nearly all Shewanella strains
produced similar values of total charge over the two lactate feedings, with the exception
of S. baltica OS155, which exhibited a statistically significant lower charge density
relative to the other strains. This result is reasonable based on the fact that each MFC
was given approximately the same amount of lactate as the electron donor. Total charge
reflects only the total number of electrons produced from the oxidation reaction, and
consumed by the reduction reaction; however these values do not yield any kinetic
information related to how quickly the reactions took place, or if the reactions ran to
completion, i.e., complete consumption of reductant and oxidant.
However, the coulombic efficiencies do yield some information about the
consumption of reductant and oxidant. Coulombic efficiency is calculated as the total
experimental charge divided by the charge that is theoretically possible if the electron
donor has been completely oxidized. Given that the oxidation of any compound will
result in intermediates, and sometimes different end-products, values for theoretical
charge were based on the measured organic acid concentrations in the anolyte of each
MFC after each feeding and when current densities reached baseline values.
The anaerobic oxidation of lactate to acetate yields four moles of electrons and
features pyruvate as an intermediate (Scott and Nealson 1994; Tang, Meadows et al.
2007). In most cases the molar amounts of acetate, pyruvate and formate in the anolyte at
the end of each experiment did not balance with the molar amounts of lactate introduced
into the system. Therefore, it was assumed that this carbon difference exited the system

60
as carbon dioxide and the theoretical values for charge were calculated accordingly and
applied to the calculation of coulombic efficiency
Most of the tested Shewanella strains demonstrated a coulombic efficiency of
approximately 10%. S. baltica OS155 only demonstrated a coulombic efficiency of
approximately 2% and S. denitrificans OS217 exhibited a coulombic efficiency of
approximately 6%. These results indicate that a majority of electrons resulting from the
oxidation of lactate were not transferred to the MFC electrodes. It is assumed that these
electrons were utilized by the cultures to build biomass and run other energy-requiring
biological functions; however no quantification of biomass was performed after the
termination of experiments to verify this.
These results yielded an initial understanding about how different Shewanella
strains perform in a MFC and provided some insight into system dependent performance.
However, only seven out of eight strains evaluated were able to grow anaerobically to
high enough culture densities for MFC testing using such high cell densities. S.
frigidimarina NCIMB400 never successfully grew, likely because the temperature in the
KIST lab (approximately 28°C) exceeded optimal conditions for this strain. S.
frigidimarina was isolated in the North Sea and has an ideal growth temperature of
approximately 10°C (Bowman, McCammon et al. 1997). Additionally, S. baltica OS155
was isolated from the Baltic Sea and has similar temperature limitations (growth at 4°C),
which may explain the poor performance of this strain at 28°C temperatures (Ziemke,
Hofle et al. 1998).

61
S. denitrificans OS217 was also isolated from the Baltic Sea and is reported to
grow ideally at 4°C (Brettar, Christen et al. 2002). Additionally, S. denitrificans OS217
lacks the mtr cassette of genes that encode several outer-membrane c-type cytochromes
involved with current production and this strain was expected to act as the negative
control for MFC evaluations (Bretschger, Obraztsova et al. 2007; Fredrickson, Romine et
al. 2008). However, the results clearly showed a positive response in terms of current
density and coulombic efficiency for S. denitrificans OS217. DNA extraction and
genetic sequencing were later performed for S. denitrificans OS217 and it was found that
the original stock of this strain was contaminated and therefore the results obtained at
KIST could not be attributed to strain OS217. A similar genetic analysis found that the S.
putrefaciens SP200 stock was a mixed culture and therefore the KIST MFC results were
not representative for this strain.
Given these inconsistencies, a new set of MFC evaluations were planned and
executed in collaboration with Andrea Cheung at the University of Southern California
(USC). Several changes were made to the growth and inoculation strategies that were
employed at KIST; in addition, the electrode sizes of the anode and cathode were
increased from 17.9 cm
2
to 79 cm
2
.
3.2 Expanded strain evaluations (USC)

Electrochemical impedance spectroscopy (EIS), potential sweeps,
potentiodynamic polarization of the anode and cathode, and cyclic voltammetry (CV)
were also employed during the evaluation of some strains to complement the current
density and organic acid analyses for the Shewanella strain evaluations (Manohar,

62
Bretschger et al. 2008; Manohar, Bretschger et al. 2008). Four new Shewanella strains
were incorporated into the evaluations given the problems discussed above for S.
putrefaciens SP200, S. denitrificans OS217, S. frigidimarina NCIMB400 and S. baltica
OS155. The new set of Shewanella strains included S. oneidensis MR-1, S. putrefaciens
W3-18-1, S. amazonensis SB2B, S. ANA-3, S. oneidensis MR-4, S. oneidensis MR-7, S.
putrefaciens CN32, and S. loihica PV-4. A description of each strains geographic site of
isolation is provided in Table 4.
Table 4. Shewanella strains and a description of the environment they were isolated from.
Strain
Geographic
Origin
Environment
characteristics of
isolation area
Reference
S. oneidensis
MR-1
Lake Oneida,
NY, USA
Anaerobic fresh water
sediment
(Myers and
Nealson 1988)
S. putrefaciens
W3-18-1
Pacific Ocean,
Washington
Coast, USA
Marine sediment
(630 m)
(Murray, Lies et al.
2001)
S. amazonensis
SB2B
Amazon River
delta, Brazil
Shallow marine
sediment (1 m)
(Venkateswaran,
Dollhopf et al.
1998)
S. ANA-3
Woods Hole,
MA, USA
As-treated wooden
pier in a brackish
estuary
(Saltikov, Cifuentes
et al. 2003)
S. oneidensis
MR-4
Black Sea
Sea water, oxic zone
(5 m)
(Nealson, Myers et
al. 1991)
S. oneidensis
MR-7
Black Sea
Sea water, anoxic zone
(60 m)
(Nealson, Myers et
al. 1991)
S. putrefaciens
CN-32
Albuquerque,
NM, USA
Shale-sandstone,
subsurface, (250 m)
(Fredrickson,
Zachara et al. 1998)
S. loihica PV-4
Loihi
Seamount, HI,
USA
Iron-rich mat,
hydrothermal sea vent
(1,325 m)
(Gao, Obraztova et
al. 2006)

63
3.2.1 USC methods

The first set of MFC evaluations with this strain collection were conducted using
a 20 mM concentration of lactate as the electron donor at the anode, 100 mM sodium
phosphate buffer as the anolyte and catholyte, and Nafion
®
424 membranes to separate
the anode and cathode compartments.
Additionally, a 50 mM PIPES buffer (described in Chapter 2), 5 mM lactate
concentration, and different ion exchange membranes were utilized for a complementary
set of MFC evaluations with the same strain collection. These complementary
evaluations were performed because of difficulty discerning organic acid peaks in the
chromatograms generated from the phosphate buffer anolytes, which had high
concentrations of sodium and lactate. Additionally, reproducible current density results
were difficult to achieve using the phosphate buffer. Nafion
®
117 membranes were
chosen to replace the Nafion
®
424 because autoclaving the fiber reinforced 424
membranes occasionally induced tears in the polymer along the fiber lengths, which
resulted in contamination of the MFC cathode. However, an experiment performed using
the PIPES buffer with the 424 membranes indicated that the membranes were not the
cause of inconsistent results in previous studies.
Each strain was grown aerobically for forty-eight hours in either the phosphate
buffered (100 mM) or PIPES buffered (50 mM) minimal media with 20 mM lactate as
the electron donor and defined salt, vitamin, amino acid, and mineral components as
described in Chapter 2. The cultures were harvested and injected into the MFCs such that

64
approximately 2 x 10
8
cells/mL were present in the anode compartment. The diluting
anolyte was either the phosphate or PIPES buffer.
MFCs were assembled with graphite felt anodes and platinized graphite felt
cathodes with an apparent surface area of 79 cm
2
. Either a Nafion
®
424 or 117
membrane separated the anode and cathode compartments. MFCs were assembled and
allowed to sit overnight with deionized water in the anode and cathode compartments to
saturate the porous electrodes and force out any remaining pockets of air. The MFCs
were then sterilized at 121°C for 15 minutes, allowed to cool, and filled with buffer at the
anode and cathode. Ag/AgCl reference electrodes were sterilized using 70% ethanol and
ultra-violet (UV) light (UV exposure in a clean bench for 15 minutes), and inserted into
the anode and cathode compartments such that equal distances were maintained between
each reference and the respective anode and cathode electrodes.
The anode compartment was continuously purged with sterile filtered nitrogen gas
at a rate of 20 mL/min and the cathode was purged with sterile filtered air at a rate of 40
mL/min. Sterile MFC conditions were maintained for twenty-four hours prior to MFC
inoculation to ensure anaerobic conditions at the anode. A measurement of oxygen at the
anode yielded an oxygen concentration of 41.7 ± 17.6 µg O
2
/L (1.3 ± 0.55 µM O
2
),
indicating that oxygen crossover from the cathode to anode was negligible.
Each MFC was inoculated with a Shewanella strain and injected with either 20
mM or 5 mM sodium lactate. The MFCs were allowed to sit at open circuit for
approximately twelve hours to ensure stable anodic and cathodic potentials. All MFCs
were evaluated at room temperature, approximately 22°C.

65
EIS was performed using the anode and cathode as the working electrode,
respectively. EIS measurements were performed at open-circuit potentials, for one out of
three MFCs, under sterile and inoculated anode conditions. EIS data were generated by
applying an alternating potential of 10 mV (vs. Ag/AgCl) amplitude in a frequency range
of 10
-3
to 10
5
Hz.
Additionally, potential sweeps were conducted under sterile and inoculated anode
conditions for all three MFCs. Potential sweeps began at the open-circuit cell voltage
(OCV), where I = 0, and proceeded at a scan rate of 0.1 mV/sec until zero cell voltage
and the maximum cell current was attained (I = I
max
). Maximum power (P
max
) was
calculated for each inoculated MFC according to P = I*V. The resistance (R
ext
) that
corresponded to P
max
was determined according to R = V
2
/P and applied as the load
across each cell during MFC operation. The voltage drop across the load was measured
every five minutes for each MFC over a period of about two hundred hours. These data
were used to generate I-t curves, which were then used to calculate C
e
for each MFC.
Two lactate injections occurred after EIS and potential sweeps were conducted
and during MFC operation. Anolyte and catholyte samples were extracted before and
after each lactate injection as well as when the cell voltage was observed to be a
maximum.
Electrolyte samples were centrifuged, acidified and sterile filtered to remove any
bacterial components, then analyzed using a high pressure liquid chromatography
(HPLC) machine (Agilent 1100 series) with a reversed phase C18 column (Phenomenex,
Synergi-Hydro). Chromatographs were generated using a diode array detector (Agilent

66
1100 series) set to the wavelength of 210 nm. A 20 µL sub-sample of each electrolyte
was injected into the sampling loop and the analytical method employed a 2.5 mM
sulfuric acid mobile phase (pH 2.0) running at a flow rate of 0.5 mL/min. Peak areas
were obtained for all the peaks in each electrolyte chromatogram and concentrations were
calculated based on the peak areas and retention times of known standards for each
evaluated organic compound in both buffers. The measured organic acid concentrations
were used to determine coulombic efficiencies for each MFC as described above.
Some strains were also evaluated using CV and potentiodynamic polarization.
CV was performed at a scan rate of 25 mV/sec in the limits of +750 mV to -750 mV or
+750 mV to -850 mV. Potentiodynamic polarization was performed at a scan rate of
0.167 mV/sec. Cathodic polarization of the cathode was performed from 30 mV beyond
the cathodic open circuit potential (E
c
OCP
+ 30mV) to -1 V vs Ag/AgCl. Anodic
polarization of the anode was performed from 30 mV beyond the anodic open-circuit
potential (E
c
OCP
- 30mV) to +1 V vs Ag/AgCl. Limiting current densities (i) were
determined by identifying the current density associated with the intersection of the
anodic and cathodic polarization curves for each set of potentiodynamic scans.
3.2.2 USC results

The power and current density results from the evaluation of different buffers,
electron donor concentrations and membranes systems for each Shewanella strain are
presented in Figure 22. It is apparent from these data that the PIPES buffer, 5 mM lactate
and Nafion 117 membrane (System 2) MFC evaluations resulted in higher power and
current densities than those strain evaluations performed with the phosphate buffer, 20

67
mM lactate and Nafion 424 membrane (System 1). The relative power density results
(Figure 22 a and b) between System 1 and System 2 show that each strain is differently
affected by the system parameters. For example, S. oneidensis MR-7 demonstrated a
60% increase in power density when tested under System 2 conditions; however S.
oneidensis MR-1 power densities only increased by 30%. Overall, S. putrefaciens W3-
18-1, S. putrefaciens CN-32 and S. putrefaciens MR-7 were the strains that showed the
greatest changes in power densities between System 1 and System 2 evaluations; the
performance of each strain relative to the other was within measurement deviations for all
other strains.

68
0.0
0.5
1.0
1.5
2.0
2.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Power density (µW/cm
2
)
Power density (µW/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Current density (µA/cm
2
)
Current density (µA/cm
2
)
a) Power density (µW/cm
2
), System 1 b) Power density (µW/cm
2
), System 2
c) Current density (µA/cm
2
), System 1 d) Current density (µA/cm
2
), System 2
0.0
0.5
1.0
1.5
2.0
2.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Power density (µW/cm
2
)
Power density (µW/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Current density (µA/cm
2
)
Current density (µA/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Power density (µW/cm
2
)
Power density (µW/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
0
0.0
0.5
1.0
1.5
2.0
2.5
0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Power density (µW/cm
2
)
Power density (µW/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Current density (µA/cm
2
)
Current density (µA/cm
2
)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Current density (µA/cm
2
)
Current density (µA/cm
2
)
a) Power density (µW/cm
2
), System 1 b) Power density (µW/cm
2
), System 2
c) Current density (µA/cm
2
), System 1 d) Current density (µA/cm
2
), System 2

Figure 22. Power densities for a) System 1, and b) System 2; and current densities for c) System 1,
and d) System 2. Power density averages and standard deviations were calculated using the
maximum power generated during potential sweep experiments for three MFCs. Current densities
were calculated as the average stable current for ten hours of operation across an applied load
determined from the potential sweep experiments for three MFCs. System 1 parameters included
100 mM phosphate buffer, 20 mM lactate, and Nafion 424 membranes. System 2 parameters
included a 50 mM PIPES buffer, 5 mM lactate, and Nafion 117 membranes. All other MFC
components including electrode size, compartment volume, and gas diffusion rates were the same.
The power density relationships between the strains show that S. putrefaciens
W3-18-1 is the highest power producer while S. oneidensis MR-4 and S. loihica PV-4 are
the lowest power producers. All other strains demonstrated equivalent power densities to
S. oneidensis MR-1 in System 2 and equivalent or less power density in System 1.
The operational current densities appeared to be most impacted in System 1
versus System 2 as shown in Figure 22 c) and d). The deviations resulting from the

69
operational current measurements in System 1 were especially high. For most System 1
evaluations, only two out of three MFCs were able to generate current above the
baseline values, and consistency between the three MFCs tested for each strain was
achieved only with S. ANA-3. The trends observed for the relative strain performance in
terms of current densities were very similar to that observed for power density in System
2. However, this was not the case for System 1.
The average open-circuit cell voltage (OCV) obtained for each strain evaluation
from the V-I curves generated for three MFCs are shown in Table 5. Open circuit
potentials (vs. Ag/AgCl) for the anode (E
a
OCP
) and cathode (E
c
OCP
) are also shown. These
values were recorded prior to EIS and V-I measurements for only one of the three MFCs
used for each strain, and are not reported as averages.
The E
a
OCP
and E
c
OCP
values were fairly consistent between System 1 and System 2
evaluations. It is apparent that the E
c
OCP
values for System 1 evaluations were about 20
mV lower over all compared to the System 2 evaluations. The same cathodes, with
relatively the same platinum loading, were utilized for System 1 and System 2
evaluations, with the System 1 evaluations taking place first. The lower E
c
OCP
values for
System 1 may be due to a difference in cleaning procedures of the cathodes during these
two evaluations. System 1 evaluations reused each cathode without cleaning them prior
to reassembly. During the System 2 evaluations, all of the cathodes were
electrochemically cleaned to remove any adsorbed chemicals from the platinum catalyst
prior to MFC use and this technique appeared to have an overall impact on the E
c
OCP

values. Electrochemical cleaning methods are provided in Appendix A.

70
The E
a
OCP
values remained within 10% agreement between System 1 and System
2 evaluations of S. oneidensis MR-1, S. putrefaciens W3-18-1, S. amazonensis SB2B and
S. ANA-3. However, the E
a
OCP
values diverged by 10% or more, for the remaining strains
S. oneidensis MR-4 and MR-7, S. putrefaciens CN32, and S. loihica PV-4. S. loihica PV-
4 and S. putrefaciens W3-18-1 demonstrated the highest average OCV values over all
three MFCs in both systems tested. This result corresponds with the high power and
current density observed for S. putrefaciens W3-18-1, but S. loihica PV-4 was one of the
poorest generators of power and current density. This relative discrepancy between
power density and average OCV for S. loihica PV-4 may indicate that this strain is
limited by the MFC design, i.e., high internal resistance, and may perform better under
different operational conditions as is indicative from the results obtained at KIST (see
above).
Table 5. Open-circuit potentials (vs. Ag/AgCl) recorded for the anode and cathode of one of three
MFCs and the average open-circuit cell voltage found for all three MFCs.
System 1 System 2
Strain
E
a
OCP
vs
Ag/AgCl
(V)
E
c
OCP
vs
Ag/AgCl
(V)
Avg. OCV,
3 MFCs
(V)
E
a
OCP
vs
Ag/AgCl
(V)
E
c
OCP
vs
Ag/AgCl
(V)

Avg. OCV,
3 MFCs
(V)
S. oneidensis MR-1 -0.44 0.29 0.73 ± 0.02 -0.42 0.33 0.77 ± 0.00
S. putrefaciens W3-18-1 -0.50 0.24 0.78 ± 0.04 -0.50 0.30 0.83 ± 0.01
S. amazonensis SB2B -0.48 0.28 0.75 ± 0.04 -0.44 0.30 0.75 ± 0.01
S. ANA-3 -0.45 0.28 0.72 ± 0.01 -0.44 0.29 0.74 ± 0.00
S. oneidensis MR-4 -0.40 0.28 0.73 ± 0.04 -0.45 0.30 0.75 ± 0.02
S. oneidensis MR-7 -0.45 0.13 0.47 ± 0.23 -0.51 0.28 0.79 ± 0.01
S. putrefaciens CN-32 -0.51 0.26 0.68 ± 0.16 -0.45 0.25 0.73 ± 0.02
S. loihica PV-4 -0.46 0.27 0.85 ± 0.14 -0.54 0.30 0.84 ± 0.03

Anolyte samples from each strain evaluation in both systems were collected,
however poor peak separation was observed in every chromatogram generated from the

71
System 1 anolytes. These chromatogram data suggest that the high sodium
concentrations in the lactate and phosphate buffer electrolytes interfered with organic
acid separation in the HPLC column (Guo, Srinivasan et al. 2007). Additionally, the
current density values were so low and inconsistent, that coulombic efficiency
calculations for System 1 would be unreliable. Therefore organic acid analyses, charge
density and coulombic efficiency calculations were only performed for System 2.
The charge density and coulombic efficiency results for each strain evaluation in
System 2 are presented in Figure 23.
0
5
10
15
20
25
30
35
40
45
50
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Charge density (Coulomb/cm
2
)
Coulombic efficiency (%)
a) Charge density (C/cm
2
), System 2 b) Coulombic efficiency (%), System 2
0
5
10
15
20
25
30
35
40
45
50
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
Charge density (Coulomb/cm
2
)
Coulombic efficiency (%)
a) Charge density (C/cm
2
), System 2 b) Coulombic efficiency (%), System 2

Figure 23. a) Charge density, and b) coulombic efficiency for eight Shewanella strains evaluated in
triplicate using a 50 mM PIPES buffer and 5 mM lactate anolyte with a Nafion 117 membrane.
Charge density averages and standard deviations for each strain were calculated based on the total
charge measured for each lactate feed in each MFC. Average coulombic efficiencies and
corresponding deviations were calculated as the total charge from each MFC divided by the total
theoretical charge available from the sum of each lactate feed to the corresponding MFC. Anolyte
concentrations for acetate, formate and pyruvate were determined using high pressure liquid
chromatography.
The charge densities for each strain in System 2 follow closely to the relative
trends of stable current density except for CN-32 and PV-4. S. putrefaciens CN-32
showed a lower charge density relative to S. oneidensis MR-7, but their respective stable

72
current densities were equivalent. S. loihica PV-4 had a much higher average charge
density relative to all other strains in comparison with the relative current density results,
which showed PV-4 as a low current producer. S. loihica PV-4 also showed the most
variability across the three MFCs relative to the other strains.
The coulombic efficiencies for all strains ranged between 20% and 40%, with
W3-18-1 being the most efficient overall with an average coulombic efficiency of 35%.
MR-1 was the least efficient strain with an average coulombic efficiency of 18%.
Coulombic efficiencies followed patterns of organic acid oxidation. Those strains that
were more efficient were those able to oxidize lactate to completion, i.e., oxidized to CO
2
.
The I-t curves for each System 2 MFC and corresponding organic acid profiles are shown
in Figure 24, Figure 25, Figure 26, Figure 27, Figure 28, Figure 29, Figure 30 and Figure
31.

