Microbial fuel cell-based biosensor for redox potential detection in

fermentation process

A Thesis Submitted to the

College of Graduate and Postdoctoral Studies
In Partial Fulfillment of the Requirements
For the Degree of

Master of Science

In the Department of Chemical and Biological Engineering

University of Saskatchewan
Saskatoon

By

Siyang Shen

© Copyright Siyang Shen, January 2022. All rights reserved.

Unless otherwise noted, copyright of the material in this thesis belongs to the author.
 PERMISSION TO USE

In presenting this thesis in partial fulfilment of the requirements for a Master of Science
degree from the University of Saskatchewan, I agree that the Libraries of this University may make
it freely available for inspection. I further agree that permission for copying of this thesis in any
manner, in whole or in part, for scholarly purposes may be granted by the professor or professors
who supervised my thesis work or, in their absence, by the Head of the Department or the Dean of
the College in which my thesis work was done. It is understood that any copying or publication or
use of this thesis or parts thereof for financial gain shall not be allowed without my written
permission. It is also understood that due recognition shall be given to me and to the University of
Saskatchewan in any scholarly use which may be made of any material in my thesis.
Requests for permission to copy or to make other uses of materials in this
thesis/dissertation inwhole or part should be addressed to:

Head of the Department of Chemical and Biological Engineering

University of Saskatchewan
57 Campus Drive
Saskatoon, SK, S7N 5A9, Canada
OR
Dean
College of Graduate and Postdoctoral Studies
University of Saskatchewan
116 Thorvaldson Building, 110 Science Place
Saskatoon, SK, S7N 5C9, Canada

i
 ABSTRACT

Currently, fermentation research's primary objective is to develop economic and

sustainable processes by increasing the product yield and reducing the operating cost. A
tremendous effort had been made in the past few decades by discovering and developing various
fermentative strains. The fermentation process measurement, control, and supervision are the next
battlefield for further fermentation technology development. Fermentation manipulation is a
complex process from both biological and engineering points of view. Different factors, including
substrates type, substrates concentration, strain type, and operating modes, need to be taken into
account. These aspects encourage the scientific community to find a robust, sophisticated, and
versatile measure for controlling fermentation.
Redox potential, also known as the oxidation-reduction potential (ORP), reflects the overall
biochemical reactions, electron transfer, and redox balance within the fermentation broth. The
biological significance of ORP includes indirectly explicating metabolic activities during
fermentation and regulating the metabolic network, affecting the metabolic pathway and gene
expression. The monitoring and controlling environmental ORP provides a thoughtful
understanding to help control the intracellular metabolic activities and fermentation process. This
approach has been proven and considered as a real-time approach to increase fermentation
efficiency in scientific communities and industrial sectors. In particular, the ORP measurement
can provide an online and consistent signal during fermentation. It can be used at any stage of
fermentation, providing both high signal-integrity and measurement reliability. Although
effective, the relatively high fabrication cost of these sensors has so far made it impractical for
extensive applications in large-scale contaminated soil monitoring, in particular.
Cost-effectiveness is an endless effort in engineering, an economically optimized ORP
monitor tool is urgently needed. With the development of bio-electrochemical research, the
microbial fuel cell (MFC) as an old technology has been adopted as a biosensor to produce power
and electricity by bacteria catalyzation. In this work, an MFC-based biosensor device was designed
and developed as a fermentation biosensor using an indigenous microorganism and modified
Nernst equation to integrate among the MFC voltage output, fermentation ORP, and fermentation
stages.

ii
 This study investigated different factors on the performance of an MFC-based biosensor.
These factors include strain types (Bacillus subtilis and Pseudomonas fluorescens) in the presence
or absence of methylene blue mediator, cathodic treatments (sparging, aerated cathode, and
potassium ferricyanide solution), and anodic aeration rates (0, 11.32, and 22.64 vvm). After
optimal conditions were established, this study used turbidimetric measurement as the indicator
for microbial growth. The correlational between microbial growth and ORP, voltage, potential
parameter (X) were investigated. Results showed that B. subtilis exhibited superior performance
under MFC condition. Sparging cathodic treatment provided a feasible and sustainable supply of
electron acceptor. Keeping anodic aeration rate at 11.32 vvm constructed a suitable anodic
environment not only to support B. subtilis growth but also to sustain the voltage generation from
MFC device. After all conditions had been settled, the voltage signal was projected to a linear
increase, and the ORP signal was likely to generate a bathtub-shaped curve. Two peaks occurred
on the curve by integrating both signals into the potential parameter X and plotting the potential
parameter X over time. Three distinct growth phases were revealed by comparing the potential
parameter X with microbial growth information and ORP profile. Potential parameter X indicated
the endpoint of lag phase, mid-point of exponential phase, and the starting point of stationary
phase. Such a result proved that the potential parameters could provide fruitful and high-resolution
information that enables precious and real-time fermentation monitoring and controlling. To
conclude, this thesis demonstrated the development of a novel fermentation biosensor by utilizing
an MFC device as a biosensor. By applying the unitless parameter (i.e., potential parameter)
derived from the modified Nernst equation, an MFC device equipped with an ORP sensor could
successfully unveil the hidden internal information during the course of fermentation and explicate
microbial growth dynamics.
Although successful, many questions were also raised during the course of this research.
One main limiting factor the choice of the microorganism. In this research, Bacillus subtilis was
selected as it could generate extracellular electrons which then pick up by the carbon electrode in
anodic chamber, resulting in voltage flow. When different microbes were chosen, one needs to
investigate whether such a microbial strain could export electron in MFC device. If not, one could
attempt to supplement electron mediator to assist electron movement from microbial surface to
carbon electron. The operating condition for this developed device needs to be optimized as
different strains possess different growth requirements.

iii
 ACKNOWLEDGEMENTS

I want to express my deepest gratitude to my supervisor, Dr. Yen-Han Lin, for his insightful
advice, patience, and invaluable support. I feel very fortunate to have him as my advisor and to
have the opportunity to work with him for the past two years. He has always been available and
willing to provide unerring advice. Working with him made this journey very pleasantly, and for
that, I am deeply grateful.
Thanks to my committee: Professor Hui Wang and Professor Catherine Niu, whose
suggestions have also shaped this thesis, their supervision allowed develop and elaborate new
ideas. I am deeply grateful to all the faculty and staff in the department, especially Kevin Carter,
Majak Mapiour, RLee Prokoishyn, and Dushmanthi Jayasinghe, for their help during my research.
I would also like to recognize my friend and colleague Xiaoyan Huang spent countless
hours educating me with myriad skills and knowledge and providing unfailing advice and
encouragement during my research.
I also feel it is important to mention my gratitude to my girlfriend Yingxin Wang, who has
offered unconditional support and provided care in my daily life. Thank you for the endless love,
encouragement, support, and patience.
Last but not least, my love and gratitude to my parents, Wangtao Ruan and Jiqin Shen, for
raising me to always strive for success, providing endless love and financial support. At the same
time, I work toward my bachelor's and master’s degrees.

iv
 TABLE OF CONTENTS
PERMISSION TO USE................................................................................................................. I
ABSTRACT .................................................................................................................................. II
ACKNOWLEDGEMENTS ....................................................................................................... IV
TABLE OF CONTENTS ............................................................................................................ V
LIST OF FIGURES .................................................................................................................. VII
NOMENCLATURE ................................................................................................................. VIII
1 INTRODUCTION ............................................................................................................... 1
2 LITEATURE REVIEW ....................................................................................................... 3
2.1 Background ..........................................................................................................................................3
2.1.1 Introduction to redox potnetial.................................................................................. 3
2.1.2 Fermentation redox potnetial .................................................................................... 5
2.1.3 Curreent redox potential application in fermentation ............................................... 7
2.1.4 Introduction to microbial fuel cell .......................................................................... 10
2.1.5 Fundmental mechanism .......................................................................................... 14
2.1.6 Electron transfer mechanism in MFC ..................................................................... 14
2.1.7 MFC performance and optimization ....................................................................... 18
2.1.8 Cell potential modeling -The Nernst equation ........................................................ 19
2.1.9 Electrochemical biosensing and MFC-based biosensor ......................................... 20
2.2 Knowledge gap ..................................................................................................................................22
2.3 Hypotheses and objectives .............................................................................................................22
3 MATERIALS AND EXPERIMENTAL METHODS........................................................ 24
3.1 Strain and cultivation ......................................................................................................................24
3.2 Microbial fuel cell-based biosensor design and construction .............................................24
3.3 Analytical procedures .....................................................................................................................26
4 RESULTS AND DISCUSSION ........................................................................................ 27
4.1 Modification of Nernst equation and MFC application.......................................................27
4.2 Effect of different microorganisms on MFC-based biosensor performance ...................29
4.3 Effect of catholyte and catholyte treatments on the MFC-based biosensor
performance .......................................................................................................................................31
4.4 Effect of aeration in the anodic chamber on the MFC-based biosensor performance 34

v
 4.5 Correlation between growth curve and redox potential change in MFC ........................36
4.6 Correlation between growth curve and voltage output from MFC ...................................38
4.7 Correlation analysis from potential parameter (𝒙 = 𝒆 − ∆𝑶𝑹𝑷∆𝑽) ..............................40
5 CONCLUSION AND RECOMMENDATIONS ............................................................... 42
6 REFERENCES .................................................................................................................. 44
7 APPENDICES ................................................................................................................... 56
7.1 Appendix A MFC apparatus and instrumentation ..................................................................56
7.2 Appendix B pH profile during fermentation .............................................................................57

vi
 LIST OF FIGURES

Figure 2.1 Diagram of a laboratorial ORP sensor........................................................................... 5