73
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
MFC1 MFC3
S. oneidensis MR1
MFC1 MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
I (mA)
I (mA)
I (mA)
MFC2
I (mA)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
050 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
[mM]
[mM]
[mM]
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
0 50 100 150 200
Time (hrs)
MFC1 MFC3
S. oneidensis MR1
MFC1 MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
I (mA)
I (mA)
I (mA)
MFC2
I (mA)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
050 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
[mM]
[mM]
[mM]

Figure 24. S. oneidensis MR-1 I-t and HPLC profiles for each MFC tested.

74
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. putrefaciens W3-18-1
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. putrefaciens W3-18-1
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):

Figure 25. S. putrefaciens W3-18-1, I-t and HPLC profiles for each MFC tested.

75
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
MFC1 (EIS) MFC2 MFC3
S. S. amazonensis SB2B SB2B
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
MFC1 (EIS) MFC2 MFC3
I (mA)
I (mA)
I (mA)
Organic acid concentrations:
Current Production (at max power):
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
I (mA)
MFC1 (EIS) MFC2 MFC3
S. S. amazonensis SB2B SB2B
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
2
4
6
8
10
12
14
16
18
20
0 50 100 150 200
Time (hrs)
Concentration (mM)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
MFC1 (EIS) MFC2 MFC3
I (mA)
I (mA)
I (mA)
Organic acid concentrations:
Current Production (at max power):

Figure 26. S. amazonensis SB2B I-t and HPLC profiles for each MFC tested.

76
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
MFC1 (EIS) MFC2 MFC3
S. Ana 3
MFC1 (EIS) MFC2 MFC3
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
Organic acid concentrations:
Current Production (at max power):
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
MFC1 (EIS) MFC2 MFC3
S. Ana 3
MFC1 (EIS) MFC2 MFC3
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
Organic acid concentrations:
Current Production (at max power):

Figure 27. S. ANA-3 I-t and HPLC profiles for each MFC tested.

77
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. oneidensis MR4
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. oneidensis MR4
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):

Figure 28. S. oneidensis MR-4 I-t and HPLC profiles for each MFC tested.

78
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
MFC1 MFC2 MFC3
S. oneidensis MR7
MFC1 MFC2 MFC3
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
Organic acid concentrations:
Current Production (at max power):
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
MFC1 MFC2 MFC3
S. oneidensis MR7
MFC1 MFC2 MFC3
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
Organic acid concentrations:
Current Production (at max power):

Figure 29. S. oneidensis MR-7 I-t and HPLC profiles for each MFC tested.

79
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
S. putrefaciens CN32
I (mA)
I (mA)
I (mA)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
S. putrefaciens CN32
I (mA)
I (mA)
I (mA)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.6)
formate (RT = 5.8)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):

Figure 30. S. putrefaciens I-t and HPLC profiles for each MFC tested.

80
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200 250 300 350
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200 250 300 350
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.5)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.5)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. loihica PV4
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200 250 300 350
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
0 50 100 150 200
Time (hrs)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200 250 300 350
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.5)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.8)
acetate (RT = 8.9)
pyruvate (RT = 6.5)
formate (RT = 5.8)
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0 50 100 150 200
Time (hrs)
lactate (RT = 7.0)
acetate (RT = 8.5)
pyruvate (RT = 5.8)
formate (RT = 5.4)
S. loihica PV4
I (mA)
I (mA)
I (mA)
[mM]
[mM]
[mM]
MFC1 (EIS) MFC2 MFC3
MFC1 (EIS) MFC2 MFC3
Organic acid concentrations:
Current Production (at max power):

Figure 31. S. loihica PV-4 I-t and HPLC profiles for each MFC tested.
The first lactate peak for each strain corresponds to the first lactate feeding that
took place prior to electrochemical evaluations including EIS and polarization resistance
measurements, when I-t data were not collected. The second and third lactate peaks
corresponded directly in time with the I-t curves generated while operating each MFC at
P
max
(Figure 22).
It is apparent that almost every Shewanella strain was able to oxidize lactate to
CO
2
with varying degrees of completion. For example, S. amazonensis SB2B was only
able to oxidize lactate (shown in blue) to acetate (shown in red), while S. putrefaciens
CN-32 and S. oneidensis MR-7 were able to oxidize lactate almost entirely to CO
2
over
the course of about one-hundred and fifty hours.

81
There were also some concentrations of formate and pyruvate apparent in each
electrolyte sample, shown in black and green respectively. However, the concentrations
of these organic acids remained stable and were fairly negligible relative to the acetate
concentrations for each stain. The exception to this trend was seen for S. loihica PV-4,
which demonstrated a fairly high concentration of formate in the anolyte samples from
MFC 1.
Scanning electron microscopy (SEM) images of the selected MFC anodes
exposed to each Shewanella strain are shown in Figure 32, Figure 33 and Figure 34.
Images are shown for anodes that were operated in System 1 and System 2, respectively.
It is apparent from these images that surface attachment of each strain was impacted by
the system into which they were inoculated. System 1 electrodes were scarcely populated
relative to System 2 electrodes, and the attached bacteria demonstrated morphologies
associated with stressed conditions, e.g., round globules (MR-1, W3-18-1, MR-4, MR-7
and CN-32) and end-to-end alignment of rod shaped cells (PV-4). S. ANA-3 was the only
strain that demonstrated a high density of rod-shaped bacteria and individual colonies
attached to the anode electrodes for System 1.

82
MR-1
10 µm
System 2
W3-18-1
10 µm
System 2
SB2B
10 µm
System 2
MR-1
20 µm
System 1
W3-18-1
20 µm
System 1
SB2B
20 µm
System 1
MR-1
10 µm
System 2
W3-18-1
10 µm
System 2
SB2B
10 µm
System 2
MR-1
20 µm
System 1
W3-18-1
20 µm
System 1
SB2B
20 µm
System 1

Figure 32. Scanning electron microscopy images of graphite fiber anode electrodes exposed to S.
oneidensis MR1, S. putrefaciens W3-18-1 and S. amazonensis SB2B, respectively, in System 1 and
System 2.

83
System 2
System 2
System 2
System 1
System 1
ANA-3
10 µm
MR-4
10 µm
MR-7
10 µm
ANA-3
20 µm
MR-4
10 µm
MR-7
20 µm
System 1
System 2
System 2
System 2
System 1
System 1
ANA-3
10 µm
MR-4
10 µm
MR-7
10 µm
ANA-3
20 µm
MR-4
10 µm
MR-7
20 µm
ANA-3
10 µm
MR-4
10 µm
MR-7
10 µm
ANA-3
20 µm
MR-4
10 µm
MR-7
20 µm
System 1

Figure 33. Scanning electron microscopy images of graphite fiber anode electrodes exposed to S.
ANA-3, S. oneidensis MR-4 and S. oneidensis MR-7, respectively, in System 1 and System 2.
CN-32
10 µm
System 2
PV-4
10 µm
System 2
CN-32
10 µm
System 1
PV-4
20 µm
System 1
CN-32
10 µm
System 2
PV-4
10 µm
System 2
CN-32
10 µm
System 1
PV-4
20 µm
System 1

Figure 34. Scanning electron microscopy images of graphite fiber anode electrodes exposed to S.
putrefaciens CN-32 and S. loihica PV-4, respectively, in System 1 and System 2.

84
System 2 anode electrodes were generally more populated along the fiber lengths,
relative to System 1, with several strains forming colonies sporadically across the surface
including MR-1, W3-18-1, SB2B, CN-32 and PV-4. Those strains that did not show a
preference for colony formation at the anode surface in System 2 did demonstrate
monolayers of cells dispersed across the fibers. The cell morphologies for the System 2
evaluations were also more consistent, with a majority of rod-shaped bacteria populating
each electrode. The exception to this trend was MR-7, which had more globular
structures attached to the anode fibers and rod-shaped bacteria interspersed as a
monolayer between the globules.
A relationship between the S. oneidensis strains MR-4 and MR-7 was apparent
from the OCP
a
values (Table 5), and the power and current density data for System 2
evaluations (Figure 22 b and d). On average the power and current density for strain
MR-7 was higher than that observed for MR-4, with MR-7 yielding approximately 50%
more power (0.56 ± 0.12 versus 0.30 ± 0.04 µW/cm
2
) and 26% more current density
(1.47 ± 0.27 versus 1.09 ± 0.34 µA/cm
2
)

than MR-4. Additionally, the OCP
a
values for
MR-7 were roughly 50 mV higher than for MR-4 in both system evaluations (shown in
Table 5).
These results are interesting because S. oneidensis strains MR-4 and MR-7 were
isolated from the same water column in the Black Sea, but at different depths. S.
oneidensis MR-4 was isolated in the suboxic zone (approximately 5 m below the
surface) and S. oneidensis MR-7 was isolated in the anoxic zone (approximately 60 m
below the surface). These organisms are thought to contribute to the Mn(IV)-reduction

85
and Mn(II)-oxidation processes that occur in the Black Sea by way of respiration
(Nealson, Myers et al. 1991; Dollhopf, Nealson et al. 2000). Given that the process of
electron transfer to solid electrodes appears to be most efficient in the absence of oxygen
(Biffinger, Byrd et al. 2008), it makes sense that an organism that is naturally adept at
respiring metal oxides in an anaerobic environment would be better suited to perform
well in an MFC relative to an organism that has adapted to use oxygen as its primary
electron acceptor. However, this logic does not seem to apply to the power density
results obtained for S. putrefaciens W3-18-1 and S. loihica PV-4. Interestingly, both
organisms were isolated from marine sediments in the Pacific Ocean with S.
putrefaciens W3-18-1 isolated at 630 m below the ocean surface off the Washington
State Coast and S. loihica PV-4 isolated at 1,325 m below the surface by a hydrothermal
vent on the Hawaiian seamount. S. loihica PV-4 was found in an iron-rich microbial
mat, and is able to reduce a number of metal-oxides, which would suggest that PV-4
would be an ideal organism to use as a MFC catalyst; however, the results presented
here suggest otherwise.
The lack of power and current density by S. loihica PV-4 may be attributed to
MFC system limitations rather than biological limitations as was discussed above.
Additionally, the conditioning time required to achieve stable OCV values for these
MFCs may impact how each strain performs, i.e., during periods when no current is
allowed to flow across the circuit, the bacteria are limited in respiratory abilities and
those strains that are sensitive to electron acceptor limitations may be quickly affected.

86
A twelve hour open-circuit period may be too long for some strains to survive with
restricted respiration capability, especially PV-4.
For a direct view of all the evaluations performed at KIST and USC, the current
densities achieved for each system, near closed-circuit values, i.e., across a 10-ohm
resistor are compared in Table 6 for each strain tested. It is apparent that I
max
was most
impacted by the strong phosphate buffer concentrations in the USC, System 1
evaluations and that the USC, System 2 evaluations yielded the best I
max
values for each
strain. These data suggest that when the current production abilities of different strains
are evaluated and compared, that the buffer composition and inoculation strategies make
a significant difference and should be taken under careful consideration. The aerobically
grown bacterium inoculated into a 50 mM PIPES buffer were able to produce the
highest levels of I
max
.
Table 6. Current densities at I
max
for each strain evaluated in three different MFC systems. Highest
values are shown in dark grey, the next highest values are highlighted in light grey, and the lowest
values are shown in white.
Current density at I
max
for each system (µA/cm
2
)
Strain
KIST
System
USC,
System 1
USC,
System 2
S. oneidensis MR-1 3.05 ± 0.97 2.29 ± 1.37 3.21 ± 0.34
S. putrefaciens W3-18-1 N/A 2.89 ± 0.82 5.96 ± 2.23
S. amazonensis SB2B 2.44 ± 0.97 2.01 ± 1.46 3.71 ± 0.92
S. ANA-3 N/A 3.08 ± 0.66 3.71 ± 0.43
S. oneidensis MR-4 N/A 1.77 ± 0.08 2.24 ± 0.50
S. oneidensis MR-7 N/A 0.99 ± 0.88 3.39 ± 0.77
S. putrefaciens CN-32 2.64 ± 1.04 1.37 ± 0.20 3.43 ± 0.68
S. loihica PV-4 1.68 ± 0.63 1.22 ± 0.29 1.72 ± 1.02

87
3.3 Electrochemical evaluations of strain performance

A subset of the evaluated strains was more rigorously explored using different
electrochemical techniques including EIS, CV and potentiodynamic polarization. Strains
S. oneidensis MR-1, S. putrefaciens W3-18-1 and S. loihica PV-4 were selected for
additional study because these strains exhibited the highest (W3-18-1), lowest (PV-4) and
reference (MR-1) power densities (Figure 22a and b).
3.3.1 ElS

The spectra obtained from anode and cathode EIS measurements during different
strain evaluations appeared to demonstrate an inductive behavior in the frequency range
of 10 to 100 Hz as seen in Figure 35.

88
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 012345
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
MR-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
a) MR-1 anode
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
MR-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
b) Pt cathode, MR-1 test
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 01234 5
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
W3-18-1 at anode
Log |Z| (ohm)
Phase angle (degrees)
c) W3-18-1 anode
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
W3-18-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
d) Pt cathode, W3-18-1 test
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 01234 5
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
PV-4 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
e) PV-4 anode
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
Buffer only at anode
PV-4 at anode
Log f (Hz)
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
f) Pt cathode, PV-4 test
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 012345
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
MR-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
a) MR-1 anode
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
MR-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
b) Pt cathode, MR-1 test
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 01234 5
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
W3-18-1 at anode
Log |Z| (ohm)
Phase angle (degrees)
c) W3-18-1 anode
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
W3-18-1 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
d) Pt cathode, W3-18-1 test
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 01234 5
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
Buffer only at anode
PV-4 at anode
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
e) PV-4 anode
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
-3 -2 -1 0 1 2 3 4 5
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
Buffer only at anode
PV-4 at anode
Log f (Hz)
Log f (Hz)
Log |Z| (ohm)
Phase angle (degrees)
f) Pt cathode, PV-4 test

Figure 35. Electrochemical impedance spectroscopy data for the anode and cathode of MFCs
operated with sterile anodes or an anode bacterial catalyst including a) S. oneidensis MR-1, c) S.
putrefaciens W3-18-1 and e) S. loihica PV-4. Figures b), d) and f) are the sterile cathode spectra
recorded under different anode conditions for the corresponding bacterial catalyst. All spectra were
collected at the open-circuit potentials (vs. Ag/AgCl) of the anode and cathode, respectively.
Inductive behaviors exhibited in the high frequency ranges are typically due to
instrumentation, electrode or wire properties and can be seen in complex plane plots,
typically in the fourth quadrant (Barsoukov and Macdonald 2005). Inductive behaviors
in the low frequency range of impedance spectra have been observed in polymer

89
electrolyte membrane (PEM) fuel cells and are attributed to the adsorption of
intermediate chemical species to the surface of the anode electrode (Ciureanu and Wang
1999; Barsoukov and Macdonald 2005). Furthermore, the inductive behavior at low
frequencies is dependent on potential, and the inductive behavior will change based on
the potential at which spectra are collected (Bai and Conway 1991; Bai and Conway
1993; Darowicki 1997).
This behavioral shift is apparent in Figure 35a through f and is most especially
pronounced in the anode spectra measured during sterile (buffer only) and inoculated
anode conditions. The inductance behavior of the inoculated anode spectra shifted
slightly relative to the sterile anode spectra, given that the presence of bacteria at the
anode changed the E
a
OCP
for each system by nearly 400 mV.
The spectra presented in Figure 35 can be modeled according to an equivalent
circuit proposed by Harrington and Conway (Harrington and Conway 1987) shown in
Figure 36. R
s
represents the solution resistance, R
p
is the polarization resistance, C
dl
is
the double-layer capacitance, R
o
is the resistance that changes in response to phase angle,
and L is the inductance. The R
o
and L arm of the circuit is related to the
pseudocapacitance which is a function of frequency and models part of the current
response which occurs as phase angle changes (Harrington and Conway 1987).

90
R
p
R
o
L
R
s
C
dl
R
p
R
o
L
R
s
C
dl

Figure 36. Equivalent circuit for impedance spectra showing inductance behavior and low
frequencies
Fit parameters have not been obtained by applying this equivalent circuit to the
collected EIS data; however given the correct circuit elements, some general observations
can be made.
The anode spectra show that the addition of each strain to the anode compartment
generated a severe reduction in R
p
relative to the sterile anode conditions. MR-1 and
W3-18-1 showed the largest changes in R
p
relative to the sterile conditions, with R
p

values decreasing over an order of magnitude. The R
p
decrease associated with PV-4 was
considerably less than what was observed for the other strains. These results corroborate
the power and current density data, which indicate that MR-1 and W3-18-1 are better at
producing power in MFCs than PV-4 under these operating conditions. Lower R
p
values
are typically associated with lower activation energies and higher redox rates.
The anode capacitance remained relatively unchanged between the sterile and
inoculated conditions; however the inductance shifts indicate a change in potential
between these anode conditions, which is found to be true. The OCP values for these
data and the other strains are listed in Table 5.
Some shifts in the inductance behavior of the MR-1 cathode were observed and
this is also likely due to a decrease in cathodic potential. The cathode potentials all

91
decreased slightly after the addition of bacteria to the anode, this may be due to crossover
of the lactate that was delivered as the fuel.
Expectedly, The R
p
values for the cathodes did not change with time or with the
addition of bacteria to the anode; however the cathode R
p
values were all an order of
magnitude lower than the anode R
p
values, even after bacteria were present at the anode.
These data indicate that the reduction rates of oxygen catalyzed by platinum were much
higher than the rates of lactate oxidation by any of the Shewanella strains.
The capacitance relationships between the cathodes and anodes also demonstrated
a physical difference between the electrodes. All the cathode capacitance values were
significantly higher than the anode values, indicating that the cathodes featured a much
higher surface area than the anodes, which is reflective of the platinum deposits at the
cathode surface.
3.3.2 Potentiodynamic polarization

S. oneidensis MR-1, S. putrefaciens W3-18-1 and S. loihica PV-4 were also
evaluated using potentiodynamic polarization to elucidate the limiting current densities
available from these strains operating as MFC anode catalysts. Potentiodynamic
polarization was done at the termination of a set of experiments, due to the potentially
destructive consequences that extreme polarization might have on the MFC biofilms. It
is also important to note that because these scans were performed at the end of a suite of
evaluations, that the cell viability in the MFC was likely much different than after it was
first inoculated.

92
Figure 37 shows the resulting potentiodynamic curves. These curves demonstrate
that the limiting currents available from each strain and Pt-cathode configuration also
correspond to the trends observed in power density and polarization resistance. S.
putrefaciens W3-18-1 had the highest limiting current density (1.07E-2 mA/cm
2
),
followed by S. oneidensis MR-1 (7.52E-3 mA/cm
2
) and S. loihica PV-4 (1.31E-3
mA/cm
2
).

93
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00 1.0E+01
Cathode
Anode w/buffer only
Anode w/MR-1
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
cathode
Anode w/buffer only
Anode w/W3-18-1
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
Cathode
Anode w/buffer only
Anode w/PV-4
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
a) MR-1, i
limit
= 7.52e-3 mA/cm
2
b) W3-18-1, i
limit
= 1.07e-2 mA/cm
2
c) PV-4, i
limit
= 1.31e-3 mA/cm
2
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00 1.0E+01
Cathode
Anode w/buffer only
Anode w/MR-1
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
cathode
Anode w/buffer only
Anode w/W3-18-1
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
Cathode
Anode w/buffer only
Anode w/PV-4
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.0E-06 1.0E-05 1.0E-04 1.0E-03 1.0E-02 1.0E-01 1.0E+00 1.0E+01
Cathode
Anode w/buffer only
Anode w/MR-1
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
cathode
Anode w/buffer only
Anode w/W3-18-1
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
-1.20
-0.80
-0.40
0.00
0.40
0.80
1.20
1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01
Cathode
Anode w/buffer only
Anode w/PV-4
E vs Ag/AgCl (V)
Log i (mA/cm
2
)
a) MR-1, i
limit
= 7.52e-3 mA/cm
2
b) W3-18-1, i
limit
= 1.07e-2 mA/cm
2
c) PV-4, i
limit
= 1.31e-3 mA/cm
2

Figure 37. Potentiodynamic scans of the anode and cathode for sterile anode conditions and after a) S.
oneidensis MR-1, b) S. putrefaciens W3-18-1, and c) S. loihica PV-4 had been inoculated to the MFC
anode. All scans were performed at a rate of 0.167 mV/sec vs Ag/AgCl.

94
3.3.3 Cyclic voltammetry

Cyclic voltammograms can yield information about the current-potential behavior
of an electrode in an aqueous solution; however the results for subsequent CV
evaluations of the same system can be very different depending on electrode properties,
scan rates, the presence of organic substrates, and the propensity for chemisorption of
different reactive species to electrode surfaces. Therefore the interpretation of cyclic
voltammograms can be complex. The CV results for MR-1, W3-18-1 and PV-4 are
shown in Figure 38 and indicate that each strain exhibits different redox behaviors.
However, these peaks and further interpretations may be better explored using different
scan rates, electrodes and energy sources. For example, formic acid has been shown to
yield a different redox response on platinum electrodes depending on electrolyte
composition (Hamann, Hamnett et al. 1998). The HPLC data for these strains all show
that different organic acids, including formate, are present in the MFC anode
compartments during testing and these compounds may complicate the interpretation of
the CV peaks.

95
E vs Ag/AgCl (V)
I (mA)
-60
-40
-20
0
20
40
60
-1.0 -0.5 0.0 0.5 1.0
MR-1
W3-18-1
PV-4
E vs Ag/AgCl (V)
I (mA)
-60
-40
-20
0
20
40
60
-1.0 -0.5 0.0 0.5 1.0
MR-1
W3-18-1
PV-4

Figure 38. Cyclic voltammograms for S. oneidensis MR-1, S. putrefaciens W3-18-1, and S. loihica PV-
4 collected at a scan rate of 25 mV/sec. MR-1 and W3-18-1 scans were collected in a range of -750
mV to +700mV and PV-4 scans were collected from -750 mV to 850 mV vs. Ag/AgCl.

96
Chapter 4: Microbial fuel cells as a tool for studying microbial
physiology

Microbial fuel cells have been suggested as a viable technology for many
different applications including wastewater treatment (Suzuki, Karube et al. 1978),
hydrogen production (May, Blanchard et al. 1964), power production (Bennetto 1987),
bioremediation (Gregory and Lovley 2005) and biological oxygen demand (BOD)
sensors (Kim 1999). However, MFCs may also play a key role as tools for understanding
microbial physiology and ecology.
The earliest studies of electrical signals induced by microbial respiration and
fermentation by Potter and Cohen (Potter 1911; Cohen 1931) suggested the relationship
between bioelectrochemical reactions and the metabolic activities of different
microorganisms. To date, several research groups have analyzed the differences in
microbial populations found at the anodes and cathodes of sediment batteries deployed in
different habitats (Reimers, Tender et al. 2001; Tender, Reimers et al. 2002; Nielsen,
Reimers et al. 2007). Additionally, it is now common to analyze the microbial
communities associated with power production from different wastewater streams and to
study the community changes between different MFC operational conditions for these
systems (Gil, Chang et al. 2003; Lee, Phung et al. 2003; Back, Kim et al. 2004; Kim,
Park et al. 2004; Phung, Lee et al. 2004; Kim, Jung et al. 2007). However, little research
has been devoted to understanding the specific electrical responses associated with
changes to pure culture growth conditions and pure culture MFC operational conditions.
Additionally, the concept of a pure culture bioanode and biocathode employed with
different terminal oxidants at the cathode to study the energy metabolism of a pure

97
culture has not been explored. To this end, several MFC experiments were conducted
using Shewanella oneidensis MR-1 to evaluate how: 1) planktonic growth conditions
related to MFC performance; 2) carbon sources affected current production; and, 3) an
ex-situ grown MR-1 biofilm would enhance current densities. Additionally, MR-1 wild
type and mutants were utilized as biocatalysts at the cathode of a MFC, with different
terminal oxidants, to evaluate the diverse energy metabolism of Shewanella oneidensis
MR-1.
4.1 Microbial fuel cell evaluation of Shewanella oneidensis MR1 grown
in bioreactors with different dissolved oxygen tensions

The ability of S. oneidensis MR-1 to directly transfer electrons to metal oxides
occurs under anaerobic conditions for metal oxides (Myers and Nealson 1988; Myers and
Nealson 1988; Myers and Nealson 1990; Hernandez and Newman 2001; Newman 2001).
However, a study using solid Fe(III)-oxide, has shown that the reduction of solid metal
oxides is slower when MR-1 is grown aerobically prior to metal oxide exposure, than
when the cells are grown anaerobically on fumarate or soluble Fe(III)-citrate (Lies,
Hernandez et al. 2005). Although aerobically grown MR-1 demonstrated an initial lag
time in the study by Lies et al., it was also shown that these cultures would eventually
achieve reduction rates comparable to anaerobically grown cells. Lies et al., additionally
studied the impact that cell density had on Fe(III)-oxide reduction and it was found that
Fe(III) reduction decreased linearly with decreasing cell densities (Lies, Hernandez et al.
2005).

98
The reduction of Fe(III)-oxides by MR-1 and other organisms requires
extracellular electron transfer, which is a process also believed to be required for using an
MFC anode as an electron acceptor (Bond and Lovley 2003; Holmes, Bond et al. 2004;
Gorby, Yanina et al. 2006; Holmes, Chaudhuri et al. 2006).
Changes in MR-1s physiology in response to decreasing oxygen tensions in the
growth environment has also been observed by Gorby et al. (Gorby, Yanina et al. 2006).
It was found that as electron acceptor availability is limited, i.e., low DOT, the microbes
begin upregulating c-type cytochromes and form nanowires. Therefore, it was expected
that a microbe grown in a more oxygen limited environment will be better poised to give
electrons directly away to MFC anodes. Alternatively, those microbes grown in a more
oxygen rich environment will not be immediately ready to give electrons away to an
electrode and will produce less current relative to those microbes grown at more severe
oxygen limitations.
MFC tests were performed using S. oneidensis MR-1 grown under different
oxygen limitations to test these hypotheses. Cultures were grown at PNNL in
collaboration with Yuri Gorby using bioreactors defined to produce MR-1 cultures at two
oxygen limited conditions: 1) 50% (4 mg O
2
/L) dissolved oxygen tension (DOT) and 2)
5% (0.4 mg O
2
/L) DOT. Cells from each growth condition were inoculated into six dual-
compartment MFCs and the corresponding current productions were monitored over
time.

99
4.1.1 Materials and Methods

Six dual-compartment MFCs were utilized throughout testing. The MFCs were
assembled using graphite felt anodes and platinized graphite felt cathode electrodes (17.9
cm
2
apparent surface area) with the anode and cathode compartments separated by a
Nafion 424 proton exchange membrane.
Each MFC was autoclaved and the anode and cathode compartments were
injected with 30 ml of a sodium-phosphate buffer (50mM phosphate, 100mM sodium
chloride, pH 7.0) and the anode compartment was continuously flushed with sterile
filtered N
2
. Air was flushed continuously through the cathode compartment.
The voltage difference, across a 10 ohm resistor, between the anode and cathode
electrodes was recorded using a digital multimeter (Keithley Instruments, 2700). Data
were collected every five minutes over a two-day period, and used to calculate the
corresponding stable current densities. Data were recorded without MR-1 at the anode
compartment to establish a baseline cell voltage.
The MR1 wild type strain was cultivated in two bioreactors using a medium
containing 12 mM sodium lactate, 0.68 mM calcium chloride, 80 µM ferric
nitrilotriacetic acid (Fe-NTA), 30 µM manganese chloride and 1 µM sodium selenate plus
vitamins and trace mineral solutions.
Both bioreactors held a constantly limited lactate concentration, a pH of 7.03 and
an optical density (OD) of about 0.3 (at 600 nm). The dissolved oxygen tension (DOT)

100
was different in each bioreactor. One system was held at a 5% DOT (~ 0.4 mg O
2
/L) and
the other at a 50% DOT (~ 4 mg O
2
/L).
Upon establishing a baseline voltage, 15 ml of buffer was removed from each MFC and
replaced with 15 ml of MR-1 (plus media) from the bioreactor. The final optical density
at 600 nm (OD
600
) of each fuel cell was estimated to be 0.15 (approximately 5 x 10
7

cells/mL).
Approximately 1 mM of lactate was added to each MFC immediately after the
injection of MR-1 and after the first voltage minima occurred. Nitrogen flow into the
MFCs was halted after the initial addition of lactate and all gas ports at the anode were
clamped shut to prevent gas flux to and from the anode compartment.
4.1.2 Results and Discussion

The expectation that MR-1 grown in a severely oxygen limited environment
produces more current than MR-1 grown in a more oxygen rich environment was tested
and shown to be true for the discussed set of experiments. Figure 39 shows the average
stable current densities achieved for the 5% and 50% DOT growth conditions and two
consecutive lactate feedings.

101
0.0
0.5
1.0
1.5
2.0
2.5
Stable current density (µA/cm
2
)
5% DOT 50% DOT 5% DOT 50% DOT
Lactate, Feed 1 Lactate, Feed 2
0.0
0.5
1.0
1.5
2.0
2.5
Stable current density (µA/cm
2
)
5% DOT 50% DOT 5% DOT 50% DOT
Lactate, Feed 1 Lactate, Feed 2

Figure 39. Average stable current densities from six MFCs inoculated with a 5% or 50% DOT grown
MR-1 culture. Data for both cultures are shown for two lactate feeds, and the averages and standard
deviations were calculated using seven hours of data corresponding to the stable current densities for
each feeding.
These data show that the 5% DOT cultures consistently produced higher current
densities than the 50% DOT cultures. Additionally, the second lactate feeding resulted in
higher current densities for both cultures; however, the 50% DOT culture showed a much
higher increase in current density during the second lactate feeding than the 5% DOT
culture.
The results indicate that MR-1 grown at a lower DOT will generate more current
density in an MFC than a culture grown at a higher DOT. Additionally, the deviations of
current densities produced during the first and second lactate feedings for the 5% DOT
cultures are not significant, indicating that the maximum current producing abilities of
the 5% DOT culture is achieved immediately upon MFC inoculation.
It is evident from average current densities and deviations for the 50% DOT
culture that current densities produced by MR-1 grown under suboxic conditions will

102
eventually adapt to anaerobic conditions and produce current densities that are equivalent
to those produced initially from 5% DOT cultures. These results correspond to the
adaptability of aerobically grown MR-1 under Fe(III)-reducing conditions found by Lies
et al. (Lies, Hernandez et al. 2005).

4.2 Shewanella oneidensis MR-1 power production using different
reductants

An additional study was conducted using different electron donors after each
culture, grown at 50% DOT, had adapted to the anaerobic MFC conditions. Several
organic acids including acetate and succinate were added as electron donors to the MFC
anode after lactate had been completely consumed from the previous batch studies. This
was done as a preliminary test to determine the fuel flexibility of MR-1 acting as a
catalyst for the oxidation of organic compounds.
An additional study was conducted using formate and pyruvate; however these
experiments were unintentionally performed using reactor cultures grown at 5% DOT,
and much higher cell densities. The formate and pyruvate data were collected with a cell
density of approximately 2 x 10
10
cells/mL (OD = 0.8) in the MFC anode compartment,
as compared to the acetate and succinate data, which were conducted with approximately
2 x 10
7
cells/mL (OD = 0.15). Additionally, all of the MFCs fed lactate and pyruvate
lacked a consistent air flow to the cathode, relative to the other electron donor evaluations.
Figure 40a through d shows the current response, in duplicate MFCs, of each
electron donor with time. These results indicate that MR-1 cultures grown at 50% DOT

103
are able to utilize acetate or succinate to produce a current under anaerobic MFC
conditions. MR-1 also appears to be capable of successfully utilizing formate to produce
current in an MFC when grown at 5% DOT; however, pyruvate does not appear to yield a
current from MR-1 under these conditions.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
a) Lactate and Acetate, OD = 0.15 b) Lactate and Succinate, OD = 0.15
c) Lactate and Formate, OD = 0.8 d) Lactate and Pyruvate, OD = 0.8
Lactate
Lactate
Lactate
Lactate
Acetate Succinate
Formate
Pyruvate
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
a) Lactate and Acetate, OD = 0.15 b) Lactate and Succinate, OD = 0.15
c) Lactate and Formate, OD = 0.8 d) Lactate and Pyruvate, OD = 0.8
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
0 5 10 15 20 25 30 35
MFC1
MFC2
Stable current density (µA/cm
2
)
Time (hrs)
Stable current density (µA/cm
2
)
Time (hrs)
a) Lactate and Acetate, OD = 0.15 b) Lactate and Succinate, OD = 0.15
c) Lactate and Formate, OD = 0.8 d) Lactate and Pyruvate, OD = 0.8
Lactate
Lactate
Lactate
Lactate
Acetate Succinate
Formate
Pyruvate

Figure 40. I-t curves for duplicate MFCs featuring MR-1 grown at 50% DOT (OD
600
= 0.15) and fed
lactate as the initial carbon sources and a) acetate or b) succinate. Figures c) and d), respectively
show I-t curves for duplicate MFCs featuring MR-1 grown at 5% DOT (OD
600
= 0.8) and fed lactate
as the initial carbon source and either formate or pyruvate as the final carbon source. The MFCs fed
pyruvate, d), were also operating with a lower dissolved oxygen tension at the cathodes relative to
those MFCs fed other carbon sources, a), b) and c), respectively.
Additionally, it is apparent from these results that higher cell densities present in
the MFC anodes produced a higher current density when utilizing lactate as a carbon
source. Those MFCs inoculated with 2 x 10
10
cells/mL at the anode (Figure 40c) were

104
able to produce nearly six times more current than those MFCs inoculated with 2 x 10
7

cells/mL (Figure 40a and b) during the second lactate feeding.
The data presented in Figure 40d indicate that current production in an MFC is
not only limited by carbon source and anode activity, but also the amount of oxygen that
is present in the MFC cathode. Even though a high cell density was present in the MFC
anodes of these experiments, the oxygen diffusion at the cathode was too low to enable
comparable current densities to the MFCs presented in Figure 40c.
4.3 Shewanella oneidensis MR-1 power production with an ex-situ
grown biofilm

An MR-1 biofilm was grown directly onto an MFC anode electrode outside of the
MFC, to test the idea that more biomass at the anode of an MFC will result in higher
current densities. A graphite felt anode was placed horizontally into a tube, sterilized,
and inoculated with 1 mL of MR-1 cells harvested from a bioreactor operated at 5% DOT.
The anode was then exposed to a constant drip of sterile media featuring 12 mM of
sodium lactate and 20 mM of fumarate, trace salts, vitamins and minerals for four days
time. After the drip feed of media was terminated a visible biofilm could be
distinguished in the tube under the MFC anode.
An MFC was sterilized and assembled using the biofilm covered anode, a sterile
cathode and a Nafion 424 membrane separating the anode and cathode compartments.
The MFC was then filled with 30 mL of a phosphate buffer (described above) and
injected with 20 mM of lactate. After the lactate injection, the MFC was operated using a
10 ohm resistor and the voltage difference across the load was monitored as described

105
above. The current density produced from this system was calculated, and after about
one week of MFC operation, the anode was removed and imaged using a scanning
electron microscope (SEM) at PNNL. The results from the first lactate feed (Figure 41)
showed an order of magnitude increase in current density (~14 µA/cm
2
) relative to MFCs
operated with a high density (2 x 10
8
cells/mL) of planktonic cells (~3 µA/cm
2
),
consistent with the results shown in Figure 40a through c. Consecutive lactate feedings
to the ex-situ grown biofilm resulted in similar current densities near 9 µA/cm
2
.
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 5 10 15 20 25 30
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 5 10 15 20 25 30
Stable current density (µA/cm
2
)
Stable current density (µA/cm
2
)
Time (hrs) Time (hrs)
a) Ex-situ grown anode biofilm b) In-situ grown anode biofilm
5 µm 5 µm
c) Ex-situ grown anode biofilm d) In-situ grown anode biofilm
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 5 10 15 20 25 30
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 5 10 15 20 25 30
Stable current density (µA/cm
2
)
Stable current density (µA/cm
2
)
Time (hrs) Time (hrs)
a) Ex-situ grown anode biofilm b) In-situ grown anode biofilm
5 µm 5 µm
c) Ex-situ grown anode biofilm d) In-situ grown anode biofilm

Figure 41. I-t curves for the first thirty hours of operation of a) an ex-situ grown, anode biofilm; and
b) an in-situ grown anode biofilm. Scanning electron microscopy images of c) ex-situ grown, anode
biofilm; and d) in-situ grown anode biofilm. Each MFC was operated using a 10 ohm resistor and
only lactate was provided in the MFC anode compartment as a carbon source and electron donor.
The in-situ grown biofilm was achieved by exposing the anode electrode to approximately 2 x 10
9

cells/mL of planktonic MR-1 and a total of 6 mM of lactate.