Figure 2.2 NAD+/NADH cross compartment metabolism ............................................................. 7
Figure 2.3 NAD+/NADH redox couple in pyruvate fermentative degradation .............................. 9
Figure 2.4 Bathtub-shape curve of ORP profile while fermentation ............................................ 10
Figure 2.5 Schematic diagram of double-chambered microbial fuel cellError! Bookmark not
defined.
Figure 2.6 Direct electron transfer via membrane bound protein and via conductive nanowire
structure .............................................................................................................................. 16
Figure 2.7 Illustration of mediated electron transfer within a microbial cell ............................... 18
Figure 3.1 The MFC apparatus diagram ....................................................................................... 26
Figure 4.1 Voltage production and power density of two different bacterial strains over 48 hours
fermentation........................................................................................................................ 31
Figure 4.2 Comparison of voltage signal generation among three different electron acceptor
sources and treatments........................................................................................................ 34
Figure 4.3 The effect of three different aeration rate on voltage production and anodic ORP .... 36
Figure 4.4 Growth of Bacillus subtilis under batch and MFC condition and redox potential change
of anodic chamber in MFC device vs. time........................................................................ 38
Figure 4.5 The growth of Bacillus subtilis and voltage output from MFC system by bacteria cell
activity vs. time .................................................................................................................. 39
Figure 4.6 The potential parameter derivative from modified Nernst equation, Bacillus subtilis
growth curve and differential of bacteria population with respect of time vs. operation time
after inoculation. ..................................................................Error! Bookmark not defined.
Figure 7.1 Actual MFC-based biosensor laboratory setup ........................................................... 56
Figure 7.2 Spectrometer for bacteria growth monitoring ............................................................. 56
Figure 7.3 pH profile during the MFC fermentation, the potential parameter present random
dispersion of change without introduces pH regulator ....................................................... 57

vii
 NOMENCLATURE

List of Abbreviation
ATP Adenosine triphosphate
BOD Biological oxygen demand
COD Chemical oxygen demand
DO dissolved oxygen
GHG Greenhouse gases
GSSG/GSH Glutathione
LB Lysogeny broth
MFC Microbial fuel cell
NAD+/NADH Nicotinamide Adenine Dinucleotide
NADP+/NADPH Nicotinamide adenine dinucleotide phosphate
OD Optical density
ORP oxidation-reduction potential
OX oxidant
PEM Proton exchange membrane
PVA Polyvinyl alcohol
RED reductant

Roman Symbols
Symbol Description Unit
A Surface area cm2
a Gap width μm
CYnor Coulombic yield under normal condition %
CYtox Coulombic yield under toxic condition %
c Analyte concentration mM
Ecell Actual cell potential mV
Eanode Anode potential mV
Ecathode Cathode potential mV

viii
F Faraday constant C/mol
I Current mA
G Bacterial biofilm conductance μS
g Thickness of biofilm μm
K Reaction equilibrium constant Mole L-1
Ks Half-velocity constant g/mL
L Length of electrode cm
n Number of transferred electrons -
Qr Reaction quotient -
R Ideal gas constant J⋅K-1⋅mol-1
T Temperature Kelvin
V Voltage mV
vvm Aeration rate L L-1 min-1
x potential parameter -

Greek Symbols
Symbol Description Unit

ηact Activation polarization mV

ηcon Concentration polarization mV

ηohm Ohmic loss mV

μ Bacterial growth rate min-1
σ Electricity conductivity μS/cm

ix
 1 INTRODUCTION

The fermentation technology predates science. Harnessed for food production and
preservation, it appears in varied cultures from all over the world, throughout recorded history.
Before the era of modern storage, storage technologies were limited, and spoiling was a significant
issue for humanity, fermentation enables the production of long-term storable food. It has been a
crucial part of human culinary history and plays a vital role in the development of civilization
(Zilberman & Kim, 2011). Biochemically, fermentation is the most common activity that is widely
distributed in all biological systems. Microorganisms such as bacteria and eukaryotes perform
metabolic activities by degrading organic substrates to obtain energy from ATP synthesis.
Traditionally, the typical products of fermentation as foods were limited. Since the
establishment of modern microbiology by Pasteur in the 19th century, the domestication of
microorganisms became possible. Fermentation has emerged and flourished as an industry,
producing and modifying such foods as cheeses, yogurts, vegetables, fruits, meat, and alcohol.
Until today, modern industrial fermentation has become a pillar of modern industry and extended
its definition to convert raw materials into industrial products (Humphrey & Lee, 1992). The rapid
development of biotechnology accelerates the growth of fermentation industry from diverse
sectors over the last several decades, the major products of fermentation include but are not limited
to food, beverage, amino acids, biopolymers, biofuel, solvents, and drugs (Coton et al., 2017).
Economically, the size of global fermentation chemical market values around 58.68 billion U.S.
dollar in 2018 (GrandViewResearch, 2019), with alcohols, enzymes and organic acids comprising
its largest sectors.
Furthermore, fermentation is considered environmentally friendly as it utilizes sustainable
resources and emits relatively low greenhouse gas (GHG), thereby maintaining a small carbon
footprint. Increased ecological awareness among consumers is driving growth and investment in
fermentation as a key player in our increasingly bio-aware society (Asveld & Van Est, 2011).
Since 2000, studies have discovered that extremophilic microbes had special metabolic
utility for industrial fermentation. New areas of research quickly emerged in the biodiesel, biogas,

1
and bioremediation sectors (María Martínez-Espinosa, 2020). Commercial applications aim to
optimize the quality and quantity of final product in a large-scale operation through the
manipulation and monitoring of the microbial biochemical processes. Nevertheless, the
microorganism plays a crucial role in the fermentation process. While dissolved oxygen (DO)
measurement is one of the most common assays, anaerobic or micro-aerobic cultures will not
produce sufficient oxygen for this to work. Since other parameters, such as pH or temperature, are
incapable of reflecting the actual cell metabolic states within the fermenter (Liu et al., 2013), the
redox potential, or oxidation-reduction potential (ORP), has been proposed as a promising
alternative for the online monitoring of anaerobic fermentation processes (Bonan et al., 2020).
ORP is an electrochemical measure of the reactive chemical environment within a solution
that determines the rate and type of redox reaction. More importantly, the biological redox system
plays an essential role in microbial metabolism. ORP reflects the overall electron movement and
redox balance for both intracellular and intercellular of a biological system. Measurement of ORP
of the system appears to be a valid approach to investigate the dynamics of microbial metabolism.
A previous study (Liu et al., 2013) shows the enzymes related to the oxidation and reduction
reaction in a biological system are redox potential sensitive, impose redox potential regulation,
and redirect intracellular metabolic flux to desired secondary metabolite product, eventually
leading towards more effective fermentation outcome. However, a problem arises in that an ORP
sensor (constructed of an Ag/AgCl reference electrode, 99% pure platinum band encapsulated in
the potassium chloride solution, and the electrode itself) is prone to fouling. Any disturbance in
solution or metabolic residual on the sensor surface will lead to misrepresentative readings.
Furthermore, the high fabrication cost and maintenance costs of such a sensor limit the ORP
technique to the large scale of industrial fermentation.
A microbial fuel cell (MFC) is a device that generates electricity from anaerobic microbes.
In a typical MFC, the microorganisms in the anodic chamber oxidize the substrate, producing
protons and electrons. The latter then travel through an external circuit to the cathodic chamber,
generating a voltage between the two chambers. Accordingly, intracellular electron transfer and
redox balance, driven by the metabolism of microorganisms, affects the ORP. The MFC then
detects and captures the electrons within the extracellular environment and generates an electrical
signal. Intriguingly, the MFC has the dual benefit of serving as both a power generator and a cost-
effective replacement for the fermentation redox sensor. Although MFCs remain novel as

1
biosensors, ongoing research includes environmental carbon and toxin detection (Chouler & Di
Lorenzo, 2015; Cui et al., 2019; Do et al., 2020; Zhou et al., 2018) as well as use as an ORP sensor.
The overall goals of this research are to investigate and elucidate the relationship between
electricity generation and ORP variation within an MFC system and determine the suitability of
integrating an MFC device with an ORP sensor. In this thesis, Chapter 1 outlined the fermentation
and industrial fermentation from a historical perspective and briefly introduced the ORP
technology and MFC technology. Chapter 2 provided a review on the importance of ORP at the
biological level, the utilization of ORP technology for fermentation, and the application of current
MFC technology for both electricity production and biosensing. Chapter 3 listed the choice of
microorganisms, the materials used in an MFC device, data collection procedure, and analytical
methods used in this thesis. Chapter 4 illustrated the derivation of the Nernst equation, different
microbial strains, different cathodic and anodic treatments on MFC-based biosensor performance,
and the correlation among microbial growth, ORP, MFC, and potential parameter under MFC
conditions. Chapter 5 concluded this thesis by reformulating the Nernst equation, and a tight
relationship that associated between ORP and voltage production in an MFC device. Finally, some
possible future applications and improvements of the developed MFC-based biosensor were also
provided in Chapter 5.

2
 2 LITEATURE REVIEW

2.1 Background
2.1.1 Introduction to redox potnetial
Redox potential (ORP) is a specific indicator of its oxidizing or reducing power intensity
depending on the electrochemical balance. The redox potential is the outcome of the redox
(oxidation and reduction) reaction within the system. While the redox reaction is one of the most
fundamental biochemical reactions, the modern exploration of redox research only traces back to
the 18th century. After the scientific community overthrew the old phlogiston theory, Antoine-
Laurent Lavoisier proposed a widely-accepted explanation of combustion, which elucidated the
nature of oxygen, identified oxygen as an essential element for discharge, and recognized the
oxygen and hydrogen elements soon after (Karamanou & Androutsos, 2013). The theory is
widespread and accepted by the research community. Scientists concluded that reactions in which
oxygen was consumed were classified as oxidation, while those in which oxygen was lost were
term reductions. The rapid development of electrochemistry during the 19th and 20th centuries shed
further light on the redox reactions, which were soon implicated in the galvanic cell: ferrous ions
could change ferrous oxide at the anodic electrode, though the full underlying principles were not
yet clear (Ferguson, 1923).
In 1956, Rudolph A. Marcus developed the first electron transfer theory which elucidated
the driving force of redox reaction and interpreted the redox reaction’s rate by a parabola (Marcus,
1956). His theory explains oxidation and reduction reactions, laying a foundation for further
electrochemical and biological research, and he was awarded the Nobel prize in chemistry in 1992.
Integrating Marcus’s theory with the electronegativity scale, Jensen set standards of redox
potential, which could be used to monitor the redox dynamics of a system (Jensen, 1996).
A redox reaction involved two components, including oxidizing agent which gain electron
to form a reducing agent, as described by:
𝑂𝑥 + 𝑛𝑒 − = 𝑅𝑒𝑑 (2.1)