106
It is evident from the SEM images in Figure 41c and d that the ex-situ grown
biofilm demonstrated better coverage of the MFC anode graphite fibers and a highly
developed network of extra-cellular appendages, relative to the in-situ grown biofilm.
Given that no SEM images were taken of the ex-situ grown biofilm prior to MFC
inoculation, it is unknown whether these extra-cellular appendages existed before MFC
operation or evolved during the process of electron transfer to the MFC anode.
It has been speculated, but unconfirmed, that the increased presence of these
extra-cellular appendages is related to the observed increase in current density. These
appendages were not evaluated in terms of their conductive properties and therefore
cannot be conclusively discussed as nanowires. Additionally, no SEM observations were
made of the MFC anodes that corresponded to current densities of nearly 14 µA/cm
2
after
a planktonic inoculation of a high cell density (2 x 10
10
cell/mL) shown in Figure 40c, so
it is unknown if these same appendages existed at this system. However the current
densities achieved from the inoculation of high planktonic cell densities and the ex-situ
grown biofilm, in addition to the SEM images from Figure 41c and d, imply that a higher
number of bacteria attached to the MFC anode should resulted in increased current
densities.
4.4 Shewanella oneidensis MR-1 wild-type and mutants acting as
cathode catalysts in an MFC with different oxidants

Microbial fuel cell (MFC) systems employ microorganisms as catalysts to drive
electrochemical reactions and produce electricity (Bennetto, Stirling et al. 1983). One of
the advantages associated with using MFCs as power sources is the fuel flexibility due to

107
the diverse metabolic and catalytic abilities of microorganisms. The variety of fuels that
can be utilized in MFC devices has led to the exploration of these systems for wastewater
treatment, hydrogen production, and bioremediation.
To optimize MFC systems for different applications, it is necessary to understand
the limitations of catalytic activity and the mechanism(s) of electron transfer employed
by bacteria and other microorganisms. To this end, several groups have studied the use
of different bacteria and communities of microorganisms to drive the oxidation of organic
and inorganic substrates while using a MFC anode as an electron acceptor. More
recently, researchers have used bacteria and other microorganisms to catalyze the
reduction of soluble terminal electron acceptors such as oxygen, nitrate or fumarate while
using the solid cathode electrode as an electron donor.
Most of these studies have been conducted using communities of microorganisms
enriched from wastewater or sediments (López-López, Expósito et al. 1999; Bergel,
Féron et al. 2005; Zhang, Jia et al. 2005; He and Angenent 2006; He, Wagner et al. 2006;
Lowy, Tender et al. 2006; Rabaey, VandeSompel et al. 2006; ter Heijne, Hamelers et al.
2006; Clauwaert, Rabaey et al. 2007; Clauwaert, van der Ha et al. 2007; Zuo, Cheng et al.
2007; Chen, Choi et al. 2008; Freguia, Rabaey et al. 2008; Rabaey, Read et al. 2008;
Rozendal, Jeremiasse et al. 2008). A few researchers have studied the abilities of pure
cultures including Thiobacillus ferroxidans (López-López, Expósito et al. 1999),
Acidithiobacillus ferrooxidans (ter Heijne, Hamelers et al. 2007), and Leptothrix
discophora (Rhoads, Beyenal et al. 2005) to biomineralize ferric iron (T. ferroxidans and
A. ferroxidans) and manganese oxides (L. discophora) at the cathode of MFC systems.
Additionally, Geobacter metallireducens (Gregory, Bond et al. 2004) and Geobacter

108
sulfurreducens (Gregory, Bond et al. 2004) have been used as cathode catalysts that
utilize the graphite electrode as an electron donor and reduce nitrate to nitrite (G.
metallireducens), or fumarate to succinate (G. sulfurreducens). However, a majority of
these studies utilized a potentiostat or sacrificial anode to apply a constant current to the
cathode electrode, thereby generating an artificial electron source as the electron donor to
the biotic reduction reaction.
Those studies that did utilize bacterial cultures at the anode to generate current in
the MFC used either wastewater enrichment cultures featuring many different
microorganisms (Bergel, Féron et al. 2005; Goel and Flora 2005; Zhang, Jia et al. 2005;
He, Wagner et al. 2006; Rabaey, VandeSompel et al. 2006; Clauwaert, Rabaey et al.
2007; Clauwaert, van der Ha et al. 2007; Zuo, Cheng et al. 2007; Chen, Choi et al. 2008;
Freguia, Rabaey et al. 2008; Rabaey, Read et al. 2008; Rozendal, Jeremiasse et al. 2008)
or a different pure bacterial culture than what was used at the cathode (Rhoads, Beyenal
et al. 2005). Although all of these studies demonstrated the successful use of biocathodes
for different applications, little is yet known about the metabolic abilities of a pure
bacterial culture acting as both the oxidation and reduction catalyst in a MFC system.
This work introduces, for the first time, a MFC that generates power using the
same pure bacterial culture as the anode and cathode catalyst using three different soluble
oxidants at the cathode: fumarate, Fe(III)-citrate, and oxygen. These power data for S.
oneidensis MR-1, along with an analysis of the metabolic products generated in each
operating MFC, show that the stable current densities observed for each oxidant at the
cathode correspond to the free energies of each redox couple tested.

109
In addition to evaluating the S. oneidensis MR-1 (WT) response to different
oxidants at the cathode, several single-deletion MR-1 mutants were also tested as MFC
cathode catalysts including those deficient in a tetraheme cytochrome c fumarate
reductase, ∆fccA; a diheme cytochrome c oxidase, ∆cyoA; a tetraheme cytochrome c
involved with anaerobic respiration, ∆cymA; and an outer membrane protein precursor
involved with metal-oxide reduction, ∆mtrB. All of these mutants have been previously
described and tested for current-producing abilities at the anode of a MFC (Bretschger,
Obraztsova et al. 2007); however, this work presents the first data showing how these
mutants operate as cathode catalysts in the MFC with fumarate (∆fccA, ∆cymA and
∆mtrB), Fe(III)-citrate (∆mtrB), or oxygen (∆cyoA and ∆mtrB) as the oxidant.
4.4.1 Materials and Methods

The metabolic flexibility and electrical current generation of S. oneidensis MR-1
as an anode and cathode catalyst were tested in collaboration with Lewis Hsu and Carie
Frantz.
Dual-compartment MFC systems were used, featuring bare graphite felt attached
to platinum wire leads as the anode and cathode electrodes. The anode and cathode
chambers were physically separated by a proton exchange membrane (DuPont, Nafion
117). Anaerobic conditions were maintained in the anode compartments by continuous
purging with nitrogen at a flow rate of 20 mL/min; this was also done in the cathode
compartments when evaluating fumarate and Fe(III)-citrate (FeC) as oxidants at the
cathode. When oxygen was being evaluated as an oxidant, air was purged into the
cathode compartment at a flow rate of 40 mL/min.

110
All MFC experiments were conducted in triplicate using lactate as the reductant at
the anode and one of three oxidants at the cathode. Controls were run in parallel with
each experiment including: 1) a sterile anodic compartment with only lactate present in
the anolyte, and bacteria at the cathode with the given oxidant; and 2) WT and lactate at
the anode, and only the given oxidant in a sterile cathode compartment. The controls
were designed to gain some understanding about the abiotic affects of lactate oxidation at
the anode, and oxidant reduction at the cathode, respectively.
All MFCs were operated using an applied load of 10 Ohm. The voltage
difference across the load was monitored every five minutes for sixty hours, or until all of
the lactate had been oxidized at the anode. Average current densities were calculated
using the apparent electrode surface area (79 cm
2
) and the average stable current (I),
observed for twelve hours, that corresponded to the voltage drop across the resistor (R),
according to Ohms law.
MR-1 WT and mutants were grown aerobically in a buffered minimal media (pH
7.0) as described by Bretschger et. al (Bretschger, Obraztsova et al. 2007), except that a 7
mM lactate concentration was utilized instead of 18 mM to ensure that the culture was
both electron donor and acceptor limited when entering late stationary phase. This was
verified by chemical analysis of the growth media prior to inoculation. Cells were
harvested during late stationary phase and inoculated into the MFC anode and cathode
compartments such that a cell density of approximately 1 x 10
8
cells/mL was achieved in
each compartment. A PIPES buffer (Bretschger, Obraztsova et al. 2007) was used as the
diluting solution in each compartment. MR-1 WT was inoculated at the anode for each
experiment and 2 mM of lactate was added as an electron donor. Either WT or a mutant

111
was added to the cathode compartment with either 4 mM fumarate, 4 mM FeC or air
diffusing through the compartment as the oxidant.
Anolyte and catholyte samples were taken periodically during MFC operation and
analyzed using a high pressure liquid chromatography (HPLC) machine (Agilent 1100
series) with a reversed phase C18 column (Phenomenex, Synergi-Hydro).
Chromatographs were generated using a diode array detector (Agilent 1100 series) set to
the wavelength of 210 nm. A 20µL sub-sample of each electrolyte was injected into the
sampling loop and the analytical method employed a 2.5 mM sulfuric acid mobile phase
(pH 2.0) running at a flow rate of 0.5 mL/min. Peak areas were obtained for all the peaks
in each electrolyte chromatogram and concentrations were calculated based on the peak
areas and retention times of known standards for each evaluated organic compound.
4.4.2 Results and Discussion

The results for all of the WT and mutant experiments for each oxidant at the
cathode are shown in Figure 42. It is apparent that the stable current densities for all of
the WT and ∆mtrB cathodes follow a trend that reflects the energy gained from each
reduction reaction. Specifically, the oxygen and FeC cathodes had the highest current
densities reflecting the reduction potential of oxygen, O
2
/H
2
O (E
o

= 0.82 V), and
Fe
3+
/Fe
2+
(E
o

= 0.33 V) under standard conditions. Finally, the lowest current densities
resulted from the reduction of fumarate to succinate (E
o

= 0.03 V) by WT and ∆mtrB at
the MFC cathode relative

(Thauer, Jungermann et al. 1977).

112
Fe(III)-citrate cathode
(no bacteria)
0.00
0.25
0.50
0.75
1.00
1.25
1.50
WT cathode
current density (µA/cm
2
)
∆cyoA cathode
O
2
cathode
(no bacteria)
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
Fumarate cathode
(no bacteria)
O
2
as terminal
electron acceptor
at cathode
Fe(III)-citrate as
terminal electron
acceptor at
cathode
Fumarate as
terminal electron
acceptor at
cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
Fe(III)-citrate cathode
(no bacteria)
0.00
0.25
0.50
0.75
1.00
1.25
1.50
WT cathode
current density (µA/cm
2
)
∆cyoA cathode
O
2
cathode
(no bacteria)
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
Fumarate cathode
(no bacteria)
O
2
as terminal
electron acceptor
at cathode
Fe(III)-citrate as
terminal electron
acceptor at
cathode
Fumarate as
terminal electron
acceptor at
cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode

Figure 42. Average current densities obtained for Shewanella oneidensis MR-1 WT at the anode with
lactate as an electron donor, and WT (black) or mutants (grey) at the cathode with oxygen, Fe(III)-
citrate, or fumarate as the oxidant. Controls (white) were run using WT as the anode catalyst with
lactate as the electron donor, and only the oxidant at the cathode, with no bacteria. Averages and
standard deviations were calculated from triplicate experiments, using the data associated with the
observed stable current density over twelve hours of operation.
Given that the oxidation conditions of the WT anodes were the same for each
experiment, the unique results for each oxidant must be driven by the chemical reduction
potentials; however, the control experiments for each oxidant (in white) show that the
abiotic reduction of each oxidant was small relative to the bacterial contribution. These
data suggest that the differences in current density are equally a result of the energy

113
metabolism induced by each bacterial culture at the cathode, and the reduction potential
of each chemical oxidant.
Total charge and coulombic efficiency were also calculated for each cathode
catalyst and oxidant, shown in Figure 43. The average charge densities for each oxygen
and FeC experiments for the WT and mutants were all within a standard deviation of
each other, indicating that while each reaction has reached equilibrium within the given
time frame, i.e., all of the reductants and oxidants have been consumed, the total charge
yielded from the system does not reflect any kinetic information about the oxidation and
reduction reactions taking place in the MFC.
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0
20
40
60
80
100
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
N/A N/A
Charge density (C/cm
2
)
Coulombic Efficiency (%)
O
2
cathode
FeC
cathode
Fumarate
cathode
O
2
cathode
FeC
cathode
Fumarate
cathode
a) Charge density (C/cm
2
) b) Coulombic efficiency (%)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0
20
40
60
80
100
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
N/A N/A
Charge density (C/cm
2
)
Coulombic Efficiency (%)
O
2
cathode
FeC
cathode
Fumarate
cathode
O
2
cathode
FeC
cathode
Fumarate
cathode
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0
20
40
60
80
100
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
N/A N/A
Charge density (C/cm
2
)
Coulombic Efficiency (%)
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0
20
40
60
80
100
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
WT cathode
∆cyoA cathode
WT cathode
∆cymA cathode
WT cathode
∆fccA cathode
∆mtrB cathode
∆mtrB cathode
∆mtrB cathode
N/A N/A
Charge density (C/cm
2
)
Coulombic Efficiency (%)
O
2
cathode
FeC
cathode
Fumarate
cathode
O
2
cathode
FeC
cathode
Fumarate
cathode
a) Charge density (C/cm
2
) b) Coulombic efficiency (%)

Figure 43. a) Charge densities, and b) coulombic efficiencies for each evaluated cathode catalyst and
cathode oxidant. Averages and standard deviations were calculated using the total experimental and
theoretical charge yielded from triplicate MFC evaluations.
A comparison of the WT and ∆mtrB current densities for each cathode oxidant
showed that the absence of the mtrB protein precursor had no statistically significant
impact on the current densities associated with cathodic reduction of O
2
or FeC relative to

114
WT (Figure 42). The same trend was apparent in the charge densities, however the WT
cathode with oxygen as the oxidant demonstrated the lowest charge density relative to
∆mtrB using oxygen or FeC as the oxidant. This result may be due to the lack of biomass
at the anode of the WT cathode relative to the other oxygen experiments with the ∆mtrB
and ∆cyoA at the cathode (Figure 44). The absence of attached cells at the anode for this
experiment may indicate that the system was limited by the anode reaction and therefore
produced less charge density relative to the other biocathodes, even though kinetically the
initial response was equivalent as is seen in the current density data.
WT
anode
WT, O
2
cathode
10 µm 10 µm
10 µm
WT
anode
10 µm
∆cyoA, O
2
cathode
10 µm
WT
anode
∆mtrB O
2
cathode
WT
anode
WT, O
2
cathode
10 µm 10 µm
10 µm
WT
anode
10 µm
∆cyoA, O
2
cathode
10 µm
WT
anode
∆mtrB O
2
cathode

Figure 44. Scanning electron microscopy images of graphite electrode fibers exposed to WT as the
anode catalyst and WT, ∆cyoA or ∆mtrB as the cathode catalyst. Oxygen served as the oxidant at the
cathode and lactate as the reductant at the anode.

115
The ∆mtrB fumarate results show a significantly different relationship with
respect to WT fumarate reduction at the cathode, and these current densities were
significantly lower with respect to values achieved using O
2
or FeC.
It has previously been shown that mtrB is required for incorporating mtr c-type
cytochromes into the MR-1 outer membrane and that the absence of these cytochromes
will negatively affect the organisms ability to reduce solid Fe(III)- and Mn(IV)-oxides
(Beliaev and Saffarini 1998; Myers and Myers 2002), as well as to produce current in a
MFC when acting as the anode catalyst (Bretschger, Obraztsova et al. 2007). However,
the results presented in Figure 42 suggest that these outer membrane proteins are not
necessary for accepting electrons from solid electrodes and/or reducing soluble FeC and
O
2
at the cathode of a MFC.
However, the current density data imply that mtrB does appear to be involved
with fumarate reduction at the MFC cathode and may interact with the fumarate
reductase, fccA. The ∆mtrB current densities were very similar to those associated with
∆fccA, indicating that electrons may pass through MtrB to FccA under these conditions;
or that mtrB is involved with activating fccA under these conditions. In the absence of
the fccA protein, very little current density is observed when fumarate is the cathode
oxidant. MtrB and FccA appear to be the key proteins for the reduction of fumarate at the
MFC cathode and when they are not produced, no fumarate reduction can occur. These
results were expected for the fccA mutant, however the ∆mtrB data contradict reported
findings that this mutant is able to anaerobically respire fumarate as well as WT when
using lactate as the electron donor (Beliaev and Saffarini 1998). These current density
results imply that MtrB is involved with the uptake of electrons from the cathode

116
electrode when soluble fumarate is the oxidant, but that other gene products regulate this
process in the presence of FeC or oxygen as the cathode oxidant.
While the current densities suggest that mtrB is involved with fumarate reduction
at the MFC cathode, cymA does not appear to be directly linked to this process. ∆cymA
demonstrated equal abilities to produce current in an MFC when fumarate was the
cathode oxidant, relative to WT. cymA encodes a cytoplasmic membrane-bound,
tetraheme c-type cytochrome that is involved in mediating electron flow from the
cytoplasm to several terminal reductases (including outer membrane c-type cytochromes)
during anaerobic respiration (Myers and Myers 1997; Myers and Myers 2000). Schwalb
et al. proposed that cymA interacts directly with the soluble fumarate reductase protein
and therefore, when cymA is disabled, fumarate reduction is no longer possible (Schwalb,
Chapman et al. 2002). However, the results shown in Figure 42, suggest that a different
pathway for electron transfer to the fumarate reductase exists and is functions when
∆cymA utilizes a solid electron donor as an energy source. Given the current density
results in Figure 42, this alternate pathway may involve mtrB.
Interestingly, both the cymA and mtrB mutants produced very little charge density
relative to the WT when using fumarate as the oxidant. These results imply that while
the kinetics of the oxidation and reduction processes for these mutants may be initially
equivalent to the WT, the total energy gained from the process by these mutants is much
less than WT. SEM images of the cymA and mtrB system anodes and cathodes (Figure
45) show much higher biomass at the anode than at the cathode, indicating that the
current densities observed for these mutants were produced at the onset of system
operation (equivalent kinetics to WT), but that the kinetics of electron uptake at the

117
cathode became the limiting factor causing the WT cells at the anode to begin redirecting
electron flow to producing biomass. The absence of biomass at the cathodes of these
mutant evaluations corroborate this hypothesis and indicate that while some initial energy
was gained, the rate limiting step occurred at the cathode and resulted in low overall
charge densities.
WT
anode
WT,
fumarate
cathode
WT
anode
∆fccA,
fumarate
cathode
10 µm
10 µm
10 µm
10 µm
WT
anode
∆cymA,
fumarate
cathode
10 µm 10 µm
WT
anode
∆mtrB,
fumarate
cathode
10 µm 10 µm
WT
anode
WT,
fumarate
cathode
WT
anode
∆fccA,
fumarate
cathode
10 µm
10 µm
10 µm
10 µm
WT
anode
∆cymA,
fumarate
cathode
10 µm 10 µm
WT
anode
∆mtrB,
fumarate
cathode
10 µm 10 µm

Figure 45. Scanning electron microscopy images of graphite electrode fibers exposed to WT as the
anode catalyst and WT, ∆cymA, ∆mtrB or ∆fccA as the cathode catalyst. Fumarate served as the
oxidant at the cathode and lactate as the reductant at the anode.

118
The coulombic efficiencies, calculated from the organic acid concentrations at the
anode, for the cymA and mtrB mutants (Figure 43b) follow a different pattern than that
observed for charge or current density, with ∆cymA demonstrating much lower
coulombic efficiencies than ∆mtrB. These calculated efficiencies only take into account
the electron transfer process at the anode and do not include the efficiency of electron
uptake at the cathode. A different expression for coulombic efficiency should be used
when considering a fully biotic MFC.
The evaluation of oxygen reduction at the MFC cathode by WT and ∆cyoA also
yielded some unexpected results, given that ∆cyoA was able to produce equivalent
current and charge densities to WT. However, these results may be explained by the
presence of at least three other oxidase-like cytochromes that are present in the MR-1
genome (Heidelberg, Paulsen et al. 2002), these may substitute for CyoA. The ∆cymA,
∆cyoA, ∆fccA and ∆mtrB results suggest that some interesting and unique energy
pathways are induced when MR-1 is forced to use a solid electrode as an electron
acceptor. Future work will explore the regulation of certain c-type cytochromes under
the conditions presented and will hopefully yield a greater understanding of the unique
energy metabolism associated with MR-1. However, these preliminary results indicate
that the energy metabolism(s) employed by MR-1 at the cathode of an MFC are very
different from the pathways induced at the anode; and may also differ greatly from other
anaerobic reduction processes.

119
4.4.3 Conclusions

These results demonstrate for the first time that one bacterial strain can utilize a
MFC anode as an electron acceptor, and simultaneously use the MFC cathode as an
electron donor while generating unique current densities with different cathode oxidants.
The MR-1 mutant results indicate that very distinct mechanisms of respiration are
induced when these bacteria use the cathode electrode as an electron donor. Under these
conditions mtrB, cymA and cyoA do not appear to be critical for the reduction of certain
soluble oxidants.
These results have far reaching implications for the use of MFCs to study
microbial physiology by utilizing electrical signals to map electron movement through
various metabolic pathways.

120
Chapter 5: Conclusions

Microbial fuel cells are a unique technology that could be used for sustainable and
renewable power sources, wastewater treatment systems, bioremediation strategies, BOD
sensors, hydrogen production techniques, and/or used as tools to study microbial ecology
and physiology.
Given the flexibility of MFC systems and their many applications, it is important
to understand the fundamental mechanisms that facilitate MFC processes and apply this
knowledge to optimizing application specific systems. To this end, the work reported
herein contributes to an understanding of the electron transfer capabilities of the genus
Shewanella and lends insight into how MFCs can be optimized for power production, and
used to study microbial ecology and physiology.
The Shewanella genus has been studied for over three decades. It is believed that
Shewanellae are key contributors to global iron and manganese cycles, given their
prevalence in many different ecosystems, and the diverse carbon and energy metabolisms
that these organisms posses (Nealson and Scott 2006). The energy metabolisms of these
organisms also enable them to interact with MFC electrodes and produce electrical
energy. Shewanella oneidensis MR-1 was the first pure culture, metal-reducing bacteria,
to be studied as an MFC anode catalyst (Kim, Kim et al. 1999) and has become one of the
model organisms for subsequent bioelectrochemical evaluations.
5.1 Evaluations of Shewanella oneidensis MR-1 and mutants

This work identifies, for the first time, the MR-1 genes responsible for current
production in MFC systems and also demonstrates that these genes are not implicated in

121
the same way for soluble Fe(III) reduction or the reduction of Fe(III)- or Mn(IV)-oxides.
Forty-six different MR-1 deletion mutants and complementations were evaluated as MFC
anode catalysts, and for their ability to reduce soluble Fe(III)-NTA, solid Fe(III)-oxide
and solid Mn(IV)-oxide.
It was found that the mtr cassette of genes including mtrA, mtrB, mtrC and omcA
are all required for current production when MR-1 is the anode catalyst using lactate as
the electron donor. Additionally, current production is severely inhibited when the
tetraheme cytochrome c encoded by cymA is not present.
It was also found that c-type cytochromes are not the only proteins that play a role
in current production. When the general secretion proteins GspG and GspD are
eliminated, current production is severely inhibited. Similarly, very little current
production is observed when the type 4 prepilin-like proteins leader peptide processing
enzyme, PilD, is not present in the organism.
There were some gene products that participate in electron transfer to several
solid substrates including MFC electrodes, Mn(IV)-oxides and solid Fe(III)-oxides,
however, the overall picture of how specific gene products participate in electron transfer
to these substrates is not yet clear. For example, ∆mtrA, ∆mtrB, ∆omcA/∆mtrC, ∆cymA,
∆gspG and ∆pilD were all limited in their abilities to produce current and reduce HFOM
relative to the WT strain. However, the ∆mtrC/∆omcA mutant was not limited in
Mn(IV)-oxide reduction, implying that the ∆mtrC/∆omcA genes are perhaps not directly
involved with solid phase Mn(IV)-oxide reduction. The trends in the solid substrate data
(Figure 3, Figure 8, and Figure 10) suggest that ∆mtrC/∆omcA is not as limited in
electron transfer abilities to solids as the ∆mtrA and ∆mtrB mutants. This is especially

122
apparent in the Mn(IV)-oxide reduction data. Additionally, the ∆mtrC/∆omcA mutant
shows higher concentrations of Fe(II) (approximately two-times higher) from the
reduction of HFOM after twenty-four (Figure 10) and forty-eight hours, relative to ∆mtrA
and ∆mtrB. Additionally, higher values of current production (approximately five-times
higher) were observed for ∆mtrC/∆omcA relative to ∆mtrA and ∆mtrB (Figure 3) during
the experimental duration. Higher Mn(IV) and Fe(III) reduction rates of an mtrC mutant,
relative to mutants lacking only mtrA or mtrB, have also been reported by Beliaev et al.
(Beliaev, Saffarini et al. 2001).
The data presented herein suggest that the omcA and mtrC gene products are not
the only outer-membrane cytochromes serving as terminal reductase(s) for Mn(IV) or
Fe(III). Furthermore, the solid metal oxide reduction data taken together with the current
production data, (which demonstrated similar limitation patterns relative to the mtr and
omc mutants), indicate that MR-1 features more than one pathway for electron transfer to
solids.
The ∆cymA and ∆gspD mutants show the same patterns as ∆mtrB and ∆mtrA; i.e.,
limited in current production, HFOM reduction and Mn(IV)-oxide reduction. These
results are in agreement with Myers and Myers (Myers and Myers 1997) who suggest
that the cymA cytochrome is a necessary component in the electron transport chain that is
common to fumarate, nitrate, Fe(III) and Mn(IV) oxides.
Interestingly, the two mutants deficient in type II protein secretion, ∆gspD and
∆gspG yielded different results relative to each other. Relative to ∆gspD, the ∆gspG

123
mutant showed limited ability to reduce HFOM, Mn(IV)-oxide, and to produce current in
an MFC.
Another surprising result revealed by this data set was the number of cytochrome
deletion mutants that were able to exceed the WT levels of current production, HFOM,
Fe(III)-NTA and Mn(IV)-oxide reduction. There are many possibilities for why this
might occur including the re-direction of electron flow to alternate terminal reductase(s)
in the occurrence of a road-block to a desired pathway or reductase. It may also be the
case that the cytochromes themselves are negative regulators of the transcription and/or
translation of factors involved with these processes. An alternative possibility is that
these mutants are enhanced in their abilities to form biofilms and therefore yield higher
current production and metal oxide reduction rates due to the presence of more bacteria at
the surface.
The results from the soluble Fe(III)-NTA data show a very different pattern of
reduction by these mutants, relative to the solid-oxide and MFC data. It is expected that
cytochromes and other proteins might play unique roles in the reduction of soluble
compounds relative to how they would be employed to interact with solid compounds.
Soluble compounds like Fe(III)-NTA can permeate the cell membrane and interact with
the organisms electron transport chain within the confines of the cell membrane. In
contrast, the reduction of solid compounds requires either direct physical contact by the
organism or the production of an extra-cellular electron shuttle to mediate electron
transfer. These mutant data corroborate the conclusions of Ruebush et al. (Ruebush,
Brantley et al. 2006) indicating that MR-1 utilizes different mechanisms for solid and

124
soluble Fe(III) reduction. However, further exploration of the abilities of ∆mtrA,
∆mtrC/∆omcA, ∆mtrF, ∆ifcA-1, ∆dmsC, and WT to reduce different forms of soluble
ferric iron led to a very complex picture of pH relationships and perhaps unique affinity
for different soluble iron compounds.
The ability of WT and ∆mtrA to increase the pH of every form of soluble ferric
media tested may play a role in increased Fe(III)-reduction over time. Interestingly,
∆mtrA demonstrated the ability to shift the Fe(III)-NTA(H
+
) media from pH 4.1 to pH
5.4 in four days, while WT was only able to increase the pH from 4.1 to 4.6 during the
same period. The mtrA mutant also achieved higher cell densities than WT at the end of
a four day exposure to Fe(III)-citrate (pH 5.6) and Fe(III)-NTA(Na
+
) (pH 7.2). It should
be noted that cell viability at pH 4.1, the pH of the Fe(III)-NTA media during each
mutant experiment, was very poor. Both WT and ∆mtrA cultures were completely non-
viable after four days of exposure to these acidic conditions, accounting for the fact that
WT showed such low values of Fe(II) production after twenty-four hours of exposure to
the Fe(III)-NTA(H
+
) media.
Unexpectedly, the pH 4.1 media had a very positive effect on several of the
mutants including ∆ifcA-1 and ∆dmsC, which had previously demonstrated very low
Fe(III) reduction abilities with Fe(III)-citrate at a pH of 6.1. These results now introduce
a new question about how various cell mutations in MR1 may affect the organisms pH
tolerance and to what degree does abiotic reduction play a role.
As a whole these data yield very useful information about the genetic machinery
involved with electron transfer by Shewanella oneidensis MR-1. It may now be possible

125
to conduct genetic engineering of MR-1 to enhance current producing abilities in the
MFC. For example, the over-expression of the mtr cassette of genes and cymA, in
addition to the deletion of some other genes including napB, nrfA, dmsC, sorB and cyoA
may lead to the creation of a highly efficient power producing organism that can be used
as an MFC anode catalyst.
5.2 Shewanella strain evaluations

To date, nineteen strains of the genus Shewanella have been genetically
sequenced and it is known that a subset of these strains posses the mtrA, mtrB and mtrC
genes involved with current production in an MFC. These strains include S. putrefaciens
SP200, S. frigidimarina NCIMB400, S. baltica OS155, S. putrefaciens W3-18-1, S.
amazonensis SB2B, S. ANA-3, S. oneidensis MR-4, S. oneidensis MR-7, S. putrefaciens
CN-32, S. loihica PV-4, and the model organism S. oneidensis MR-1.
Several different MFC evaluations were conducted using these strains to
characterize the ability of different Shewanellae to transfer electrons to MFC anodes,
produce anode biofilms, and metabolize lactate as an electron donor under different
system conditions.
An initial screening of S. oneidensis MR-1, S. putrefaciens CN-32, S. putrefaciens
SP200, S. amazonensis SB2B, S. denitrificans OS217, S. loihica PV-4, S. frigidimarina
NCIMB400 and S. baltica OS155 was performed at the Korea Institute of Science and
Technology (KIST). The KIST evaluations suggested that some strains of Shewanella
may perform better than the model organism MR-1, and that the anaerobic carbon
metabolism of these strains at MFC anodes is diverse. However, temperature constraints,

126
rigorous inoculation strategies, and bacterial strain contamination complicated the
interpretation of these results. Therefore a new set of evaluations was performed with
genetically authenticated Shewanella strains, new growth and inoculation strategies,
larger electrodes, different electrolyte buffers, and different ion exchange membranes.
The new MFC evaluations also employed a different test and operation strategy by
incorporating electrochemical impedance spectroscopy to quantify internal system
parameters including polarization resistance, solution resistance and capacitance.
Additionally, polarization resistance scans were performed to generate V-I curves and to
determine maximum power, maximum current and open-circuit cell voltage.
While the KIST evaluations operated each MFC close to the maximum current
available from the system by using a load of 10 ohms, the USC evaluations utilized the
V-I curves to determine the resistance that should be applied to the system so that each
MFC would operate at maximum power.
The KIST evaluations demonstrated that MR-1 was the best current producer
when operating at close to short-circuit conditions in MFCs. However, S. putrefaciens
CN-32 and SP200 and S. amazonensis SB2B were very near MR-1 levels. Additionally,
CN-32 and SP200 demonstrated the highest coulombic efficiencies, and on average,
exceeded MR-1 efficiencies. These data indicated that the S. putrefaciens strains CN-32
and SP200 were much better at transferring electrons to MFC anodes relative to the other
strains.
The KIST data also suggested that the MFC system parameters had an effect on
strain performance. The data produced from one of the three MFCs always demonstrated

127
higher current densities than the other two. This may have been due to platinum loading
at the cathode, flow rates of nitrogen or air to the MFC compartments, or membrane
functionality given that all of these systems had been utilized for previous evaluations.
Bacterial growth conditions and MFC operational temperatures also appeared to
impact strain performance, especially for S. frigidimarina NCIMB400 and S. baltica
OS155. Each Shewanella strain was grown anaerobically and washed with anaerobic
phosphate buffer prior to MFC inoculation. This was done to remove any trace carbon
and energy sources that may have existed in the supernatant before the bacteria were
transferred to the MFC anode. This washing process required pelleting and vortexing
each culture multiple times, which may have disrupted and/or removed any existing
outer-membrane cytochromes and required that the bacteria utilize energy from lactate to
rebuild these functions while operating in the MFC. This hypothesis was supported by
the observation that current production was always higher for each strain during the
second lactate feeding, indicating a recovery time was required in the MFC; and that the
overall coulombic efficiencies were very low, roughly 10%.
Temperature limitations during growth were especially apparent with strains
NCIMB400 and OS155. Both of these strains have optimal growth temperatures between
10°C and 18°C, feature the mtr cassette of genes and can reduce metal-oxides. Although,
these strains demonstrated some growth at 28°C, neither was able to facilitate electron
transfer to MFC anodes with great proficiency.
Based on lessons learned during the KIST evaluations, the USC evaluations
replaced the NCIMB400, OS155, OS217 and SP200 strains with four other Shewanella

128
strains that were all sequenced and found to be pure cultures, grew at 30°C, featured the
mtr cassette of genes and had been implicated in metal oxide reduction. Additionally, the
USC evaluations expanded upon the KIST findings that MFC system parameters and
bacterial growth conditions may affect current production.
The USC evaluations for both buffer systems, operating at P
max
, showed that S.
putrefaciens W3-18-1 was the best power producer and the most efficient strain in terms
of transferring electrons to the MFC anode. Strain performance, in terms of power and
current density, was varied for some of the other strains relative to MR-1 depending on
the system employed for evaluation. For example, all of the strains were enhanced in
power producing abilities when evaluated in the PIPES buffer relative to the phosphate
buffer. However, S. oneidensis MR-7 and S. putrefaciens CN-32 showed a much greater
improvement in performance relative to the other strains, such that the relative trends
between MR-7, CN-32 and the other strains changed. This result may indicate that MR-7
and CN-32 have a greater sensitivity to phosphate than the other strains.
The phosphate buffer appeared to have a severe impact on stable current
production for each Shewanella strain. Current densities, operating at P
max
, did not even
reach 0.5 µA/cm
2
for any of the strains, and very little repeatability was observed
between the triplicate MFC evaluations during operation. These results suggest that the
phosphate buffer impacts system performance more severely with time, given that the
power curves were generated only twelve hours after MFC inoculation, and the current
density data were collected twenty-four hours after inoculation and beyond.

129
Phosphate is known to disrupt iron uptake in bacteria and therefore a strong (100
mM) phosphate buffer may inhibit the upregulation of the decaheme c-type cytochromes
involved with electron transfer to MFC anodes. It is possible that the bacteria were
unable to respire the anode after the initial electron transfer executed during the
polarization of the MFC; and that while the MFCs were allowed to sit at open-circuit
prior to MFC operation, any cells that had attached to the MFC electrode were no longer
able to respire or regulate cytochromes to the outer surface. This hypothesis is
corroborated by the SEM images of the MFC anodes that show an absence of attached
cells, and cell morphologies that indicate stress responses in rod-shaped bacteria.
The current densities for the PIPES buffer evaluations were significantly higher
for each strain than the phosphate buffer values and also followed similar relative trends
to the power densities collected for each strain. The SEM images collected for the PIPES
evaluations of each strain showed that more bacteria were attached to the MFC anodes,
relative to the phosphate images. Although thick biofilms were not developed, several
colonies and monolayers of rod-shaped bacteria were observed along the anode surfaces
exposed to S. oneidensis MR-1, S. putrefaciens W3-18-1, S. amazonensis SB2B, S.
putrefaciens CN-32 and S. loihica PV-4. S. ANA-3 and S. oneidensis MR-7 SEM images
indicate that the anode fibers exposed to these cultures were sparsely populated with
clusters of biomass interspersed with rod-shaped cells. S. oneidensis MR-4 anodes
appeared to have the fewest number of attached cells with individual bacteria widely
scattered along the length of the carbon fibers.

130
The coulombic efficiencies observed for each strain evaluated in the PIPES buffer
indicate that the carbon metabolism employed by several strains had a significant impact
on the over all efficiency of MFC performance. For example, MR-7 and CN-32
demonstrated high coulombic efficiencies relative to MR-1 and were the strains that most
effectively oxidized lactate all the way to CO
2
, i.e., very little acetate or formate remained
in the system at the termination of the experiment. This trend was apparent over all
lactate feedings, but is most easily visualized during the second lactate feeding (Figure
46).
0
5
10
15
20
25
30
35
40
45
50
Coulombic Efficiency (%)
Coulombic efficiency for only the second lactate feeding
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4
0
5
10
15
20
25
30
35
40
45
50
Coulombic Efficiency (%)
0
5
10
15
20
25
30
35
40
45
50
Coulombic Efficiency (%)
Coulombic efficiency for only the second lactate feeding
MR-1
W3-18-1
SB2B
ANA-3
MR-4
MR-7
CN-32
PV-4

Figure 46. Coulombic efficiency for each strain during the second lactate feeding only.
S. putrefaciens W3-18-1 and S. ANA-3 also demonstrated a high charge density
and coulombic efficiency relative to the other strains indicating that these strains were
very adept at transferring charge to the electrode surface, even though they were less
efficient than MR-7 at utilizing acetate.

131
The SEM images for the strains W3-18-1, CN-32, ANA-3 and MR-7 imply an
interesting relationship between how these cultures transferred electrons and formed
biofilms at the anode. S. putrefaciens W3-18-1 was the only strain of the four to
demonstrate well developed colonies and moderate surface coverage of the anode fibers.
Anodes exposed to S. putrefaciens CN-32 had a much patchier network of colonies that
were unevenly distributed along the length of the fibers; and S. ANA-3 and S. oneidensis
MR-7 both lacked colonies along the fibers, but instead featured globular biomass
distributed as monolayers along the length of the fibers with rod-shaped cells individually
interspersed.
The MR-7, CN-32 and ANA-3 results imply that relatively high coulombic
efficiencies can be achieved without a significant amount of biomass attached to the
anode surface, indicating that the coulombic efficiency of each individual cell is high.
The W3-18-1 results imply a similar phenomena, given that the highest coulombic
efficiency, current density and power density was demonstrated by this strain, which also
had the best biofilm coverage at the MFC anode.
The relatively lower coulombic efficiencies, moderate power and current densities,
and higher accumulations of biomass demonstrated by MR-1 and SB2B may be
indicative of the preference of these strains to utilize energy toward building biomass
rather than moving electrons to the surface of the anode. Similarly, the relatively low
coulombic efficiencies, low power and current densities, and limited biomass at the
anodes of MR-4 and PV-4 may be indicative that these strains either suffered from

132
systems limitations, i.e., the external load across the circuit was too high, or are not well
suited to act as anode catalysts in MFC systems.
An additional parameter to consider when evaluating relative system performance
is the viability of each culture and at what point along the respective growth cycle each
strain was harvested. All strains were grown aerobically for forty-eight hours prior to
MFC inoculation and the appropriate cell densities were inoculated into the MFC anode,
such that approximately 2 x 10
8
cells/mL were present in the anode compartment for each
evaluation. However, growth curves collected for each strain indicate that forty-eight
hours of growth did not correspond to the same growth phase for each strain (Figure 47).
0
10
20
30
40
50
60
0 10 2030405060
MR1
W3-18-1
SB2B
ANA
MR4
MR7
CN32
PV4
Time (hrs)
Cell counts x10
8
(cells/mL) at 30°C
0
10
20
30
40
50
60
0 10 2030405060
MR1
W3-18-1
SB2B
ANA
MR4
MR7
CN32
PV4
Time (hrs)
Cell counts x10
8
(cells/mL) at 30°C

Figure 47. Growth curves over a forty-eight hour period for S. oneidensis MR-1, S. putrefaciens W3-
18-1, S. amazonensis SB2B, S. ANA-3, S. oneidensis MR-4, S. oneidensis MR-7, S. putrefaciens CN-32
and S. loihica PV-4. All strains were grown aerobically at 30°C at an agitation rate of 150 rpm using
PIPES buffered minimal media with 18 mM lactate as the electron donor.

133
The most efficient strains including W3-18-1, MR-7 and CN-32 were all at late
log phase or early stationary phase when harvested for MFC inoculation. ANA-3
appeared to be in late stationary phase after forty-eight hours of growth, but interestingly
demonstrated two log phases, one during each twenty-four hour period. SB2B and MR-1
also appeared to be in late stationary phase after forty-eight hours of growth, but also
demonstrated the lowest culture densities relative to the other seven strains. PV-4 was
also apparently in late stationary phase after forty-eight hours growth, but had higher cell
densities than SB2B and MR-1. MR-4 appeared to be in the death phase after forty-eight
hours of growth, which may explain why MFC performance was so low relative to the
other strains.
There is an apparent relationship between how strains performed in an MFC and
at point in their growth cycle they were transferred into the MFC. Those strains that were
harvested in late log phase were the most efficient and demonstrated the highest charge
densities. The strains harvested in late stationary phase demonstrated similar charge
densities to each other, and on average, lower charge densities and coulombic efficiencies
relative to those strains harvested in late log phase. Finally MR-4, the one strain that was
harvested while in its death phase, demonstrated the lowest charge density and average
coulombic efficiency.
These results yield valuable information about how to evaluate and utilize
different strains of Shewanella in MFC environments. Additionally, several strains have
been identified to be either more efficient at oxidizing carbon sources, or produce more
power, than the model organism S. oneidensis MR-1.