3
In a redox reaction, the electron is transferred between the redox pair. Both oxidation and reduction
reactions take place simultaneously. For example, consider the reduction of copper by zinc, which
can be viewed as two half-reactions:
Oxidation reaction: 𝑍𝑛 → 𝑍𝑛2+ + 2𝑒 − (2.2)
Reudction reaction: 𝐶𝑢2+ + 2𝑒 − → 𝐶𝑢 (2.3)
Overall reaction: 𝑍𝑛 + 𝐶𝑢2+ → 𝐶𝑢 + 𝑍𝑛2+ (2.4)
For the reduction half reaction, the equilibrium constant for the reaction can be represented as:
𝐾 = [𝐶𝑢]/[𝐶𝑢2+ ][𝑒 − ]2 (2.5)
Rearranging the above equation, we see that the electron activity of redox potential can be isolated
as:
1
𝑒 − = [ [𝐶𝑢]/(𝐾 ∗ [𝐶𝑢2+ ])]2 (2.6)
or
1 (𝐶𝑢)
𝑝𝐸 = − 𝑙𝑜𝑔 𝑒 − = 2 [𝑙𝑜𝑔𝐾 − 𝑙𝑜𝑔 (𝐶𝑢2+ )] (2.7)

The standard redox potential between two different chemical species can be predicted by
redox half-reaction. By determining the potential difference between the chemical reaction and a
standard hydrogen electrode, a standard redox potential table of any chemicals and chemical
species can be established. The standard redox potential is a systematic measurement that
represents the intrinsic capacity of molecules or chemical half reactions’ tendency to be either
reduced or oxidized. Similarly, using this relationship, a standard redox potential table can be
produced to give the intrinsic capacity of molecules or chemical half-reactions to be reduced or
oxidized. A lower redox potential indicates a reductive state prone to losing electrons and vice
versa. The difference in redox potential between an electron donor and an acceptor determines the
driving force of the electron transfer reaction:
𝛥G = −𝑛 ∗ 𝐹 ∗ 𝛥E (2.8)
where n is the number of electrons transferred, F is the Faraday constant, and ΔE is the difference
in redox potential. Generally, the standard redox potential for a half-reaction in an aqueous solution
is based on the assumption that the concentration of all chemical species is 1 molarity. In practice,
the reaction kinetics are complex and depend on the specific molarities. Typically, the Nernst
equation is used to define the redox potential in an aqueous system (Chang et al., 2004):
∅ 𝑅𝑇
𝐸𝑐𝑒𝑙𝑙 = 𝐸𝑐𝑒𝑙𝑙 − ln𝑄𝑟 (2.9)
nF

4
 On the other hand, the actual measurement of the redox potential typically uses an ORP
sensor, which is constructed with a chemically inert electrode (such as platinum, nickel, and gold)
immersed in the solution as a reference. A reference electrode made with silver-silver chloride (a
structure similar to a pH sensor) is connected to saline solution and poses a neutral half cell
potential. A simplified diagram and ORP sensor are present in Figure 2.1.
The redox potential is measured in volt (V) or millivolt (mV), the inert electrode sensing
the dissolving electrode ion and returning electrode ion to define the potential and compare the
difference between a solution with a stable reference electrode to generate redox reading. Because
platinum is highly conductive yet has low chemical and biological reactivities, it is an ideal
material for the redox probe. However, any impurity or (in)organic buildup on the probe will lead
to a false reading. Thus, frequent maintenance, recalibration, and replacement of redox sensor are
required for long-term measurement stability when the redox monitoring project goes online
(Suslow, 2004).

Figure 2.1 Diagram of a laboratorial ORP sensor.

2.1.2 Fermentation redox potnetial
Redox reaction is a fundamental type of chemical reaction in biological systems (from cells
to flora and fauna). Living organisms obtain energy for maintenance, growth, reproduction motility,

5
and other functions from the redox reaction (Shapiro, 1972). Redox reaction involves both energy
metabolism and the biological redox process, which take place simultaneously. The significance
of redox potential for cellular metabolism study was predicted by Joseph Needham in 1926, and
great research interests have been drawn to the scientific community since then (Cohen et al., 1928;
Needham & Needham, 1926).
The shift of redox potential relates to the electron movement in the biological system, and
the role of electron transfer has been investigated and established (Lande et al., 2012). However,
those studies only focused on the underlying biochemical mechanism of electron transfer, and a
standard protocol to observe the electron transfer under experimental conditions is still under
investigated. There are several difficulties in measuring electron transfer in the biological systems,
which include: (1) reactions are mostly not self-exchange; (2) structural information is often
obscured; (3) environments are often inhomogeneous; (4) standard free energy remains undefined;
(5) reactions might occur across membranes; and (6) reations might involve protein conformation;
(7) a fixed reaction site is absent (Marcus & Sutin, 1985).
On the biochemical level, energy storage/release is one of the essential examples of redox
electron transfer. Adenosine triphosphate (ATP) synthesis relies on the degradation of organic
matter via oxidative phosphorylation. The pathways of ATP synthesis impact a full range of
biomolecules, including carbohydrates, amino aicds, fatty acids, and nucleic acids. A balance must
be maintained between oxidizing and reducing agents; otherwise adversely biological effects will
arise (Foyer & Noctor, 2005). For example, an unbalanced redox state leads to reactive oxygen,
sulfur, and nitrogen species, which can impair protein translation, modification, and cellular
signaling (Jones & Sies, 2015). Redox homeostasis is fundamental for maintaining regular cellular
function, and redox potential offer an essential window for perceiving and evaluating the state of
a biological system. Owing to the complexity of biological redox system, there is no direct
approach for sensing the intercellular redox potential. Instead, the redox state may be summarized
by the presence of a redox couple within the system (Schafer & Buettner, 2001).
There are three prominent redox couples that can be used as the indicators for defining the
intercellular redox potential, which are NAD+/NADH, NADP+/NADPH, and GSSG/GSH.
NAD+/NADH and NADP+/NADPH are used to estimate the reduction potential, and GSSG/GSH
facilitates the oxidation potential evaluation. (Wang et al., 2018). In these redox couples,
NAD+/NADH is the most abundant in the intercellular environment. In the sense of bioenergetic

6
metabolism, the cycle between NAD+ and its reduced form, NADH, contributes a reducing
potential to the mitochondrial electron transport, fuelling phosphorylation in ATP synthesis. The
NAD+/NADH cycle also plays roles in the nucleus and cytosol as well as extracellularly. Cantó et
al. (2015) concluded that the NAD+ and NADH equilibrium played an important role in regulating
cellular flux rate and direction. The concentration of NAD+ and NADH have typically maintained
between 0.2 to 0.5 mM, but significant shifts might occur under environmental change and energy
stress (Fig 2.2). This discovery showed the potential for scientists to control cellular metabolism
by altering the intercellular and intracellular redox environments.

.
Figure 2.2 NAD+/NADH cross compartment metabolism (Cantó et al., 2015)

2.1.3 Curreent redox potential application in fermentation

Traditionally, fermentation refers to cellular respiration without oxygen as the terminal
electron acceptor. However, with the improvement of biotechnology, the definition of
fermentation has been broadened to include the production of biomolecules such as amino acids,
proteins and polysaccharides, and substance such as ethanol and organic acids, by microorganisms
under both aerobic and anaerobic environments (María Martínez-Espinosa, 2020). Specifically,
fermentation is a microorganic extraction of energy via the glycolysis pathway, where pyruvate is
produced, followed by reduction to the secondary metabolites. Figure 2.3 shows the electron
movement within the redox pair of NADH and NAD+. This conversion directly alters the

7
intercellular redox potential, affecting the abundances of some important biomolecules such as
ATP, amino acids, fatty acids, and nucleic acids. Since redox potential represents the sum of total
redox reactions undergoing metabolism within the cells, the redox potential is also strongly
affected by environmental factors such as pH, dissolved oxygen levels, and response equilibria
(Ishizaki et al., 1974). Further effects include the exchange of free oxygen and hydrogen through
(semi)permeable membranes due to shuttle-protein activity. Thus, redox potential is manipulable
externally and also provides an alternative approach to understanding and monitoring the
fermentation status.
Elaboration is first on monitoring and then on manipulating: (1) The solution redox
potential measures the concentration of nutrients and reducing agents as well as the medium's
oxygen level. Typically, in batch fermentation, the redox potential profile follows a bathtub-shaped
curve (Liu et al., 2017). As presented in Figure 2.4 (Liu et al., 2017), the microbial population size
remains small during “the lag phase”, but the medium maintains high oxygen and nutrient levels.
The redox potential remains at its high initial point, undergoing a subtle decline. After entering
“the log growth phase,” the microorganism population surges. Nutrients and oxygen deplete with
the rapid increase in cell number, resulting in a dramatic decrease in solution redox potential.
Eventually, due to the depletion of nutrients and a reduction of culture vitality, the older cells
undergo apoptosis, releasing the organic matter back into the environment. At this point, an abrupt
increase of redox potential indicates that the batch fermentation is approaching the endpoint. This
feature, with its rapid response time and high sensitivity to the redox reaction, has become a key
heuristic for online fermentation monitoring. (2) The internal processes can be re-directed by
imposing a redox potential regulation strategy in the external environment, which include gene
expressions and enzyme activities (Allen, 1993) to optimize desired secondary metabolites – thus
achieving a higher input-output ratio for feedstock and reducing total operation time (Liu et al.,
2017).
Sparging is the most common method to control redox potential during fermentation. For
aerobic strains, by using dissolved oxygen acts as a terminate electron acceptor and by controlling
the aeration rate and the air-nitrogen concentration, a target redox potential level can be achieved.
With semi-aerobic or anaerobic strains, the oxygen must be replaced by a different oxidizing agent,
such as potassium ferrocyanide. In both cases, the optimum redox potential level depends on many
factors: specific strain, substrate, and objective metabolites. As examples:(a) Bonan et al. (2020)

8
reported about xylose fermentation with yeast (Spathaspora passalidarum) and found an optimum
yield (28.61g/L) occurs for a redox potential reading of -100mV. (b) A study of ethanol production
in high gravity found that with a highly concentrated 250g/L glucose substrate, a redox potential
of -150mV led to both an optimized yield of 131g/L and assisted yeast survival under the harsh
osmotic pressures. (c) In commercial wine fermentation, Killeen et al. (2018) showed that a redox
potential of 215mV accelerated the fermentation process and maintained yeast cell viability
compared to a control group. More importantly, the redox potential control strategy was shown to
enhance exogenous gene expression during fermentation via redirecting the metabolic flux. (d)
Zhu et al. (2009) utilized a two-stage redox potential control strategy to increase production of 3-
hydroxy propionic acid via a recombinnant Klesiella pneumoniae strain.