134
5.3 MFCs as tools for studying microbial physiology

MR-1 cultures grown at low DOT (5%) exhibit higher and more consistent
current densities in batch operated MFCs than MR-1 cultures grown at a higher DOT
(50%). These results indicate that the treatment of microorganisms prior to MFC
inoculation will impact total performance. Additionally, these results demonstrate that
the electrical signals produced by the microbial catalysts can yield valuable information
about the physiology and adaptability of these organisms.
Several different carbon sources were tested as fuels in MR-1 MFCs and it was
found that acetate, formate and succinate can all be utilized by MR-1 under anaerobic
anode conditions after lactate has been consumed from the system. However, pyruvate
does not appear to be a good reductant for MFC use when MR-1 is the anode catalyst.
The electrical signal yielded from each carbon source was unique and demonstrates that
the I-t profiles may provide unique information about the kinetics and thermodynamics of
the redox reactions taking place within the MFC system. Lactate produced the highest
current and charge densities of the carbon sources evaluated indicating that it is the
preferred energy source. Acetate and succinate yielded lower current and charge
densities relative to lactate, but higher charge densities relative to formate. Formate
yielded a current density equivalent to lactate, however a much lower charge density.
Further study should investigate the interpretation of these signals, given that it is
possible to infer useful physiological information from the mesas and valleys of current
produced by the biocatalyst in MFC systems.
High cell planktonic cell densities (~10
10
cells/mL) and ex-situ grown anode
biofilms demonstrated the highest current densities (operating at I
max
) relative to those

135
systems inoculated with lower cell densities of planktonic bacteria. It is believed that a
well established biofilm with an interconnected network of extra-cellular appendages
improves the current producing capacity of an MFC, however the anodes injected with a
high cell density of planktonic bacteria were not imaged, and it is not known if the
nanowire network was established in the MFC under these conditions. Future work
should be devoted to better understanding this issue and capturing information about the
charge that is transferred per bacterial cell. Understanding the relationships between
biomass and nanowire density will hopefully yield a more complete picture of energy
transfer through complex networks of microorganisms.

A study of MR-1 wild type and mutants as cathode catalysts with different
oxidants demonstrated that MR-1 can utilize the solid electrode as an electron donor
while reducing soluble terminal electron acceptors. The current densities produced from
each evaluation appear to correspond with the magnitude of energy that can be gained
from the oxidant at the cathode. The mutant data suggest that different energy pathways
are induced when MR-1 uses the cathode as an electron donor, than when using the anode
as an electron acceptor or when exposed to soluble electron acceptors outside of the MFC
environment.
It appears as though cymA and mtrB reverse roles relative to the incorporation of
the fumarate reductase, fccA. Under anaerobic bottle experiments the absence of cymA
will inhibit fumarate reduction (Myers and Myers 1997); however the absence of mtrB
will not (Myers and Myers 2002). In the MFC cathode, the absence of cymA does not
affect the kinetics of fumarate reduction relative to WT; however the absence of mtrB

136
does. These data represent the first investigation utilizing a pure culture as the anode and
cathode catalyst in an MFC system and imply that very unique energy metabolisms are
employed by Shewanella to donate, and accept, electrons from solid substrates.
5.4 Future Work

Although progress has been made relative to understanding the mechanisms
associated with electron transfer in MFCs, there are still many unanswered questions and
continuing need for optimization. From the biological prospective there are several
outstanding questions related to what gene products are involved with electron transfer to
and from solid substrates. To answer this question gene expression and proteomic data
should be collected under different MFC conditions, and using different MFC cell
populations including the attached biofilm and the planktonic cells. These data could
hopefully yield more information about differential gene expression and how proteins are
regulated under different MFC conditions, i.e. bacteria attached to the anode vs
populations attached to the cathode. Additionally, it would be very interesting to learn
how gene expression is different between a biofilm population and a planktonic
population. This type of data might also identify other gene products involved in electron
transfer and highlight target genes that might be overexpressed and/or removed from an
organism to genetically engineer a better power producer.
In addition to collecting data about differential gene expression and protein
regulation, more of the MR-1 single deletion mutations should be evaluated at the
cathode to yield a greater understanding about which cytochromes may be involve in
accepting electrons from solid surfaces, and how these processes may change the other

137
cytochrome interactions with soluble electron acceptors. Specifically, the other putative
cytochrome c oxidase mutants should be evaluated and the cymA mutation should be
screened with other oxidants at the cathode.
Given the observed relationship between biofilm coverage at the anode surface,
coulombic efficiency and power production, it is very important to devise methods for
determining the number of electrons yielded per individual bacterial cell. Once this
relationship is well characterized a more focused effort can be spent on designing MFC
systems that fully exploit the available charge produced from a bacterial population, i.e.
better current collectors and high surface area electrodes that are specifically designed to
maximize electron transfer from bacteria.
The studies reported here utilized MFCs assemble with graphite felt electrodes
and Nafion proton exchange membranes. Although graphite felt is an inexpensive
material, the current distribution of these electrodes appeared to be less than optimal and
the thickness of the felt may also inhibit proton conduction through the system.
Furthermore, Nafion is not a cost effect material when considering scaling MFCs to a
wastewater treatment application. For these reasons, it is critical that more research be
conducted with different electrode and membrane materials, in addition to exploring
different MFC designs that may be more suitable for real applications.
The exploration of different electrode materials and membranes should include
the use of electrochemical techniques such as EIS, potentiodynamic polarization and CV;
however caution should be employed in the interpretation of these types of data,
especially EIS and CV. Little is known about how these electrochemical measurements
may affect the biological reactions and more precise and controlled experiments should

138
be conducted to explore these relationships. Periodic anolyte analysis during slow-scan
CV should be performed to clarify the reactions that produce different redox peaks that
appear when MFCs are operated under varying conditions.
Additionally, X-ray photoelectron spectroscopy and other surface analysis
techniques should be utilized to investigate the presence of adsorbed species on electrode
surfaces, and clarify the inductive behavior seen in EIS data when using large-diameter
graphite felt electrodes.
Understanding how bacteria transfer electrons to solid surfaces, how to optimize
the biological reaction, how to design MFC systems that best exploit these reactions, and
how to interpret the electrochemical data yielded from biological catalysts will ultimately
lead to the optimization of MFC systems, and the opportunity to employ these systems in
many different applications.

139
Bibliography

Aelterman, P., K. Rabaey, et al. (2006). "Continuous Electricity Generation at High
Voltages and Currents Using Stacked Microbial Fuel Cells." Environmental
Science and Technology 40(10): 3388-3394.

Akiba, T., H. P. Bennetto, et al. (1987). "Electricity Production from Alkalophilic
Organisms." Biotechnology Letters 9(9): 611-616.

Allen, R. M. and P. H. Bennetto (1992). Microbial Fuel-Cells: Electricity Production
from Carbohydrates. 14th Symposium on Biotechnology for Fuels and Chemicals,
Gatlinburg, Tennessee, Applied Biochemistry and Biotechnology.

Allen, R. M. and P. H. Bennetto (1993). "Microbial Fuel-Cells: Electricity Production
from Carbohydrates." Applied Biochemistry and Biotechnology 39/40: 27-40.

Arnold, R. G., M. R. Hoffmann, et al. (1990). "Regulation of Dissimilatory Fe(III)
Reduction Activity in Shewanella putrefaciens." Applied and Environmental
Microbiology 56(9): 2811-2817.

Arnold, R. G., T. M. Olson, et al. (1986). "Kinetics and mechanism of dissimilative
Fe(III) reduction by Pseudomonas sp. 200." Biotechnology and Bioengineering
28(11): 1657-1671.

Back, J. H., M. S. Kim, et al. (2004). "Construction of Bacterial Artificial Chromosome
Library from Electrochemical Microorganisms." FEMS Microbiology Letters
238(1): 65.

Bai, L. and B. E. Conway (1991). "AC Impedance of Faradaic Reactions Involving
Electrosorbed Intermediates: Examination of Conditions Leading to
Pseudoinductive Behavior Represented in Three-Dimensional Impedance
Spectroscopy Diagrams." Journal of The Electrochemical Society 138(10): 2897-
2907.

Bai, L. and B. E. Conway (1993). "Three-dimensional impedance spectroscopy diagrams
for processes involving electrosorbed intermediates, introducing the third
electrode-potential variable--examination of conditions leading to pseudo-
inductive behavior." Electrochimica Acta 38(14): 1803-1815.

Barsoukov, E. and J. R. Macdonald (2005). Impedance Spectroscopy Theory, Experiment,
and Applications. Hoboken, New Jersey, Wiley-Interscience.

140
Beliaev, A. S., D. M. Klingeman, et al. (2005). "Global Transcriptome Analysis of
Shewanella oneidensis MR-1 Exposed to Different Terminal Electron Acceptors."
Journal of Bacteriology 187(20): 7138-7145.

Beliaev, A. S. and D. A. Saffarini (1998). "Shewanella putrefaciens mtrB Encodes an
Outer Membrane Protein Required for Fe(III) and Mn(IV) Reduction." Journal of
Bacteriology 180(23): 6292-6297.

Beliaev, A. S., D. A. Saffarini, et al. (2001). "MtrC, an Outer Membrane Decahaem c
Cytochrome Required for Metal Reduction in Shewanella putrefaciens MR-1."
Molecular Microbiology 39(3): 722-730.

Bencheikh-Latmani, R., S. M. Williams, et al. (2005). "Global Transcriptional Profiling
of Shewanella oneidensis MR-1 during Cr(VI) and U(VI) Reduction." Applied
and Environmental Microbiology 71(11): 7453-7460.

Bennetto, H. P. (1987). "Microbes Come to Power." New Scientist 114: 36-40.

Bennetto, H. P. (1990). "Bugpower"-The Generation of Microbial Electricity. Frontiers
of Science. A. Scott, Blackwell, Oxford: 60-82.

Bennetto, H. P. (1990). "Electricity Generation by Microorganisms." Biotechnology
Education 1(4): 163-168.

Bennetto, H. P., G. M. Delaney, et al. (1985). "The Sucrose Fuel Cell: Efficient Biomass
Conversion Using a Microbial Catalyst." Biotechnology Letters 7(10): 699-704.

Bennetto, P. H., J. L. Stirling, et al. (1983). "Anodic Reactions in Microbial Fuel Cells."
Biotechnology and Bioengineering 25(2): 559-568.

Bergel, A., D. Féron, et al. (2005). "Catalysis of Oxygen Reduction in PEM Fuel Cell by
Seawater Biofilm." Electrochemistry Communications 7(9): 900-904.

Berk, R. S. and J. H. Canfield (1964). "Bioelectrochemical Energy Conversion." Applied
Microbiology 12(1): 10-12.

Biffinger, J. C., J. N. Byrd, et al. (2008). "Oxygen Exposure Promotes Fuel Diversity for
Shewanella oneidensis Microbial Fuel Cells." Biosensors and Bioelectronics
23(6): 820-826.

Biffinger, J. C., J. Pietron, et al. (2007). "A Biofilm Enhanced Miniature Microbial Fuel
Cell using Shewanella oneidensis DSP10 and Oxygen Reduction Cathodes."
Biosensors and Bioelectronics 22(8): 1672-1679.

141
Biffinger, J. C., R. Ray, et al. (2007). "Diversifying Biological Fuel Cell Designs by Use
of Nanoporous Filters." Environmental Science and Technology 41(4): 1444-1449.

Bond, D. R., D. E. Holmes, et al. (2002). "Electrode-Reducing Microorganisms That
Harvest Energy from Marine Sediments." Science 295(5554): 483-485.

Bond, D. R. and D. R. Lovley (2003). "Electricity Production by Geobacter
sulfurreducens Attached to Electrodes." Applied and Environmental
Microbiology 69(3): 1548-1555.

Bouhenni, R., A. Gehrke, et al. (2005). "Identification of Genes Involved in Cytochrome
c Biogenesis in Shewanella oneidensis, Using a Modified Mariner Transposon."
Applied and Environmental Microbiology 71(8): 4935-4937.

Bowman, J. P., S. A. McCammon, et al. (1997). "Shewanella gelidimarina sp. nov. and
Shewanella frigidimarina sp. nov., Novel Antarctic Species with the Ability to
Produce Eicosapentaenoic Acid (20:5 omega 3) and Grow Anaerobically by
Dissimilatory Fe(III) Reduction." International Journal of Systematic and
Evolutionary Microbiology 47(4): 1040-1047.

Brake, J., W. Momyer, et al. (1963). Biochemical Fuel Cells, Part 2. 17th Annual Power
Sources Conference.

Bretschger, O., A. Obraztsova, et al. (2007). "An Exploration of Current Production and
Metal Oxide Reduction by Shewanella oneidensis MR-1 Wild Type and
Mutants." Applied and Environmental Microbiology 70(21): 7003-7012.

Brettar, I., R. Christen, et al. (2002). "Shewanella denitrificans sp. nov., A Vigorously
Denitrifying Bacterium Isolated from the Oxic-Anoxic Interface of the Gotland
Deep in the Central Baltic Sea." International Journal of Systematic and
Evolutionary Microbiology 52(6): 2211-2217.

Brettar, I., E. R. B. Moore, et al. (2001). "Phylogeny and Abundance of Novel
Dentrifying Bacteria Isolated from the Water Column of the Central Baltic Sea."
Microbial Ecology 42(3): 295-305.

Bullen, R. A., T. C. Arnot, et al. (2006). "Biofuel Cells and their Development."
Biosensors and Bioelectronics 21(11): 2015.

Burdige, D. J. and K. H. Nealson (1985). Microbial Manganese Reduction by Enrichment
Cultures from Coastal Marine Sediments. 50: 491-497.

Chang, I.-S., H. Moon, et al. (2006). "Electrochemically active bacteria (EAB) and
mediator-less microbial fuel cells." J. Microbiol. Biotechnol. 16: 163-177.

142
Chang, I. S., H. Moon, et al. (2006). "Electrochemically Active Bacteria (EAB) and
Mediator-Less Microbial Fuel Cells." Journal of Microbiology and Biotechnology
16(2): 163-177.

Chaudhuri, S. K. and D. R. Lovley (2003). "Electricity Generation by Direct Oxidation of
Glucose in Mediatorless Microbial Fuel Cells." Nature Biotechnology 21(10):
1229.

Chen, G.-W., S.-J. Choi, et al. (2008). "Application of Biocathode in Microbial Fuel
Cells: Cell Performance and Microbial Community." Applied Microbiology and
Biotechnology.

Cheng, S., H. Liu, et al. (2006). "Increased Performance of Single-Chamber Microbial
Fuel Cells using an Improved Cathode Structure." Electrochemistry
Communications 8(3): 489.

Choi, Y., E. Jung, et al. (2004). "Construction of Microbial Fuel Cells Using
Thermophilic Microorganisms, Bacillus licheniformis and Bacillus
thermoglucosidasius." Bulletin of the Korean Chemical Society 25(6): 813-818.

Ciureanu, M. and H. Wang (1999). "Electrochemical Impedance Study of Electrode-
Membrane Assemblies in PEM Fuel Cells: I. Electro-oxidation of H
2
and H
2
/CO
Mixtures on Pt-Based Gas-Diffusion Electrodes." Journal of the Electrochemical
Society 146(11): 4031-4040.

Clauwaert, P., K. Rabaey, et al. (2007). "Biological Denitrification in Microbial Fuel
Cells." Environmental Science and Technology 41(9): 3354-3360.

Clauwaert, P., D. van der Ha, et al. (2007). "Open Air Biocathode Enables Effective
Electricity Generation with Microbial Fuel Cells." Environmental Science and
Technology 41(21): 7564-7569.

Cohen, B. (1931). The Bacterial Culture as an Electrical Half-Cell. Thirty-second Annual
Meeting of the Society of American Bacteriologists.

Cohn, E. M. (1963). "Perspectives on Biochemical Electricity." Developments in
Industrial Microbiology 4: 53-58.

Cornell, R. M. and U. Schwertmann (1996). Ferrihydrite. The Iron Oxides: Structure,
Properties, Reactions, Occurences and Uses. B. Bock. Weinheim Germany, VCH:
491.

Darowicki, K. (1997). "The fixed state in impedance measurements of a two-step
electrode reaction proceeding with accompanying adsorption of an intermediate
product." Electrochimica Acta 42(7): 1073-1079.

143

Davis, J. B. (1963). "Generation of Electricity by Microbial Action." Advances in
Applied Microbiology 5: 51-64.

Davis, J. B. and H. F. Yarbrough, Jr. (1962). "Preliminary Experiments on a Microbial
Fuel Cell." Science 137(3530): 615-616.

de Mele, M. F. L., H. A. Videla, et al. (1982). "Comparative Study of the Electrochemical
Behaviour of Glucose and other Compounds of Biological Interest."
Bioelectrochemistry and Bioenergetics 9(4): 469-487.

DiChristina, T. J., C. M. Moore, et al. (2002). "Dissimilatory Fe(III) and Mn(IV)
Reduction by Shewanella putrefaciens Requires ferE, a Homolog of the pulE
(gspE) Type II Protein Secretion Gene." Journal of Bacteriology 184(1): 142-151.

Disalvo, E. A. (1980). "Coupling Between Electrochemical and Biochemical Reactions."
Bioelectrochemistry and Bioenergetics 7(4): 807-815.

Disalvo, E. A. (1980). "Transference of Oxygen Through Biological Interphases and its
Relation with a Bioelectrochemical Reaction." Bioelectrochemistry and
Bioenergetics 7(4): 787-792.

Disalvo, E. A. and H. A. Videla (1979). "Relation Between Biological Parameters and the
Bioelectrochemical Fuel-Cell Response." Bioelectrochemistry and Bioenergetics
6(2): 185-195.

Disalvo, E. A. and H. A. Videla (1981). "Role of biological agents in the oxidation of
glucose on black platinum electrodes." Biotechnology and Bioengineering 23(5):
1159-1164.

Dollhopf, M. E., K. H. Nealson, et al. (2000). "Kinetics of Fe(III) and Mn(IV) Reduction
by the Black Sea strain of Shewanella putrefaciens using In Situ Solid State
Voltammetric Au/Hg Electrodes." Marine Chemistry 70(1-3): 171-180.

Elias, D. A., M. E. Monroe, et al. (2006). "Confirmation of the Expression of a Large set
of Conserved Hypothetical Proteins in Shewanella oneidensis MR-1." Journal of
Microbiological Methods 66(2): 223-233.

Fischer, D. J., A. Landes, et al. (1965). "Electrochemical Observations in Microbiological
Processes. Growth of Thiobacillus Thiooxidans. II." Biotechnology and
Bioengineering 7(4): 491-506.

144
Fredrickson, J. K., A. S. Beliaev, et al. (2004). Global and Physiological Responses to
Substrate Shifts in Continuous and Controlled Batch Cultures of Shewanella
oneidensis MR-1. GTL: Contractor-Grantee Workshop II, Washington, D.C., U.S.
Department of Energy.

Fredrickson, J. K., M. F. Romine, et al. (2008). "Towards Environmental Systems
Biology of Shewanella." Nature Reviews Microbiology Submitted.

Fredrickson, J. K., J. M. Zachara, et al. (1998). "Biogenic Iron Mineralization
Accompanying the Dissimilatory Reduction of Hydrous Ferric Oxide by a
Groundwater Bacterium." Geochimica et Cosmochimica Acta 62(19-20): 3239-
3257.

Freguia, S., K. Rabaey, et al. (2007). "Electron and Carbon Balances in Microbial Fuel
Cells Reveal Temporary Bacterial Storage Behavior During Electricity
Generation." Environmental Science and Technology 41(8): 2915-2921.

Freguia, S., K. Rabaey, et al. (2008). "Sequential Anode-Cathode Configuration
Improves Cathodic Oxygen Reduction and Effluent Quality of Microbial Fuel
Cells." Water Research 42(6-7): 1387-1396.

Gao, H., A. Obraztova, et al. (2006). "Shewanella loihica sp. nov., Isolated from Iron-
Rich Microbial Mats in the Pacific Ocean." International Journal of Systematic
and Evolutionary Microbiology 56(8): 1911-1916.

Gil, G.-C., I.-S. Chang, et al. (2003). "Operational Parameters Affecting the
Performannce of a Mediator-less Microbial Fuel Cell." Biosensors and
Bioelectronics 18(4): 327.

Goel, R. K. and J. R. V. Flora (2005). "Sequential Nitrification and Denitrification in a
Divided Cell Attached Growth Bioelectrochemical Reactor." Environmental
Engineering Science 22(4): 440-449.

Gorby, Y. A., S. Yanina, et al. (2006). "Electrically Conductive Bacterial Nanowires
Produced by Shewanella oneidensis Strain MR-1 and Other Microorganisms."
Proceedings of the National Acadamy of Sciences, United States 103(30): 11358-
11363.

Gregory, K. B., D. R. Bond, et al. (2004). "Graphite Electrodes as Electron Donors for
Anaerobic Respiration." Environmental Microbiology 6(6): 596-604.

Gregory, K. B. and D. R. Lovley (2005). "Remediation and Recovery of Uranium from
Contaminated Subsurface Environments with Electrodes." Environmental Science
and Technology 39(22): 8943-8947.

145
Guo, Y., S. Srinivasan, et al. (2007). "Investigating the Effect of Chromatographic
Conditions on Retention of Organic Acids in Hydrophilic Interaction
Chromatography Using a Design of Experiment." Chromatographia 66(3): 223-
229.

Hamann, C. H., A. Hamnett, et al. (1998). Electrochemistry, Wiley-VCH.

Harrington, D. A. and B. E. Conway (1987). "ac Impedance of Faradaic reactions
involving electrosorbed intermediates--I. Kinetic theory." Electrochimica Acta
32(12): 1703-1712.

Hartshorne, R., B. Jepson, et al. (2007). "Characterization of Shewanella oneidensis
MtrC: A Cell-Surface Decaheme Cytochrome Involved in Respiratory Electron
Transport to Extracellular Electron Acceptors." Journal of Biological Inorganic
Chemistry 12(7): 1083-1094.

Hau, H. H. and J. A. Gralnick (2007). "Ecology and Biotechnology of the Genus
Shewanella." Annual Review of MIcrobiology 61(1): 237-258.

He, Z. and Largus T. Angenent (2006). "Application of Bacterial Biocathodes in
Microbial Fuel Cells." Electroanalysis 18(19-20): 2009-2015.

He, Z., N. Wagner, et al. (2006). "An Upflow Microbial Fuel Cell with an Interior
Cathode: Assessment of the Internal Resistance by Impedance Spectroscopy."
Environmental Science and Technology 40(17): 5212-5217.

Heidelberg, J. F., I. T. Paulsen, et al. (2002). "Genome Sequence of the Dissimilatory
Metal Ion-Reducing Bacterium Shewanella oneidensis." Nat Biotech 20(11): 1118.

Hernandez, M. E. and D. K. Newman (2001). "Extracellular electron transfer." Cellular
and Molecular Life Sciences (CMLS) 58(11): 1562.

Holmes, D. E., D. R. Bond, et al. (2004). "Electron Transfer by Desulfobulbus
propionicus to Fe(III) and Graphite Electrodes." Applied and Environmental
Microbiology 70(2): 1234-1237.

Holmes, D. E., S. K. Chaudhuri, et al. (2006). "Microarray and Genetic Analysis of
Electron Transfer to Electrodes in Geobacter sulfurreducens." Environmental
Microbiology 8(10): 1805-1815.

Junter, G. A., J. F. Lemeland, et al. (1980). "Electrochemical Detection and Counting of
Escherichia coli in the Presence of a Reducible Coenzyme, Lipoic Acid." Applied
and Environmental Microbiology 39(2): 307-316.

146
Karube, I., T. Matsunaga, et al. (1976). "Continuous Hydrogen Production by
Immobilized Whole Cells of Clostridium butyricum." Biochimica et Biophysica
Acta 444: 338-343.

Karube, I., T. Matsunaga, et al. (1977). "Biochemical Fuel Cell Utilizing Immobilized
Cells of Clostridium butyricum." Biotechnology and Bioengineering 19(11):
1727-1733.

Karube, I., H. Matsuoka, et al. (1984). "Large-Scale Bacterial Fuel Cell Using
Immobilized Photosynthetic Bacteria." Annals of the New York Academy of
Sciences 434(1): 427-436.

Karube, I., S. Suzuki, et al. (1981). "Biochemical Energy Conversion by Immobilized
Whole Cells." Annals of the New York Academy of Sciences 369(1): 91-98.

Kieft, T. L., J. K. Fredrickson, et al. (1999). Dissimilatory Reduction of Fe(III) and Other
Electron Acceptors by a Thermus Isolate. 65: 1214-1221.

Kim, B. H., H. J. Kim, et al. (1999). "Direct Electrode Reaction of Fe(III)-Reducing
Bacterium, Shewanella putrefaciens." Journal of Microbiology and Biotechnology
9: 127-131.

Kim, B. H., H. S. Park, et al. (2004). "Enrichment of Microbial Community Generating
Electricity using a Fuel-Cell-Type Electrochemical Cell." Applied Microbiology
and Biotechnology 63(6): 672.

Kim, B. H., Park, H.S., Kim, H.J., Kim, G.T., Chang, I.S., Lee, J., Phung, N.T. (2004).
"Enrichment of microbial community generating electricity using a fuel-cell-type
electrochemical cell." Applied Microbiology and Biotechnology 63: 672-681.

Kim, G. T., M. S. Hyun, et al. (2005). "Dissimilatory Fe(III) reduction by an
electrochemically active lactic acid bacterium phylogenetically related to
Enterococcus gallinarum isolated from submerged soil." Journal of Applied
Microbiology 99(4): 978-987.

Kim, H. J., Moon, S.H., Chang, I.S., Kim, B.H. (1999). "A Microbial Fuel Cell Type
Lactate Biosensor Using a Metal-Reducing Bacterium, Shewanella putrefaciens."
Journal of Microbiology and Biotechnology 9(3): 365-367.

Kim, H. J., H. S. Park, et al. (2002). "A Mediator-less Microbial Fuel Cell using a Metal
Reducing Bacterium, Shewanella putrefaciens." Enzyme and Microbial
Technology 30(2): 145.

147
Kim, J. R., S. H. Jung, et al. (2007). "Electricity Generation and Microbial Community
Analysis of Alcohol Powered Microbial Fuel Cells." Bioresource Technology
98(13): 2568-2577.

Kolker, E., A. F. Picone, et al. (2005). "Global Profiling of Shewanella oneidensis MR-1:
Expression of Hypothetical Genes and Improved Functional Annotations."
Proceedings of the National Acadamy of Sciences, United States 102(6): 2099-
2104.

Kolker, E., A. F. Picone, et al. (2005). "Global profiling of Shewanella oneidensis MR-1:
Expression of hypothetical genes and improved functional annotations." PNAS
102(6): 2099-2104.

Konikoff, J. J. (1963). "Electrical Energy from Biological Systems." Aerospace medicine
34: 1129-1133.

Krumbein, W. E. and H. J. Altmann (1973). "A New Method for the Detection and
Enumeration of Manganese Oxidizing and Reducing Microorganisms."
Helgolander wiss. Meeresunters 25: 347-356.

Lane, N. (2006). "Microbiology: Batteries Not Included, What Can't Bacteria Do?"
Nature 441(7091): 274.

Lee, J., N. T. Phung, et al. (2003). "Use of Acetate for Enrichment of Electrochemically
Active Microorganisms and their 16S rDNA Analyses." FEMS Microbiology
Letters 223(2): 185.

Lewis, K. (1966). "Symposium on bioelectrochemistry of microorganisms. IV.
Biochemical fuel cells." Bacteriological Reviews 30(1): 101-113.

Lies, D. P., M. E. Hernandez, et al. (2005). "Shewanella oneidensis MR-1 Uses
Overlapping Pathways for Iron Reduction at a Distance and by Direct Contact
under Conditions Relevant for Biofilms." Applied and Environmental
Microbiology 71(8): 4414-4426.

Liu, H., S. Cheng, et al. (2005). "Power Generation in Fed-Batch Microbial Fuel Cells as
a Function of Ionic Strength, Temperature, and Reactor Configuration."
Environmental Science and Technology 39: 5488-5493.

Liu, H., S. Cheng, et al. (2005). "Production of Electricity from Acetate or Butyrate
Using a Single-Chamber Microbial Fuel Cell." Environmental Science and
Technology 39(2): 658-662.

148
Liu, H., S. Grot, et al. (2005). "Electrochemically Assisted Microbial Production of
Hydrogen from Acetate." Environmental Science and Technology 39(11): 4317-
4320.

Liu, H. and B. E. Logan (2004). "Electricity Generation Using an Air-Cathode Single
Chamber Microbial Fuel Cell in the Presence and Absence of a Proton Exchange
Membrane." Environmental Science and Technology 38(14): 4040-4046.

Liu, H., R. Ramnarayanan, et al. (2004). "Production of Electricity during Wastewater
Treatment Using a Single Chamber Microbial Fuel Cell." Environmental Science
and Technology 38(7): 2281-2285.

Logan, B. E., B. Hamelers, et al. (2006). "Microbial Fuel Cells: Methodology and
Technology." Environmental Science and Technology 40(17): 5181-5192.

Logan, B. E., B. Hamelers, et al. (2006). "Microbial fuel cells: methodology and
technology." Environ. Sci. Technol. in press.

López-López, A., E. Expósito, et al. (1999). "Use of Thiobacillus ferrooxidans in a
Coupled Microbiological-Electrochemical System for Wastewater
Detoxification." Biotechnology and Bioengineering 63(1): 79-86.

Lovley, D. R. (2006). "Bug juice: harvesting electricity with microorganisms." Nature
Rev. Microbiology 4: 497-508.

Lovley, D. R., E. J. P. Phillips, et al. (1989). "Hydrogen and Formate Oxidation Coupled
to Dissimilatory Reduction of Iron or Manganese by Alteromonas putrefaciens."
Applied and Environmental Microbiology 55(3): 700-706.

Lowy, D. A., L. M. Tender, et al. (2006). "Harvesting Energy from the Marine Sediment-
Water Interface II: Kinetic Activity of Anode Materials." Biosensors and
Bioelectronics 21(11): 2058-2063.

Luu, Y.-S. and J. A. Ramsay (2003). "Review: microbial mechanisms of accessing
insoluble Fe(III) as an energy source." World Journal of Microbiology &
Biotechnology 19: 215-225.

Manohar, A. K., O. Bretschger, et al. (2008). "The Polarization Behavior of the Anode in
a Microbial Fuel Cell." Electrochimica Acta 53(9): 3508-3513.

Manohar, A. K., O. Bretschger, et al. (2008). "The use of Electrochemical Impedance
Spectroscopy (EIS) in the Evaluation of the Electrochemical Properties of a
Microbial Fuel Cell." Bioelectrochemistry 72(2): 149-154.

149
Marshall, M. J., A. S. Beliaev, et al. (2006). "c-Type Cytochrome-Dependent Formation
of U(IV) Nanoparticles by Shewanella oneidensis." PLoS Biology 4(8): 1324-
1333.

Marsili, E., D. B. Baron, et al. (2008). "Shewanella Secretes Flavins that Mediate
Extracellular Electron Transfer." Proceedings of the National Acadamy of
Sciences, United States 105(10): 3968-3973.

Marx, C. J. and M. E. Lidstrom (2002). "Broad-Host-Range cre-lox System for Antibiotic
Marker Recycling in Gram-Negaive Bacteria." Biotechniques 33(5): 1062-1067.

Matsunaga, T., I. Karube, et al. (1979). "Electrode System for the Determination of
Microbial Populations." Applied and Environmental Microbiology 37(1): 117-121.

Matsunaga, T., I. Karube, et al. (1980). "Electrochemical Determination of Cell
Populations." European Journal of Applied Microbiology and Biotechnology 10:
125-132.

May, P. S., G. C. Blanchard, et al. (1964). Biochemical Hydrogen Generators. 18th
Annual Power Sources Conference.

Meyer, T. E., A. S. Tsapin, et al. (2004). "Identification of 42 Possible Cytochrome C
Genes in the Shewanella oneidensis Genome and Characterization of Six Soluble
Cytochromes." OMICS: A Journal of Integrative Biology 8(1): 57-77.

Min, B., J. Kim, et al. (2005). "Electricity Generation from Swine Wastewater using
Microbial Fuel Cells." Water Research 39(20): 4961.

Min, B. and B. E. Logan (2004). "Continuous Electricity Generation from Domestic
Wastewater and Organic Substrates in a Flat Plate Microbial Fuel Cell."
Environmental Science and Technology 38(21): 5809-5814.

Moon, H., I. S. Chang, et al. (2004). "Improving the Dynamic Response of a Mediator-
less Microbial Fuel Cell as a Biochemical Oxygen Demand (BOD) Sensor."
Biotechnology Letters 26(22): 1717.

Moon, H., In Seop Chang, Jae Kyung Jang, Kyung Shik Kim, Jiyoung Lee, Robert W.
Lovitt, and Byung Hong Kim (2005). "On-Line Monitoring of Low Biochemical
Oxygen Demand Through Continuous Operation of a Mediator-Less Microbial
Fuel Cell." Journal of Microbiology and Biotechnology 15(1): 192-196.

Murray, A. E., D. Lies, et al. (2001). "DNA/DNA Hybridization to Microarrays Reveals
Gene-Specific Differences Between Closely Related Microbial Genomes."
Proceedings of the National Acadamy of Sciences, United States 98: 9853-9858.

150
Myers, C. R. and J. M. Myers (1992). "Localization of Cytochromes to the Outer
Membrane of Anaerobically Grown Shewanella putrefaciens MR-1." Journal of
Bacteriology 174(11): 3429-3438.

Myers, C. R. and J. M. Myers (1997). "Cloning and Sequence of cymA, a Gene Encoding
a Tetraheme Cytochrome c Required for Reduction of Iron(III), Fumarate, and
Nitrate by Shewanella putrefaciens MR-1." Journal of Bacteriology 179(4): 1143-
1152.

Myers, C. R. and J. M. Myers (2002). "MtrB Is Required for Proper Incorporation of the
Cytochromes OmcA and OmcB into the Outer Membrane of Shewanella
putrefaciens MR-1." Applied and Environmental Microbiology 68(11): 5585-
5594.

Myers, C. R. and J. M. Myers (2003). "Cell Surface Exposure of the Outer Membrane
Cytochromes of Shewanella oneidensis MR-1." Letters in Applied Microbiology
37: 254-258.

Myers, C. R. and J. M. Myers (2004). "The Outer Membrane Cytochromes of Shewanella
oneidensis MR-1 are Lipoproteins." Letters in Applied Microbiology 39: 466-470.

Myers, C. R. and K. H. Nealson (1988). "Bacterial Manganese Reduction and Growth
with Manganese Oxide as the Sole Electron Acceptor." Science 240(4857): 1319-
1321.

Myers, C. R. and K. H. Nealson (1988). "Microbial Reduction of Manganese Oxides:
Interactions with Iron and Sulfur." Geochimica et Cosmochimica Acta 52(11):
2727-2732.

Myers, C. R. and K. H. Nealson (1990). "Respiration-linked proton translocation coupled
to anaerobic reduction of manganese(IV) and iron(III) in Shewanella putrefaciens
MR-1." Journal of Bacteriology 172(11): 6232-6238.

Myers, J. M. and C. R. Myers (2000). "Role of the Tetraheme Cytochrome CymA in
Anaerobic Electron Transport in Cells of Shewanella putrefaciens MR-1 with
Normal Levels of Menaquinone." Journal of Bacteriology 182(1): 67-75.

Myers, J. M. and C. R. Myers (2001). "Role for Outer Membrane Cytochromes OmcA
and OmcB of Shewanella putrefaciens MR-1 in Reduction of Manganese
Dioxide." Applied and Environmental Microbiology 67(1): 260-269.

Myers, J. M. and C. R. Myers (2003). "Overlapping Role of the Outer Membrane
Cytochromes of Shewanella oneidensis MR-1 in the Reduction of Manganese(IV)
Oxide." Letters in Applied Microbiology 37: 21-25.

151
Nealson, K. and J. Scott (2006). Ecophysiology of the Genus Shewanella. The
Prokaryotes: 1133-1151.

Nealson, K. H., C. R. Myers, et al. (1991). "Isolation and Identification of Manganese-
Reducing Bacteria and Estimates of Microbial Mn(IV)-Reducing Potential in the
Black Sea." Deep Sea Research Part a - Oceanogaphic Research Papers 38: S907-
S920.

Nealson, K. H. and D. Saffarini (1994). "Iron and Manganese in Anaerobic Respiration:
Environmental Significance, Physiology, and Regulation." Annual Review of
Microbiology 48(1): 311-343.

Newman, D. K. (2001). "MICROBIOLOGY: How Bacteria Respire Minerals." Science
292(5520): 1312-1313.

Nielsen, M. E., C. E. Reimers, et al. (2007). "Enhanced Power from Chambered Benthic
Microbial Fuel Cells." Environmental Science and Technology 41(22): 7895-7900.

Niessen, J., U. Schröder, et al. (2004). "Fluorinated polyanilines as superior materials for
electrocatalytic anodes in bacterial fuel cells." Electrochemistry Communications
6(6): 571.

Nishikawa, S., S. Sakai, et al. (1982). "Dye-Coupled Electrode System for the Rapid
Determination of Cell Populations in Polluted Water." Applied and
Environmental Microbiology 43(4): 814-818.

Obraztsova, A. and Y. Wang (2008). Physiological Differences Among Different
Shewanella Strains.

Obuekwe, C. O. and D. W. S. Westlake (1982). "Effects of Medium Composition on Cell
Pigmentation, Cytochrome Content, and Ferric iron Reduction in a Pseudomonas
sp. Isolated from Crude-Oil." Canadian Journal of Microbiology 28: 989-992.

Oh, S. and B. E. Logan (2005). "Hydrogen and Electricity Production from a Food
Processing Wastewater using Fermentation and Microbial Fuel Cell
Technologies." Water Research 39: 4673-4682.

Oh, S. and B. E. Logan (2006). "Proton Exchange Membrane and Electrode Surface
Areas as Factors that Affect Power Generation in Microbial Fuel Cells." Applied
Microbiology and Biotechnology 70: 162-169.

Oh, S., B. Min, et al. (2004). "Cathode Performance as a Factor in Electricity Generation
in Microbial Fuel Cells." Environmental Science and Technology 38(18): 4900-
4904.

152
Park, D. H. and G. J. Zeikus (2002). "Impact of Electrode Composition on Electricity
Generation in a Single-Compartment Fuel Cell using Shewanella putrefaciens."
Applied Microbiology and Biotechnology 59(1): 58.

Park, D. H. and G. J. Zeikus (2003). "Improved Fuel Cell and Electrode Designs for
Producing Electricity from Microbial Degradation." Biotechnology and
Bioengineering 81(3): 348-355.

Park, H. S., B. H. Kim, et al. (2001). "A Novel Electrochemically Active and Fe(III)-
reducing Bacterium Phylogenetically Related to Clostridium butyricum Isolated
from a Microbial Fuel Cell." Anaerobe 7(6): 297.

Pham, C. A., S. J. Jung, et al. (2003). "A Novel Electrochemically Active and Fe(III)-
Reducing Bacterium Phylogenetically Related to Aeromonas hydrophila, Isolated
from a Microbial Fuel Cell." FEMS Microbiology Letters 223(1): 129.