Figure 2.3 NAD+/NADH redox couple in pyruvate fermentative degradation

9
 Figure 2.4 Bathtub-shaped curve of ORP profile during fermentation (Liu et al., 2017)
2.1.4 Introduction to microbial fuel cell
Due to the increasing energy demand and concern for long-term energy security, the study
of the microbial fuel cell (MFC) has been flourishing in recent years. This technology integrates
biophysical chemistry and electrochemistry to utilize the biocatalytic capability of the
microorganism to stimulate a bioreactor that spontaneously converts organic matter to electrical
energy through the metabolic activity of microbe (Logan et al., 2006).
Microorganisms, usually bacteria, continually catalyzes organic substrate into the
inorganic matter and renewable electricity in the MFC. The substrate selection varies depending
on the choice of bacteria or bacterial consortium and the growing environment. Furthermore,
microbes could contribute to the bioremediation of industrial waste as well. Multiple studies had
been reported that the MFC device could assist in the bioremediation of pollutants and the
economic feasibility of utilizing microbial fuel cell (Chen et al., 2016; Li et al., 2019; Morris &
Jin, 2008; Pous et al., 2013). With the dual benefits of generating renewable energy and
bioremediation by degrading pollutants, microbial fuel cell seems to be a promising yet
challenging technique for the future.

10
 The following section will discuss the history of MFC technology, the principle,
mechanism, configuration of different types of primary microbial fuel cells, and the applications
of this technique.
2.1.4.1 Development of microbial fuel cell technology
In 1791, the Italian physician and physicist Luigi Galvani birthed the field of bioelectricity
when he detected electricity in frog muscle tissue (Park et al., 2013). After a century, Potter (1912)
constructed the first microbial fuel cell when he demonstrated that electrical energy could be
liberated from organic compounds by a culture of Escherichia coli and Saccharomyces spp.
Inspired by Potter, Cohen (1931) connected multiple MFCs in series, producing over 35 volts of
electricity.
Davis and Yarbrough (1962) utilized hydrocarbon as the carbon source for the MFC to
generate electricity. Although no electrical output was observed, this is the first attempt to apply
MFCs to the problems of biodegradation and bioremediation. In the same year, Rohrback et al.
(1962) successfully generated hydrogen by Clostridium butyricum culture within a glucose oxidase
electrode system.
In the 1980s, it was understood that a critical limitation of microbial fuel cells lies in
electron transport efficiency. Peptidoglycans and lipopolysaccharides in the bacterial membrane
greatly impeded the movement of electrons was recognized (Du et al., 2007). Several efforts were
made to chemically mediate the electron transfer (Weibel & Dodge, 1975; Yaropolov et al., 1976).
In 1993, a functioning mediated MFC that could transport electrons without an endogenous
electron shuttle was developed (Allen & Bennetto, 1993). Those improvements yielded a
significant improvement in power output and spurred widespread interest in MFCs worldwide.
Unfortunately, applications were limited because the necessary mediator concentration was
poisonous to the microorganism (Logan et al., 2006). The development of the mediator-less MFC
sparked a revolution.
In exploring how electron transfer within an MFC occurs, researchers discovered certain
species of bacteria capable of transferring electrons directly – without a mediator. Kim et al. (1999)
reported the first self-electron transfer bacterium Shewanella putrifaciens. Chaudhuri and Lovely
(2003) confirmed their finding by demonstrating that Rhodoferax ferrireducens can oxidize
glucose to carbon dioxide by self-shuttling the electron to the electrode without the assistance of
any mediator. Gorby et al. (2006) reported that a special nanowire is produced by a mutated strain

11
of S. oneidensis, which enables the direct electron transfer in bacteria. These findings led to many
practical applications of the mediator-less MFC such as: long-term and remote energy-generation
(deep-sea sensors, military and robotic devices) and in wastewater treatment (Singh & Yakhmi,
2014).
2.1.4.2 Type of microbial fuel cell
MFC architecture can vary by size, design, and material yet broadly fall into two classes:
single- and double-chambered configurations. As illustrated in Figure 2.5, the double-chambered
MFC consists of anodic and cathodic chambers divided by a proton exchange membrane (PEM),
which permits proton-crossover from the anodic chamber yet blocks oxygen diffusion from the
cathodic compartment. The electron provided by the anode combines with the proton as an electron
acceptor at the cathode to complete an external circuit. Double-chambered MFCs often operate in
a batch mode with the use of defined medium (Logan et al., 2006).
The “H-shape” is one of the most commonly double-chambered MFC configurations for
research as the design can be easily modified for different objectives by altering the electrode,
chamber size, material, and/or half-cell condition. Sangeetha & Muthukumar (2012) reported that
closely spaced electrodes yield elevated power output and chemical oxygen demand (COD)
removal when using wastewater as a substrate. By contrast, the wide electrode separation of H-
configured MFCs results in a greater internal resistance, leading to poor power output (Du et al.,
2007).
Miniature MFCs are another double-chambered configuration that has intrinsic advantages
for research. Ringeisen et al. (2006) reported that a 1.2 cm3 working volume of MFC is capable of
producing 10 mW/m2 of maximum power density, which is sufficient for autonomous sensors
located in remote areas. Other advantages of miniature MFCs include: a short startup time, more
negligible material consumption, and higher throughput of power. (Jiang et al., 2015). Such
features enable MFC utilization in the technical area, such as the remote robot, submarine camera,
or self-sustainable energy devices.
Up-flow MFCs evolved from an up-flow anaerobic sludge blanket reactor design into a
two-chamber configuration, yielding a maximum power output of 170 mW/m2(He et al., 2005).
This device received a lot of attention because it can be easily scaled to treat large volumes of
wastewater (Du et al., 2007). A modified version with a U-shape PEM, featuring a larger surface
area while decreasing the electrode separation, and achieved a significant power boost of 29.2

12
W/m2 as the internal resistance declined from 84 Ω to 17Ω (2006). Note that the up-flow MFC
design is not viable as a bio-generator because the recirculation pumps consume more power than
the device generates (Du et al., 2007).
In single-chamber MFCs, the cathode is either outside the fuel cell cavity or inside without
using a PEM (Logan et al., 2006). Eliminating PEM, which costs roughly US$1400/m2 (Nafion®),
removes a major obstacle for the large-scale commercialization of MFCs. Additionally, Oliveira
et al. (2013) suggested that during continuous operation, the presence of a PEM can cause
acidification of the anode compartment leading to degraded performance. Using wastewater as a
substrate, Liu and Logan (2004) tested these ideas and achieved similar efficiencies with and
without a PEM. However, with glucose as a substrate, the PEM-equipped designs were much more
efficient (40-55% vs 9-12%).
Follow-on studies investigated novel and low-cost substitutes for the PEM. Li et al. (2011)
classified separator technology into ion-exchange membranes, size-selective separators, and salt
bridges. Chang et al. (2017) produced 38 mW /m2 electricity using a low-cost PVA separator. Even
though single-chamber MFCs typically generate lower energy output, they remain preferable for
real-world applications because their simplified design is cost-effective, scalable, and lowering
operational difficulty.

Figure 2.2.5 Schematic diagram of a double-chambered microbial fuel cell

13
 .

2.1.5 Fundamental mechanism of MFC

The configurations of the MFC are varied significantly depending on the size, material
selection, and applications. A typical MFC consists of an anodic and a cathodic chamber separated
by a half-cell separator. Figure 2.5 provides a schematic view of one of the most widely used dual
chamber MFC configurations. The substrate is oxidized by microorganisms anaerobically in the
anodic chamber. The proton passes freely through a half-cell separator, inhibiting oxygen and ions
crossover through a proton gradient.
The microbe liberates electrons from the organic substrate at the anode. The generated
electrons are transferred through an external circuit, subsequently consumed by the cathodic
compartment. The electrical potential is brought about by the bacterial activity in the anodic
chamber and the redox reaction in the cathodic chamber. In an anodic half-cell, the organic
compound is exoelectrogenically, which is to say, the electrochemically active microorganism is
capable of transferring the electron extracellularly (Logan, 2009). For a long time, this was thought
to be impossible. To illustrate this, the reaction which occurs in the anodic compartment with a
glucose substrate can be represented as below (Mahadevan et al., 2014):
𝐶6 𝐻12 𝑂6 + 6𝐻2 𝑂 → 6𝐶𝑂2 + 24𝐻 + + 24𝑒 − (2.10)
Here, glucose catabolism releases 24 protons and 24 electrons into the extracellular environment.
The bacterium arrives at the anode and donates an electron which travels through the external
circuit to the cathodic chamber. Meanwhile, the proton will diffuse through the half-cell separator
into the cathodic chamber where oxygen typically is used as the terminal electron acceptor due to
its high redox potential (i.e., “+1.229 V”), limitless availability, and low cost. (Song et al., in press).
The proton, electron, and oxygen react at the cathodic chamber to form water in an oxygen
reduction reaction, which can be represented by:
6𝑂2 + 24𝐻 + + 24𝑒 − → 12𝐻2 𝑂 (2.11)
2.1.6 Electron transfer mechanism in MFC
The microorganism continuously undergoes substrate catabolism in MFC, liberating
electrons and protons into the anodic compartment. The selectively permeable half-cell separator
allows proton crossover to the cathodic chamber through proton gradient. The non-conductive
structure of most microorganisms cannot transfer those electrons to the outside of the cell and

14
donate them to the anode. But electrochemically active bacteria or exoelectrogen are adapted to
directly transfer an intracellular electron to the extracellular environment without the assistance of
a mediator (Logan, 2009). This microorganism either developed a modified structure or
synthesized particular molecules that aid electrons to relocate outside the cell membrane.
On the other hand, certain redox-active species could facilitate electron transfer outside of
the cellular membrane for the electrochemically inactive bacteria. Such redox-active species refer
to the mediator. Schroder (2007) defined the mediator as a species of chemical that must be soluble,
has physical contact with the electrode, and possess a low reduction potential. Its potential should
come as close as possible to the organic substrate in the anodic chamber as well. Regardless of the
bacterial type employed in the MFC, there are only a few electron transfer methods. Those methods
can be classified into direct electron transfer and mediated electron transfer (Schroder, 2007).
2.1.6.1 Direct electron transfer
Through a process called extracellular respiration, exoelectrogenic bacteria are capable of
transferring electrons to the extracellular medium. C-type cytochromes are one of the redox
proteins that facilitate this transfer, typically found in metal-reducing bacteria such as Geobacter
and Shewanella (Shi et al., 2007 In brief, the cytochrome proteins utilize a cascade of membrane-
bound complexes, with a solid terminal electron acceptor, in this case, is Fe (III) or Mn (III, IV).
This direct electron transfer results in forming a monolayer of cell membrane or organelle on the
electrode surface. The specific mechanisms behind these processes are not well understood. One
recent study suggests that the electricity generated in MFC can be elevated by overexpressing the
cymA gene, which facilitates cytochromes translation in S. oneidensis MR‐1 (Vellingiri et al.,
2019).
A major bottleneck in redox protein electron transfer is the total number of bacteria in the
monolayer to contact the electrode (Torres et al., 2010). Some bacteria have evolved a conductive
pilus, also known as a nanowire, to aid in extracellular respiration (Figure 2.6). Reguera et al.
(2005) observed extracellular electron transfer in the C-type cytochrome-deficient species G.
sulfurreducens via this faculty. One year later, another metal-oxidizing bacterium S. oneidensis
had been reported to produce a nanowire as well (Yuri A. Gorby et al., 2006). The physiology of
the microbial nanowire remains not fully understood. It is believed that the subunit-protein PilA,
consisting of five aromatic amino acids, enables/controls electron transfer through the nanowire
(Vargas et al., 2013).

15
 Importantly, this nanowire structure significantly improves the overall efficiency of charge
transfer to the electrode (Schroder, 2007). It acts as a conduit that allows a significantly increased
number of bacterial cells to contact the electrode's surface area, leading to a thicker bacterial
biofilm. Malvankar et al. (2011) calculated the conductivity of the nanowire to be:
2𝑎
𝜎 = 𝐺(𝑔𝐿) (2.12)

where G is the bacterial biofilm conductance, 2a is gap width, L is the length of the electrode, and
g is the thickness of biofilm; and they present a method for measuring G and g.
Extensive studies have been focused on two nanowire-producing bacteria G.
sulfurreducens and S. oneidensis (Creasey et al., 2018). Beyond these, only a limited number of
microbes capable of producing nanowires have been discovered. A broader search for nanowire-
producing microbes is required to elucidate the genetic background and deepen our understanding
of the electro-physiology of nanowires.

Figure 2.6 Direct electron transfer via membrane bound protein and via conductive nanowire
structure

2.1.6.2 Mediated electron transfer

Most bacteria do not produce electrochemically active proteins or nanowires, in which case
a mediator (or redox agents) is needed to facilitate electron transportation. This electrode-
contactless method can achieve superior performance on substrate removal while effectively
generating electricity (Uria et al., 2017). Two types present themselves: exogenous-mediated and
metabolite-mediated (Schroder, 2007).
An exogenous mediator completes the external circuit between electrode and bacteria.
Oxygen would be a superior mediator (because of its superior electronegativity), but the anaerobic

16
bacteria require a low-oxygen solution. A good mediator must be highly permeable to the bacterial
cell; it must have a favorable overpotential to drive oxidization at the electrode; and ditto in its
reduced form to lure electrons from the bacterial cell (Mahadevan et al., 2014). Specifically, the
oxidized mediator must penetrate the cell, procure an electron, migrate back across the membrane,
and give that electron to either the electrode or a terminal electrode acceptor. A schematic of
exogenous-mediated electron transfer is provided in Figure 2.7.
Several exogenous redox mediator compounds have been investigated, including dyes such
as methylene blue, natural red, resorufin, and thionine – as well as color-shifting inorganic
compounds such as potassium ferricyanide (Emde et al., 1989; Rahimnejad et al., 2015).
While multiple studies have demonstrated the efficacy of redox mediators (Evelyn et al.,
2014; Lohar et al., 2015), there are also intrinsic limitations posed by this method. For example,
the requirement for continuous addition of mediator, the inability to penetrate thick biofilms, net
economic inefficiency, and potential toxicity. In sum, exogenous mediation seems impractical for
commercial and large-scale applications (Mahadevan et al., 2014).
Some microbes, however, have adapted to produce their own endogenous mediators by
employing primary and secondary metabolites low-molecular shuttles. One example is
Pseudomonas spp., which excretes pyocyanin to serve as a shuttle/mediator (Rabaey et al., 2004),
though at the cost of some ATP (Lovley, 2006).
Self-mediating bacteria present a competitive advantage under batch conditions in which
the substrate is continually replaced. In such a system, a steady replacement of mediators is
necessary. Sulfate/sulfide cycling is quite suitable for electron self-mediation, yielding a standard
biological potential of -0.22V (Schroder, 2007). During respiration, the sulfate reduces sulfide;
upon contact with the electrode, the sulfide oxidized to complete the cycle. Primary metabolites
derived from fermentation and anaerobic respiration could assist this self-mediating process. An
energy-rich reduced substrate (such as hydrogen and ammonia) is directly used as a mediator in
fermentation. The energy-rich substance is oxidized by bacteria with the aid of the electrocatalytic
anode – thereby avoiding the interruption of other biological processes (Schroder, 2007).

17
 Figure 2.7 Illustration of mediated electron transfer within a microbial cell
2.1.7 MFC performance and optimization
Recently, intensive research has been focused on optimizing power generation through a
multidisciplinary approach to MFC devices.
Carbon-based electrodes have drawn lots of attention because of their excellent
biocompatibility, conductivity, and competitiveness. Surface area and porosity are essential to
reduce internal resistance and ohmic losses (Kalathil et al., 2018). More recently, nanotechnology
has led to the emergence of electrodes based on carbon nanotubes, which present a 3D
nanostructure and large porosity and generate superior MFC (Delord et al., 2017).
Energy losses within the MFC are primarily due to three mechanisms: activation
polarization (𝜂𝑎𝑐𝑡 ), ohmic processes (𝜂𝑜ℎ𝑚 ), and concentration polarization (𝜂𝑐𝑜𝑛𝑐 ) (Rismani-
Yazdi et al., 2008). Activation loss dominates in the low current densities, where the initiation of
electron transport must overcome the energy barrier of the reacting species. These losses may be
mitigated by increasing the anodic surface area, introducing redox mediators, and increasing the
operating temperature. (Mahadevan et al., 2014). Ohmic losses are caused by the MFC’s internal
resistance and can be easily calculated by ohm’s laws. These losses can be reduced by shortening
the distance between electrodes and employing better conductive material (Mahadevan et al.,
2014). At high current densities, the incapability of maintaining the mass transfer rate of
substrate/product results in an increase/decrease in the anodic/cathodic potentials. Concentration

18
polarizations occur when high current density and cell potential drop depart from the linear
relationship with the current density profile. A longer growth time of biofilm on the anode could
result in higher limiting current and fewer polarization losses when other condition remains the
same (Zhao et al., 2014)
Collectively, the energy output of an MFC can be predicted by (Rismani-Yazdi et al., 2008):
𝑉𝑜𝑝 = 𝐸𝑐𝑒𝑙𝑙 − (𝜂𝑎𝑐𝑡 + 𝜂𝑜ℎ𝑚 + 𝜂𝑐𝑜𝑛𝑐 ) (2.13)
2.1.8 Cell potential modeling -The Nernst equation
Microbial fuel cell models have been studied by Zhang and Halme (1995): an
oversimplified model will lead to detail loss from MFC, and an overcomplex model will lead to
higher difficulty in data analysis. Considerations must be made for biological, electrochemical,
physical, and diffusive processes. Nonetheless, the redox reaction is considered as first-order
reaction in the modeling by Zhang and Halme (1995):
𝐸𝑐𝑒𝑙𝑙 = 𝐸𝑐𝑎𝑡ℎ𝑜𝑑𝑒 − 𝐸𝑎𝑛𝑜𝑑𝑒 (2.14)
The potential of the redox reaction of bacteria degrading the substrate in an anodic chamber is
equal to the standard potential between two half-reactions. The standard cell potential can be
calculated from the Gibbs free energy:
ΔG
𝐸0 = −𝑛𝐹 (2.15)

Where ΔG is the Gibbs free energy, n is the number of electrons transferred during the reaction,
and F is the faraday constant.
However, the cell's overall potential depends on the concentration of reducing/oxidizing
agents and environmental factors. The Nernst equation accounts for these effects (Feiner &
McEvoy, 1994):
∅ 𝑅𝑇
𝐸𝑐𝑒𝑙𝑙 = 𝐸𝑐𝑒𝑙𝑙 − ln𝑄𝑟 (2.16).
nF

Where R is the universal gas constant, T is the temperature in Kelvin, n is the mole of electrons,
and Qr is the reaction quotient, defined as the relative amounts of product and reactant present
during the reaction at particular time. For example, for the reaction:
𝑎𝐴 + 𝑏𝐵 → 𝑐𝐶 + 𝑑𝐷 (2.17)
The reaction quotient, Qr, is defined as:
[𝐶]𝑐 [𝐷]𝑑
𝑄𝑟 = [𝐴]𝑎[𝐵]𝑏 (2.18)

19
Additionally, other biological processes, including complex enzyme reactions and microbial
growth in the fuel cell, are highly dependent on substrate availability. Such process can be modeled
by Monod-type kinetics (Kovárová-Kovar & Egli, 1998):
[𝑆]
𝜇 = 𝜇𝑚𝑎𝑥 𝐾 +[𝑆] (2.19)
𝑠

Where μ is the growth rate, S is the concentration of substrate, and 𝐾𝑠 is the half-velocity constant.
Two diffusive processes occur in the fuel cell: hydrogen ions diffuse through PEM into the
cathodic chamber and external oxygen from diffusing/dissolving into the catholyte. Those
processes can be modeled by Ficks’ second law of diffusion (Paul et al., 2014). However, Zhang
and Halme (1995) demonstrated that the diffusion process is fast enough to affect neither cell
potential nor power generation and can thus be neglected under normal circumstances.
2.1.9 Electrochemical biosensing and MFC-based biosensor
Sensors in modern industrial applications cover the entire spectrum of input parameters
converted to an electrical signal. In recent years, the strong demand for real-time, on-site, simple,
cost-effective designs has led to the use of biosensors in fields as diverse as pharmaceuticals,
medical applications, foods, and the environment.
The majority of biosensors operate either enzymatically or electrochemically. Enzymatic
biosensors have the advantages of specificity and high sensitivity yet suffer from complex
preparation protocols and a short lifetime (Mulchandani, 1998). On the other hand, an MFC-based
biosensor is an electrochemical approach. It is a versatile tool, sensitive to a large variety of
analytes, and offers a simplified preparation protocol, sustainable and cost-effective at scale. As
such, it is a viable future alternative for enzymatic biosensors (Su et al., 2011).
An MFC device generates an electrical signal that effectively monitors microorganisms'
metabolic activity kept under steady-state conditions. The microorganism within the anodic
compartment acts as the recognition agent that responds to any alteration (pH, solution
concentration, specific compound) in the environment, affects the flow rate of electrons of the
external circuit, and transduces into a measurable voltage signal. Thus, any environment
disturbance results in a change in voltage/ampere, making the MFC an in-situ biosensor (Cui et
al., 2019). Any fluctuations are securely coupled to the redox reactions within the MFC – an
advantage not available to other types of amperometric biosensors (Lovley, 2008).
In an unsaturated anodic solution, the change of concentration of biodegradable substrate
will directly affect the metabolic rate of microorganisms and, consequently, the flow rate of the

20
electrons and the signal output. This approach is mainly used for in-situ environment monitoring
for labile carbon such as BOD, COD, or toxic carbon monoxide (Peixoto et al., 2011; Zhou et al.,
2018). On the contrary, in a saturated and steady anodic environment with constant external factors,
a sudden change of the MFC biosensor's signal output may be caused by the emergence of a new
bioactive compound within the feed stream. This approach is mostly used to detect toxic
components in water (Chouler & Di Lorenzo, 2015), as the inhibition of microbial activity reduces
the amount of electricity generated. This technique is favored by the wastewater treatment industry
as it can detect and prevent four major types of toxicant discharge, including heavy metals,
antibodies, and organic and acidic toxicants (Cui et al., 2019). It is important to note that the
toxicant concentration may exceed the detection range of MFC biosensor; therefore, the sensitivity
of the microbe(s) vis-à-vis the inhibition effect of the toxin are important pre-considerations.
Sensitivity is defined as the electrical signal change per unit change of concentration of target
analyte:
𝛥𝐼
𝑆𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦 = (2.20)
𝛥𝑐∗𝐴

where ΔI is the change in current output, Δc is the change in the target analyte concentration, and
A is the anode surface
(Di Lorenzo et al., 2014). The inhibition effect can be calculated by modified Michaelis-Menten
Equation:
|𝐶𝑌𝑛𝑜𝑟 −𝐶𝑌𝑡𝑜𝑥 |
𝐼(%) = ∗ 100 (2.21)
𝐶𝑌

Where CY is the coulombic yield and is obtained by integrating the electrical signal generation
over time, and CYnor and CYtox are the coulombic yields in normal and toxic conditions.
Furthermore, the MFC-based biosensor could be used as a monitor for microbial dynamics.
Liu et al. (2011) utilized an MFC-based biosensor to monitor the anaerobic digestion of a microbial
consortium in a wall-jet MFC with varied feedstock concentration and process configuration. The
MFC biosensor signal presents a clear correlation between signal and various parameters. A recent
study shows that by incorporating high-throughput Illumina® sequencing technology with the
voltage generated by an MFC bioreactor one could successfully monitor the microbial community
growth dynamics (Pepè Sciarria et al., 2019).
So long as the details of the bacterial mechanisms remain elusive, most research continues
to focus on bacteria-consortia of the naturally occurring environment or of industrial wastewater

21
(Vejarano et al., 2019). So far, real-world implementations of MFC biosensors have been limited
to these consortia or to single microbial strains.
2.2 Knowledge gap
Historically, the fermentation process can be monitored by microscopical presence,
counting microbial population, or measuring the level of a key metabolite in the fermenter. Such
monitoring techniques can easily be accomplished without specialized skills and equipment. In
recent decades, the rapid development of bioelectrochemical technology and the emergence of
digital sensors enable a more effective and precise fermentation monitoring process. ORP
measurement by utilizing the ORP sensor is one of the most popular fermentation monitoring and
control approaches. Such a method provides an online and consistent signal during fermentation.
It also can be used in any stage of operation, providi ng high signal integrity and measurement
reliability. However, the high fabrication cost of an ORP sensor and instrumentation makes it
impractical for large-scale applications in industrial fermentation. Cost-effectiveness is an endless
effort in engineering study, and an economically optimized ORP monitor tool is urgently needed.
MFC is an electrochemical system. It utilizes an active microorganism as the biocatalyst
for bioelectricity production. Besides power generation, it can be used as a tool as a biosensor.
Most MFC-based biosensor research focuses on an environmental study, such as detecting the
substrate concentration or the presence of the substance of interest in the environment.
Furthermore, the Nernst equation provides a relation to associate the signal output from an MFC
device with the level of redox potential from microbial activities. Valiadation of such correlation
derived from Nernest equation under actual MFC conditions is nececarry for the development of
fermentation process equipped with MFC-based biosensors.
.
2.3 Hypotheses and objectives
The following hypotheses will be tested as a part of this research:
• The microbial growth stage is associated with the signal output from an MFC device.
• The voltage output presents a correlation with ORP level within an MFC device.
• The correlation between voltage output from an MFC device and redox potential level from
an ORP sensor could be used to monitor the progress of a fermentation process.

22
 The purpose of this study is to establish a cost-effective approach for a large-scale fermentation
process that is capable of real-time monitoring strategy by utilizing an MFC device as an ORP
sensor. The specific objectives include:
• To correlate a relationship among voltage output, redox potential within MFC and
microorganism growth stage.
• To develop a new approach for fermentation monitoring by biosensing technology.

23
 3 MATERIALS AND EXPERIMENTAL METHODS

3.1 Strain and cultivation

Bacillus. subtilis is a commercial bacterial strain isolated in Taiwan (Yanten Biotechnology
Co., Ltd.). Our growth medium was a modified Lysogeny broth (LB), consisting of peptone 10g/L,
NaCl 10g/L, yeast extract 5g/L, MgSO4 2g/L, and 1M NaOH 1 mL/L. The medium was sterilized
at 121°C for 50 minutes prior to use. For long-term strain preservation, the modified Lysogeny
broth was mixed with 20% glycerol and stored in a -80°C freezer.
Frozen seed culture of B. subtilis from the -80°C freezer was thawed and inoculated into
50 mL of modified LB broth with the addition of 15g/L of glucose. After incubation on the shaker
bed (180rpm) at 25°C for 12 h, the regenerated culture was inoculated into a sterilized 250 mL
beaker filled with 230 mL modified LB broth and incubated on the shaker bed at (180rpm) at 25°C
for another 24 h.
Pseudomonas fluorescens local strain, ICW1057, was isolated from an oil-impacted field
from Saskatchewan, Canada in our lab (Huang & Lin, 2020). This strain will be used as a
comparator to B. subtilis in order to evaluate their respective performance in MFC device. A
modified minimal salt medium is used for bacterial replication and incubation. The minimal salt
medium is sterilized at 121°C for 50 minutes prior to use. All chemicals used in strain cultivation
and replication were obtained from VWR International (Edmonton, AB, Canada) and Fisher
Scientific (Ottawa, ON, Canada).
3.2 Microbial fuel cell-based biosensor design and construction
The MFC-based biosensor is designed based on a double-chambered MFC device, which
is most easily controlled and modified in the laboratory environment. A proton exchange
membrane located at the middle dividing the device into anodic and cathodic compartments. The
microbial fermentation process is carried out at the anodic chamber, and the reduction process is
conducted at the cathodic chamber. Two low-cost graphite rods are employed as electrodes and
inserted into each respective compartment. The anodic chamber is equipped with an ORP probe
for environmental redox potential detection. Two air pumps connect to each compartment to
control the aeration rate and oxygen level. A pH regulation system is also connected to the anodic
chamber to prevent fluctuation of redox potential triggered by pH change. The anode and cathode
are connected to an external circuitry for voltage signal detection and data collection. The

24
schematic of the experimental setup is illustrated in Figure 3.1., and specific construction materials
and the experimental procedure are followed:
A standard H-shaped, two-chamber MFC with a 250 mL reactor (Wenoote Inc, China) and
a 5cm-diameter PEM (Nafion 117, DuPont, USA) was used. The PEM was subjected to a three-
stage pre-treatment (Ghasemi et al., 2013): first, immersed in 80°C water for 30 mins; second,
immersed in an 80°C solution of 3% hydrogen peroxide; and finally immersed in 0.5M sulfuric
acid. Subsequently, the PEM was rinsed with deionized water three times and stored in 0.5M
hydrochloric acid. Both electrodes were graphite rods, 6mm in diameter and 100mm in length. The
projection area of each rod was approx. 1725mm2 (90mm length and bottom surface submerged).
Rods were pre-treated by baked at 300°C for 20 minutes to remove impurities, soaked with acetone
and methanol successively, and finally washed with deionized water three times.
The anodic chamber was inoculated with 24 mL of the B. subtilis and P. fluorescens (O.D.
adjusted to 2.0-2.4), along with 226 mL of modified LB used as substrate (adjusted initial O.D. of
anolyte between 0.2-0.24). The pH of the substrate was adjusted to 8 with 1M NaOH or HCl
solutions, and a peristaltic pump was connected to the anodic chamber to maintain this pH
throughout the operation. The cathodic chamber was filled with 240 mL distilled and deionized
water. An air pump (Tetra Inc., China) was connected to each chamber for sparging. A gas
flowmeter (SHO-RATE, Brook instrument CO., USA) was maintained an aeration rate of 0,11.32,
and 22.64 vvm in the anodic chamber, while flowrate through the cathodic chamber was kept over
10 gallons/min without control. A bottle of deionized water and air filters were connected between
the air pumps and each chamber to provide humid and sterile sparging condition in order to prevent
electrolyte evaporation. Additionally, a disk-shaped stir disk was placed in the anodic chamber,
and the MFC was placed on a stir plate (110rpm) to allow air to dissolve into the anolyte evenly.
An external 324 kΩ resistor was connected across the anode and cathode with titanium wire. Both
ORP and pH sensors (Van London Co., USA) were inserted into the anodic chamber for data
acquisition. All components of this apparatus were sterilized in an autoclave at 121°C for 50
minutes before use.
Additionally, the effect of the supplement of mediator and catholyte had been tested in this
set of experiments, where 1% 1M of methyl blue was added into the anodic chamber to facilitate
the electron transfer. Also, 1% 1M potassium ferricyanide was added into the cathodic chamber as
an alternative for oxygen as terminal electron acceptor.

25
 Figure 3.1 The MFC apparatus diagram
3.3 Analytical procedures
The electrical voltage across the resistor was measured by the Phidgets VINT hub (Phidgets
Inc, Canada) and recorded using a computer-based data acquisition system (National Instruments,
LabVIEW, version 19.0f2) at 10s intervals. The current was calculated by Ohm’s law: I=V/R.
The redox potential measurement used an ORP sensor connected to a
pH/Temperature/mV/ISE multi-meter (VWR Scientific, USA). The data was recorded by the
computer-based data acquisition system mentioned above at 10s intervals.
Biomass concentration was determined based on the optical density (OD600). The bacterial
suspension within the anodic chamber was sampled every hour after the 5-hr lag period. Each
sample was withdrawn with a 5-mL sterile medical syringe and tested for optical density and
biomass change by a benchtop spectrophotometer (UVmini-1240, Shimadzu Inc, Japan) of 600nm
wavelength.

26
 4 RESULTS AND DISCUSSION

The motivation of this research was to prove the concept of an alternative ORP
measurement method and to design a prototype of an actual MFC-based fermentation sensor.
Different factors, including microbial types, fermentation conditions, and environmental factors,
were considered in this work. The results were broken down into a few parts, including testing two
types of microorganisms with or without the supply of artificial electron mediator, using different
catholyte and catholyte treatments, applying different anodic chamber aeration rates, and analyzing
the ORP change, voltage output, and (unitless) potential parameter that we derived from the Nernst
equation, respectively.
4.1 Modification of the Nernst equation for MFC application
The Nernst equation is well established and had been proven by a large amount of
experimental evidence. This equation can be derived directly from basic thermodynamic principle
and applied to two major cases, including the equilibrium potential of an electrode (redox electrode
or metal ion electrode) in a Galvanic cell and the potential (voltage) of a cell reaction of an
electrochemical cell (Scholz, 2017). In this experiment, the redox potential and voltage of the
microbial fuel cell are both subjected to the Nernst law. The Nernst equation is applied to our work
and can be described as in Equation 3.1:
𝑹𝑻
𝑬𝒄𝒆𝒍𝒍 = 𝑬∅𝒄𝒆𝒍𝒍 − ln𝑄𝑟 (3.1)
nF
∅
Where the 𝐸𝑐𝑒𝑙𝑙 is the actual cell potential, 𝐸𝑐𝑒𝑙𝑙 is the theoretical cell potential, R is ideal gas
constant, T is temperature, n is the number of electrons, F is faraday constant, and 𝑄𝑟 is reaction
quotient that represents stoichiometric measurement of relative amounts of oxidant and reductant
of the redox reactions in the MFC devices.
∅
By moving the 𝐸𝑐𝑒𝑙𝑙 to the left side of the equation, the change of the cell potentials within
the MFC could can be calculated by:
𝑹𝑻
𝑬𝒄𝒆𝒍𝒍 − 𝑬∅𝒄𝒆𝒍𝒍 = − nF ln𝑄𝑟 (3.2)

27
 ∅
Where the 𝐸𝑐𝑒𝑙𝑙 − 𝐸𝑐𝑒𝑙𝑙 , is change of the initial cell potential and in the MFC-based fermentation
reaction, and the redox potential is a quantitative indicator of the degree of completion of the
∅
biochemical reactions. In this case, the difference between 𝐸𝑐𝑒𝑙𝑙 and 𝐸𝑐𝑒𝑙𝑙 can be consider as the
change of redox potential in the MFC:
𝑹𝑻
∆𝑶𝑹𝑷 = − nF 𝑙𝑛𝑄𝑟 (3.3)
𝑅𝑇
Where − nF 𝑙𝑛𝑄𝑟 is equal to the change of the redox potential in the MFC device, Qr is the reaction

quotient that represents the stoichiometric balance of oxidants and reductants involved in the MFC-
based fermentation. However, fermentation consists of a series of complicated redox reactions,
and many intermediate products and derivatives are formed. In this case, we could assume that the
oxidant is equal to the reductant with the electron activities. And the ultimate goal of all those
transformations and conversions of chemicals and matters is for microbial cell metabolism and
propagation. Hence, the Qr is a unitless term that reflects matter exchange between a microbial
strain and biomass gain during the course of fermentation and is conclusively called potential
parameter X. As lnQr is a constant, the RT/nF represents an electric voltage measured in an MFC
device and its unit is V or mV. This voltage is generated by the electrons that emit from the redox
reactions during fermentation and is captured by the MFC device. Hence the RT/nF can be
considered as the electron movement between anodic and cathodic compartment, consequently
resulting in the voltage production from the MFC device (∆V):
∆𝑶𝑹𝑷 = −∆𝑽 𝑙𝑛 𝑿 (3.4)
Reorganized the Equation 3.4:
∆𝑶𝑹𝑷
𝑿 = 𝒆− ∆𝑽 (3.5)
Where the ∆𝑂𝑅𝑃 is the change of redox potential from the microbial fermentation in anodic
chamber, ∆𝑉 is the voltage between the anode and cathode. Because parameter X is a constant that
correlated two potentials in the MFC device (ORP and voltage), it is named as potential parameter
X.
Equation 3.5 demonstrates a relationship that could correlate the redox potential reading and
voltage detection from the MFC device. Integrating such a relationship with the bacterial growth
dynamic makes it possible to monitor and control the fermentation process.

28
4.2 Effect of different microorganisms on MFC-based biosensor performance
In this project, pursuing maximum electricity harvesting was not a priority. Biosensor
research emphasized sensitivity, stability, and reliability instead. Due to the complex interactions
in the microbial consortium, this study only examined the performance of the pure culture of
bacteria. Two pure cultures of different bacterial strains, Bacillus. subtilis, and Pseudomonas
fluorescens were the candidates for the MFC-based biosensor, and both performances with
mediator presence or absence were examined in this section. Figure 4.1 showed the plots of voltage
production and power density with the addition of a 0.1M methylene blue mediator. For P.
fluorescens, it was observed that without the supplement of mediator, the voltage and power
density occurred after the 8 h of fermentation and sharply climbed to 0.02V and 700 µW/m2,
respectively. Both values then gradually reached a peak around 30 h, giving out a maximum
voltage production of 0.04 V and maximum power density of 2602 µW/m2. The voltage and power
density started to descend slowly until the end of fermentation. For the methylene blue addition
group, the first signal appeared after 10 h of the experiment, and a rapid increase from 10 h to 15
h was observed. After that, both changes of voltage and power density became stable, and a
comparatively slight increase until 48 h were observed. The maximum voltage was around 0.11V,
and the maximum power density was around 17000 µW/m2. For B. subtilis, without the
supplement of mediator, the first voltage and power density produced at 6 h of fermentation, and
the voltage and power density increased from 0 to 0.16 V and 63000 µW/m2, respectively, from 6
h to 10 h of fermentation. Both values then remained stable until the end of fermentation, which
gave out 0.19 V maximum voltage and 69356 µW/m2 maximum power density. For B. subtilis
with the mediator, the voltage and power density curves presented a similar pattern to the no
mediator counterpart. Noticeably, B. subtilis with mediator generated the voltage and power
density at the 3 h, which was the quickest among all four groups.
Regardless of the addition of the mediator, B. subtilis outcompeted the voltage and power
production from the P. fluorescens. The Pseudomonas family, such as P. aeruginosa and P.
alcaliphila, has been proven to be successful electrogens (Yong et al., 2011); however, the
scientific community hardly investigated the P. fluorescens’ electricity harvesting ability and its
physicochemical properties in MFC. By comparing two plots of P. fluorescens without the aid of
a mediator, the voltage and power density produced from this strain (P. fluorescens) presented
discontinuousness and instability. The acceleration (8-13 h), deceleration of increase (14-32 h),

29
and acutely decreased phase (after 35 h) of the signal can be observed. A possible explanation for
this performance might be that P. fluorescens lacked a shuttle to transfer the electron to the
intercellular environment efficiently. Only a trace number of electrons can be released into the
anodic chamber and detected by the electrode. When the mediator was added, P. fluorescens
produced relatively higher voltage and power density because the methylene blue facilitated the
electron shuttle to complete the half cell reaction.
Moreover, the voltage became stable around 20 h (Figure 4.1a), with average voltage
production below 0.1 V, which presented a weak power generating ability for P. fluorescens. For
B. subtilis, the similar pattern of voltage and power density generation regardless of the addition
of mediator elucidated that B. subtilis was capable of producing sufficient shuttle to transfer the
electron from intracellular to the extracellular environment. Nimje et al. (2009) examined the
performance of B. subtilis in MFC. They proved that B. subtilis, which produced active and soluble
redox compounds, could facilitate the extracellular electron transportation for MFC reactions. The
electrochemical activity of B. subtilis was governed by the excreted metabolites redox compound,
which reflexed the intercellular homeostasis and ORP balance. Furthermore, the endospore
structure helped B. subtilis to overcome the environmental stress and to regulate the cell
germination, which accelerated the cell multiplication phase and shortened the lag phase of
bacterial growth (Mckenney et al., 2013). Additionally, the biofilm-forming adaptation of B.
subtilis allowed fast sensing and constructed a stable communication between bacterial cells and
anode (Ismail & Jaeel, 2013). In the presence of mediator, B. subtilis was able to produce power
at 3 h of operation, contributing to the fast and thriving growth habitat; the bacterial cells reached
enough population and carried out normal catabolism to liberate electrons in the substrate.
Furthermore, the delayed emergence of power when B. subtilis without the addition of mediator
may be caused by the biofilm and redox compound that can not be formed in the early fermentation
stage.
P. fluorescens lacked an endogenous method to shuttle electrons and required an external
mediator for electron transfer. The insufficient or exceeded addition of mediators would lead to
the adverse effect on bacteria themselves and inaccurate MFC sensing response (Mohan et al.,
2007), which caused more serious operational difficulty and economic cost. Furthermore, B.
subtilis presented a significantly shorter period for cell multiplication. The growth-promoting
feature and the fast-building bacterial colony enabled the biofilm formation and robust growth of

30
bacterial cultures. Considering the sensitivity of bacteria, life cycle, and bacterial behavior, B.
subtilis presented superior performance as an MFC-based biosensor experiment candidate.

Figure 4.1 Voltage production and power density of two different bacterial strains over 48 h.
fermentation.
4.3 Effect of catholyte and catholyte treatments on the MFC-based biosensor performance
As an important part of MFC, cathodic compartment was to provide a confined space and
to serve as terminal electron acceptor to complete the fuel cell half-reaction. Oxygen is a ubiquitous
electron acceptor used in the cathodic chamber due to its low cost and high oxidation potential.
Another alternative approach is using a high oxidative chemical compound. Furthermore, oxygen

31
in the air can be directly used by an air cathode, but construction and well-conducted air-cathode
require costly catalysts and material (Kakarla et al., 2015). Based on the scope of the MFC-based
biosensor, this section compared a sparged air cathode and a potassium ferricyanide solution
cathode, and a simple aerated cathode was used as the control group.
In Figure 4.2, the voltage produced from the three groups (sparging treatments, potassium
ferricyanide catholyte, and aerated cathode) has been plotted. The potential generated from aerated
cathode around 6 h after inoculation, and the potassium ferricyanide catholyte voltage developed
at 7-8 h after inoculation. Sparging treated group signal appeared at 9-10 h of operation. All three
groups' voltage increased until around 20 h, and the air cathode voltage production suddenly
dropped to 0 at 22 h. The voltage from the potassium ferricyanide group kept a sharp increase until
24 h. The increment became more gradual around 24-46 h, and a maximum voltage around 0.21
V was observed. The voltage decreased slightly afterward and was stable around 52-53 h, with a
final voltage around 0.2 V at the end of the experiment. The sparging group had significant growth
of voltage production until 20 h, and then the slope became flat from 20-40 h. A peak occurred
around 42 h with 0.14 V maximum voltage generation. Then the voltage started to decrease and
remained stable until the end of fermentation with approximately 0.13V. The air-cathode MFC
had lower internal resistance than MFC with ferricyanide.
Each cathode treatment exhibited rapid start-up, but the potential generated by the aerated
cathode group vanished and could not provide a stable signal over a long period of time. This may
be due to the dissolved oxygen in the catholyte being quickly consumed by electrons and protons
emitted from the anodic compartment. As the oxygen was exhausted in the cathodic chamber, there
was insufficient electron acceptor to the support fuel cell half-reaction. More importantly, the
oxygen dissolution and diffusion rate into the catholyte was much lower than the consumption rate
from the fuel cell. After oxygen depletion, every oxygen molecular was dissolved into a cathodic
chamber, quickly offset electron accumulated in the circuit, which resulted in minor potential
difference to overcome the internal resistance for proton transportation. Eventually, voltage
sharply dropped to and remained at 0 after oxygen depletion.
By comparing the sparging and potassium ferricyanide catholyte, the potassium
ferricyanide exhibited superior overall power and voltage production than the sparging group. The
difference in power generation may be attributed to the different types of electron acceptors.
Lawson et al. (2020) reported that when the electrode resistance was similar for ferricyanide and

32
oxygen as the electron acceptor, the ferricyanide carried a more significantly experimental working
potential. When utilizing oxygen as the electron acceptor, the half-reaction at cathode would be,
and the theoretical potential can be calculated by Nernst equation:
𝑅𝑇 1
𝐸 = 𝐸 ∅ − 4F 𝑙𝑛 [𝑂 + ]4
(3.6)
2 ][𝐻

On the other hand, the half-reaction of potassium ferricyanide would be 𝐹𝑒(𝐶𝑁)3− −

6 + e →

𝐹𝑒(𝐶𝑁)4−
6 , the corresponding potential is:
𝑅𝑇 𝐹𝑒(𝐶𝑁)4−
𝐸 = 𝐸∅ − 𝑙𝑛 𝐹𝑒(𝐶𝑁)63− (3.7)
F 6

The oxygen possed a higher redox potential of 1.23 V, and ferricyanide had a lower redox
potential of 0.36 V. However, when utilizing oxygen as an electron acceptor, the actual cell
potential was exhibited an exponential relationship to oxygen concentration, which consequently
caused the cathode polarization. In contrast, the potassium ferricyanide contributed to the cell
potential by the concentration of ferricyanide. Using the potassium ferricyanide as the electron
acceptor led to a faster reduction rate and higher voltage production when there was much lower
overpotential and higher solubility within potassium ferricyanide (Ucar et al., 2017). Oh and Logan
(2006) reported a similar result, 1.5-1.8 times of increased power production in MFC was observed
when the ferricyanide was used as electron acceptor.
In contrast to the power production-oriented MFC, signal clarity, real-time response, and
stability emphasized the MFC-base sensor. Besides the voltage production, both the sparging
group and potassium ferricyanide group were comparable in producing stable, clear, and similar
voltage signals. Using non-sustainable potassium ferricyanide as a catholyte could increase the
operational cost and difficulty and also could make the MFC biosenosr less productive. The aerated
cathode required minimum attention, but the oxygen diffusion rate in the cathodic chamber is much
lower than the electron production rate in the anodic chamber. After the existing oxygen molecule
is exhausted, the voltage drop to zero. Although the oxygen is continuously diffused into the
catholyte, the trace amount of oxygen is suddenly consumed by the half-reaction, and there is no
detectable voltage signal generated due to there is not enough power density to overcome the
overpotential for activation polarization. All these results suggested that the sparging group could
be the optimal cathodic treatment for this MFC-based biosensor.

33
Figure 4.2 Comparison of voltage signal generation among three different electron acceptor
sources and treatments in cathode.
4.4 Effect of aeration in the anodic chamber on the MFC-based biosensor performance
To investigate the effect of the growth of B. subtilis under semi-aerobic or anaerobic
conditions on the MFC-base biosensor performance, the voltage production and ORP change under
different aeration rates in the anodic chamber were tested. As presented in Figure 4.3, the left
scatters chart portrayed the voltage generated from the MFC device. The plot on the right side
showed the ORP change over time in the anodic chamber under no aeration, aeration at 11.32 vvm,
and aeration at 22.64 vvm aeration condition, respectively. To eliminate technical variability and
convenience for visualizing the data, both voltage and ORP were normalized. What stands out in
the figure was that the voltage generated from the 11.32 vvm outcompeted the other two groups
with two folds of voltage. For the 11.32 vvm aeration group, there was a steep rise from 2 to 8 h
of operation. After that, the voltage leveled off from 9 to 16 h. Another noticeable increase of
voltage appeared from 20-40 h. For the no aeration group, the voltage production showed a sharp
rise in the first 5 h of fermentation, and then the curve gradually increased from 5-50 h, followed
by the a slight drop at the end of the experiment. In the 22.64 vvm aeration group, the voltage
increased gradually almost during 72 h of experiment, with only a slight drop observed in 30 h.
Regarding ORP, all three groups expressed a similar pattern of change. The ORP markedly
declined in the first 10-12 h and gradually decreased over time. However, closer inspection of the

34
figure showed that the no aeration group ORP dropped significantly after 12 h, but the ORP started
to increase when the experiment was near the end.
For a long time, B. subtilis has been widely considered as a strict aerobe. At the end of the
last century, the aerobic growth habitat of B. subtilis was identified, and specific genes that could
facilitate anaerobic respiration were isolated by several studies (Glaser et al., 1995; Nakano et al.,
1997; Nakano & Zuber, 1998). As the present study was designed to determine the effect of semi-
aerobic or anaerobic anodic environment on the MFC-based biosensor, different aeration rates
were introduced to the experiment. Among three different aeration groups, the 22.64 vvm aeration
group presented the lowest voltage generation. This low voltage generation could be attributed to
the exceeded oxygen that remained in the anodic environment. And exceeded oxygen act as the
electron acceptor to complete the half cell reaction that was supposed to happen in the cathodic
chamber, which adversely affected the voltage production of the MFC-base biosensor.
Furthermore, the voltage curve from the 22.64 vvm aeration group presented a near-linear increase
and then flat growth, and the sensitivity of the biosensor may also be affected by the high aeration
rate. Kolodkin-Gal et al. (2013) have demonstrated that high oxygen levels triggered the regulatory
pathway and inhibited B.sbutilis synthesis of extracellular matrix. Similar observation were also
seen in this study. A high aeration rate in the anodic chamber impaired the biofilm formation,
leading to MFC carrying lower efficiency in electron transfer and lower sensitivity. When
comparing to no aeration and the 11.32 vvm groups, both voltage curves presented a similar pattern,
except that the 11.32 vvm group showed a more significant change in the voltage. As mentioned
above, the exceeded electron acceptor in the anodic chamber would cause low efficiency of
electron transfer rate to the cathodic chamber and led to a lower potential difference. A possible
explanation for the more significant change of voltage in 11.32 vvm group might be that B. subtilis
was both active aerobic and anaerobically, but anaerobic growth of B. subtilis presented a weak
population multiplication and electron-producing ability. Nakano and Zuber (1998) reported that
B. subtilis undertook either nitrate respiration or fermentation under strictly anaerobic conditions,
and when oxygen was limited, this strain used nitrate or nitrate as a terminal electron acceptor
instead. Nevertheless, nitrate respiration led to poor growth of bacteria and loss of the ability for
sporulation (Hoffmann et al., 1995). The ORP result further supported the idea of anaerobic
respiration. In the no aeration growth, the ORP in the anodic chamber sharply decreased as the
nitrate in the substrate was quickly consumed. As the population reached the maximum capacity,

35