Pham, T. H., K. Rabaey, et al. (2006). "Microbial Fuel Cells in Relation to Conventional
Anaerobic Digestion Technology." Engineering in Life Sciences 6(3): 285-292.

Phung, N. T., J. Lee, et al. (2004). "Analysis of Microbial Diversity in Oligotrophic
Microbial Fuel Cells Using 16S rDNA Sequences." FEMS Microbiology Letters
233(1): 77.

Pitts, K. E., P. S. Dobbin, et al. (2003). "Characterization of the Shewanella oneidensis
MR-1 Decaheme Cytochrome MtrA: Expression in Escherichia coli Confers the
Ability to Reduce Soluble Fe(III) Chelates." Journal of Biological Chemistry
278(30): 27758-27765.

Potter, M. C. (1911). "Electrical Effects Accompanying the Decomposition of Organic
Compounds." 84(571): 260-276.

Rabaey, K., N. Boon, et al. (2004). "Biofuel Cells Select for Microbial Consortia That
Self-Mediate Electron Transfer." Appl. Environ. Microbiol. 70(9): 5373-5382.

Rabaey, K., G. Lissens, et al. (2003). "A Microbial Fuel Cell Capable of Converting
Glucose to Electricity at High Rate and Efficiency." Biotechnology Letters
25(18): 1531.

Rabaey, K., S. T. Read, et al. (2008). "Cathodic Oxygen Reduction Catalyzed by Bacteria
in Microbial Fuel Cells." ISME J.

Rabaey, K., J. Rodriguez, et al. (2007). "Microbial Ecology Meets Electrochemistry:
Electricity-Driven and Driving Communities." ISME J 1(1): 9-18.

153
Rabaey, K., K. VandeSompel, et al. (2006). "Microbial Fuel Cells for Sulfide Removal."
Environmental Science and Technology 40(17): 5218-5224.

Rabaey, K. and W. Verstraete (2005). "Microbial Fuel Cells: Novel Biotechnology for
Energy Generation." Trends in Biotechnology 23(6): 291.

Rao, J. R., G. J. Richter, et al. (1976). "The Performance of Glucose Electrodes and the
Characteristics of Different Biofuel Cell Constructions." Bioelectrochemistry and
Bioenergetics 3(1): 139-150.

Reimers, C. E., L. M. Tender, et al. (2001). "Harvesting Energy from the Marine
Sediment-Water Interface." Environmental Science and Technology 35(1): 192-
195.

Reynolds, L. W. and J. J. Konikoff (1963). "A Preliminary Report on Two
Bioelectrogenic Systems." Developments in Industrial Microbiology 4: 59-69.

Rhoads, A., H. Beyenal, et al. (2005). "Microbial Fuel Cell using Anaerobic Respiration
as an Anodic Reaction and Biomineralized Manganese as a Cathodic Reactant."
Environmental Science and Technology 39(12): 4666-4671.

Ringeisen, B. R., E. Henderson, et al. (2006). "High Power Density from a Miniature
Microbial Fuel Cell Using Shewanella oneidensis DSP10." Environmental
Science and Technology 40(8): 2629-2634.

Rohrback, G. H. (1962). Biochemical Fuel Cells. 16th Annual Power Sources Conference.

Roller, S. D., P. H. Bennetto, et al. (1984). "Electron-Transfer Coupling in Microbial
Fuel Cells: 1. Comparison of Redox-Mediator Reduction Rates and Respiratory
Rates of Bacteria." Journal of Chemical Technology and Biotechnology 34(B): 3-
12.

Rozendal, R. A., A. W. Jeremiasse, et al. (2008). "Hydrogen Production with a Microbial
Biocathode." Environmental Science and Technology 42(2): 629-634.

Ruebush, S. S., S. L. Brantley, et al. (2006). "Reduction of Soluble and Insoluble Iron
Forms by Membrane Fractions of Shewanella oneidensis Grown under Aerobic
and Anaerobic Conditions." Applied and Environmental Microbiology 72(4):
2925-2935.

Saffarini, D. A., R. Schultz, et al. (2003). "Involvement of Cyclic AMP (cAMP) and
cAMP Receptor Protein in Anaerobic Respiration of Shewanella oneidensis."
Journal of Bacteriology 185(12): 3668-3671.

154
Saltikov, C. W., A. Cifuentes, et al. (2003). "The ars Detoxification System Is
Advantageous but Not Required for As(V) Respiration by the Genetically
Tractable Shewanella Species Strain ANA-3." Applied and Environmental
Microbiology 69(5): 2800-2809.

Schröder, U., J. Nießen, et al. (2003). "A Generation of Microbial Fuel Cells with Current
Outputs Boosted by More Than One Order of Magnitude." Angewandte Chemie
International Edition 42(25): 2880-2883.

Schwalb, C., S. K. Chapman, et al. (2002). "The Membrane-Bound Tetrahaem c-type
Cytochrome CymA Interacts Directly with the Soluble Fumarate Reductase in
Shewanella." Biochemical Society Transactions 30(4): 658-662.

Scott, J. H. and K. H. Nealson (1994). "A Biochemical Study of the Intermediary Carbon
Metabolism of Shewanella putrefaciens." Journal of Bacteriology 176(11): 3408-
3411.

Shaw, M. (1963). Biochemical Fuel Cells, Part 1. 17th Annual Power Sources
Conference.

Sisler, F. D. (1961). "Electrical energy from Biochemical Fuel Cells." New Scientist 12:
110-111.

Sisler, F. D. (1962). "Electrical Energy from Microbiological Processes." Journal of the
Washington Academy of Sciences 52: 181-187.

Stookey, L. L. (1970). "Ferrozine - A New Spectrophotometric Reagent for Iron."
Analytical Chemistry 42(7): 779-781.

Suzuki, S., I. Karube, et al. (1978). Application of a Biochemical Fuel Cell to
Wastewaters. Biotechnology and Bioengineering Symposium, United States.

Suzuki, S., I. Karube, et al. (1980). "Biochemical Energy Conversion using Immobilized
Whole Cells of Clostridium butyricum." Biochimie 62(5-6): 353-358.

Tang, Y. J., A. L. Meadows, et al. (2007). "Anaerobic Central Metabolic Pathways in
Shewanella oneidensis MR-1 Reinterpreted in the Light of Isotopic Metabolite
Labeling." Journal of Bacteriology 189(3): 894-901.

Tender, L. M., C. E. Reimers, et al. (2002). "Harnessing Microbially Generated Power on
the Seafloor." Nature Biotechnology 20(8): 821.

ter Heijne, A., H. V. M. Hamelers, et al. (2007). "Microbial Fuel Cell Operation with
Continuous Biological Ferrous Iron Oxidation of the Catholyte." Environmental
Science and Technology 41(11): 4130-4134.

155

ter Heijne, A., H. V. M. Hamelers, et al. (2006). "A Bipolar Membrane Combined with
Ferric Iron Reduction as an Efficient Cathode System in Microbial Fuel Cells."
Environmental Science and Technology 40(17): 5200-5205.

Thauer, R. K., K. Jungermann, et al. (1977). "Energy Conservation in Chemotrophic
Anaerobic Bacteria." Microbiology and Molecular Biology Reviews 41(1): 100-
180.

Thompson, D. K., A. S. Beliaev, et al. (2002). "Transcriptional and Proteomic Analysis
of a Ferric Uptake Regulator (Fur) Mutant of Shewanella oneidensis: Possible
Involvement of Fur in Energy Metabolism, Transcriptional Regulation, and
Oxidative Stress." Applied and Environmental Microbiology 68(2): 881-892.

Turner, A. P. F. and G. Ramsay (1983). "Applications of Electron Transfer Between
Biological Systems and Electrodes." Biochemical Society Transactions 11: 445-
448.

van Hees, W. (1965). "A Bacterial Methane Fuel Cell." Journal of The Electrochemical
Society 112(3): 258-262.

Venkateswaran, K., M. E. Dollhopf, et al. (1998). "Shewanella amazonensis sp. nov., A
Novel Metal-Reducing Facultative Anaerobe from Amazonian Shelf Muds."
International Journal of Systematic Bacteriology 48(3): 965-972.

Venkateswaran, K., D. P. Moser, et al. (1999). "Polyphasic Taxonomy of the Genus
Shewanella and Description of Shewanella oneidensis sp. nov." International
Journal of Systematic and Evolutionary Microbiology 49(2): 705-724.

Videla, H. A. and A. J. Arvía (1975). "The Response of a Bioelectrochemical Cell with
Saccharomyces cerevisiae Metabolizing Glucose Under Various Fermentation
Conditions." Biotechnology and Bioengineering 17(10): 1529-1543.

von Felten, P., H. Zurrer, et al. (1985). "Production of Molecular Hydrogen with
Immobilized Cells of Rhodospirillum rubrum." Applied Microbiology and
Biotechnology 23: 15-20.

von Wolzogen Kuhr, C. A. H. and L. S. van der Vlugt (1934). "Graphitization of Cast
Iron as an Electro-Biochemical Process in Anaerobic Soils." Water (The Hague)
18(16): 147-165.

Wan, X.-F., N. C. VerBerkmoes, et al. (2004). "Transcriptomic and Proteomic
Characterization of the Fur Modulon in the Metal-Reducing Bacterium
Shewanella oneidensis." Journal of Bacteriology 186(24): 8385-8400.

156
Weber, K. A., L. A. Achenbach, et al. (2006). "Microorganisms Pumping Iron: Anaerobic
Microbial Iron Oxidation and Reduction." Nature Reviews Microbiology 4(10):
752-764.

Wilkins, J. R. (1978). "Use of Platinum Electrodes for the Electrochemical Detection of
Bacteria." Applied and Environmental Microbiology 36(5): 683-687.

Wilkins, J. R., D. C. Grana, et al. (1980). "Combined Membrane Filtration-
Electrochemical Microbial Detection Method." Applied and Environmental
Microbiology 40(4): 852-853.

Yahiro, A. T., S. M. Lee, et al. (1964). "I. Enzyme Utilizing Bio-Fuel Cell Studies."
Biochimica et Biophysica Acta 88: 375-383.

Yang, M. Y. and Z. Yang (1988). "Development of Microbial Electrode and Studies on
its Cathodic and Anodic Reactions." Chinese Journal of Biotechnology 4: 138-
144.

Yost, C., L. Hauser, et al. (2003). "A Computational Study of Shewanella oneidensis
MR-1: Structural Prediction and Functional Inference of Hypothetical Proteins."
OMICS: A Journal of Integrative Biology 7(2): 177-191.

Young, T. G., L. Hadjipetrou, et al. (1966). "The Theoretical Aspects of Biochemical
Fuel Cells." Biotechnology and Bioengineering 8(4): 581-593.

Zhang, L.-h., J.-p. Jia, et al. (2005). "Electrochemical Effect on Denitrification in
Different Microenvironments Around Anodes and Cathodes." Research in
Microbiology 156(1): 88-92.

Ziemke, F., M. G. Hofle, et al. (1998). "Reclassification of Shewanella putrefaciens
Owen's Genomic Group II as Shewanella baltica sp. nov." International Journal of
Systematic and Evolutionary Microbiology 48(1): 179-186.

Zuo, Y., S. Cheng, et al. (2007). "Tubular Membrane Cathodes for Scalable Power
Generation in Microbial Fuel Cells." Environmental Science and Technology
41(9): 3347-3353.

157
Appendix A Electrode and membrane pretreatment procedures

Graphite Felt Electrodes

The graphite felt electrodes (Electrolytica, GF-S6-06, 6mm thick) were cut to an
appropriate size and pretreated to make them hydrophilic. Platinum leads were attached
to each electrode and some were electroplated with platinum to be cathodes, while the
rest remained anodes. Anode and cathode electrodes were cleaned after each use and
recycled.
Electrode pretreatment

The pretreatment included a thirty minute soak in 95% ethanol followed by a
thorough rinse with deionized water. The rinsed electrodes were then placed in an
agitated 0.1N HCL solution and allowed to soak overnight. Following the HCL soak the
electrodes were rinsed thoroughly with deionized water and dried in an 80ûC oven.
Platinum leads were attached to each pretreated electrode by weaving platinum
wire (Alfa-Aesar, 0.3 mm, platinum wire) through the graphite felt and adhering the wire
to the felt with a mixture of graphite epoxy (Electrolytica, EPOX-4), diluted with toluene.
The graphite epoxy mixture was applied to an exposed portion of the platinum wire,
pressed into the felt with a Teflon sheet, and allowed to cure in an 80ûC oven overnight.
The resistance of each electrode assembly was tested and found to be between 1 and 7
ohm.

158
Electrode Cleaning

Graphite felt electrodes were cleaned by sonnication for two to three hours, then
soaking in an agitated 0.1N HCl solution overnight. Electrodes were rinsed thoroughly
and soaked overnight in a 2M KOH solution. Each electrode was electrochemically
cleaned in a fresh 2M KOH solution using the graphite felt as the working electrode, a
Ag/AgCl reference electrode (Bioanalytical Systems, RE-5B, 6mm diameter, 7.5 cm length)
and platinum wire as the counter electrode. A potential of +1V (vs. Ag/AgCl) was
applied potentiostatically to the working electrode for ten minutes. The
electrochemically cleaned electrodes were sonicated in deionized water for thirty minutes
with two exchanges of water, to remove any residual KOH, and dried in an oven
overnight at 80ûC.
Cathode electroplating

Some pretreated and electrochemically cleaned electrodes were converted to
cathodes by electroplating. Each electrode was weighed and recorded prior to the plating
step. Plating solution was composed of citric acid, sulfuric acid, nitric acid and 1g/100
mL of dihydrogen hexachloroplatinate(IV) hexahydrate (Alfa-Aesar, #11051). The
solution was sonicated for fifteen minutes and added to a clean beaker featuring the
graphite electrode as the working electrode, a Ag/AgCl reference electrode and a
platinum wire counter electrode. The solution was agitated using a magnetic stir bar and
a potentiostat (Gamry, Reference 600) was used in potentiostatic mode to apply a
constant potential of -0.2 V (vs Ag/AgCl) for seven minutes.

159
The graphite felt cathodes were rinsed in deionized water to remove traces of the
plating solution and cured in an 80ûC oven, overnight. The cathodes were weighed after
curing to determine the amount of platinum plated on each electrode.
Nafion membranes

The Nafion membranes (424 and 117) were pretreated to make them proton
selective. Membranes were soaked in a 5% (by volume) HCl solution for two to three
hours to hydrate and expand the dimensions. The membranes were then rinsed in
deionized water and cleaned in a 3% solution of boiling hydrogen peroxide. After the
cleaning step, the membranes were boiled in 0.5M sulfuric acid for protonation.
Membranes were finally rinsed and stored in deionized water until use.

160
Appendix B SEM preparation of biological samples

Fix wet sample in 2.5% gluteraldehyde solution under the same anaerobic/aerobic
conditions as the experiment was conducted. Let samples sit in gluteraldehyde for
several hours or overnight depending on the material that the bacteria are sitting on.
Store at 4° C and keep away from light while samples are sitting in gluteraldehyde.

Gently rinse by consecutively exchanging gluteraldehyde solutions with:
1. 100% buffer
2. Buffer/DI-H2O or DI-H2O
Slowly dehydrate the fixed sample:
1. 10% EtOH (15 min)
2. 25% EtOH (15 min)
3. 50% EtOH (15 min)
4. 75% EtOH (15 min)
5. 95% EtOH (15 min)
6. 100% EtOH (2x 15 min)
7. Critical Point Drying or HMDS
Mount sample using carbon tabs and Al stubs

161
Appendix C Organic acid sample preparation for HPLC analysis

1) Autoclave HPLC vials with caps and septa loosely attached (liquid cycle).
2) Extract at least 0.7 mL of sample and place in sterilized microcentrifuge tube.
3) Centrifuge samples for 5 minutes at 13000 rpm or 4 min at 14000 rpm to pellet
biomass.
4) Extract supernatant and transfer to clean microcentrifuge tube. Throw out pellet.
5) Freeze (at -20°C or -80°C) if necessary, otherwise move to step 6.
6) Acidify sample using 10 µL of 1.25 M H
2
SO
4
per 1 mL of sample. 1.25 M H
2
SO
4
is
prepared in a small bottle in the corrosives cabinet.
7) If you anticipate that a precipitate will form in the sample after acidification (e.g.
PIPES is contained in the media) let acidified samples sit overnight at 4°C.
8) Individually, filter sterilize every sample into autoclaved HPLC vials using 0.2 µm
syringe filters. Filters, syringes, vials and caps are kept on the shelves next to the
HPLC.
9) Place vials in the HPLC tray in the order that you have defined your sequence table.

Abstract (if available)

Abstract The mechanisms that bacteria employ to transfer electrons to their surrounding environments are diverse and not well understood. This study provides original data that begin to elucidate specific mechanisms involved with electron transfer to microbial fuel cell (MFC) electrodes, Fe(III)- and Mn(IV)-oxides using various strains and species of the genus Shewanella. Additionally, MFC were used to study the carbon and energy metabolisms of several different Shewanella strains, identify efficiencies, and study the physiology of Shewanella oneidensis MR-1. These data have implications toward the optimization of bioremediation technologies, MFC development and to the fundamental understanding of how bacteria interact with and affect their environments.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Applications of advanced electrochemical techniques in the study of microbial fuel cells and corrosion protection by polymer coatings

PDF

From fuel cells to single cells: electrochemical measurements of direct electron transfer at microbial-electrode interfaces

PDF

Survival and evolution of Shewanella oneidensis MR-1: applications for microbial fuel cells

PDF

Bacterial nanowires of Shewanella oneidensis MR-1: electron transport mechanism, composition, and role of multiheme cytochromes

PDF

Electrokinetic transport of Cr(VI) and integration with zero-valent iron nanoparticle and microbial fuel cell technologies for aquifer remediation

PDF

From single molecules to bacterial nanowires: functional and dynamic imaging of the extracellular electron transfer network in Shewanella oneidensis MR-1

PDF

Electrochemical studies of outward and inward extracellular electron transfer by microorganisms from diverse environments

PDF

From cables to biofilms: electronic and electrochemical characterization of electroactive microbial systems

PDF

Design, dynamics, and control of miniature catalytic combustion engines and direct propane PEM fuel cells

PDF

Extracellular electron transport: Investigating the diversity and mechanisms behind an understudied microbial process with global implications

PDF

Electronic, electrochemical, and spintronic characterization of bacterial electron transport

PDF

Studies on the flame dynamics and kinetics of alcohols and liquid hydrocarbon fuels

PDF

Using sequencing techniques to explore the microbial communities associated with ferromanganese nodules and sediment from the South Pacific gyre

PDF

Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction

PDF

Electrochemical investigations and imaging tools for understanding extracellular electron transfer in phylogenetically diverse bacteria

PDF

The modification of catalysts and their supports for use in various fuel cells

$On the dynamic fracture behavior of polymeric materials subjected to extreme conditions$

PDF

On the dynamic fracture behavior of polymeric materials subjected to extreme conditions

PDF

Understanding the role of electrode design in determining the electrochemical performance of high-energy/high-power lithium-ion and lithium-sulfur batteries

PDF

Phase change heterostructures for electronic and photonic applications

PDF

Reactive and quantum molecular dynamics study of materials: from oxidation and amorphization to interfacial charge transfer

Asset Metadata

Creator Bretschger, Orianna (author)

Core Title Electron transfer capability and metabolic processes of the genus Shewanella with applications to the optimization of microbial fuel cells

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Materials Science

Degree Conferral Date 2008-08

Publication Date 08/01/2008

Defense Date 06/16/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag microbial fuel cells,OAI-PMH Harvest,Shewanella

Language English

Advisor Mansfeld, Florian B. (committee chair), Nealson, Kenneth H. (committee member), Nutt, Steven R. (committee member), Prakash, Surya (committee member)

Creator Email bretschg@usc.edu,obretschger@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1509

Unique identifier UC1286574

Identifier etd-Bretschger-2182 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-98971 (legacy record id),usctheses-m1509 (legacy record id)

Legacy Identifier etd-Bretschger-2182.pdf

Dmrecord 98971

Document Type Dissertation

Rights Bretschger, Orianna

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu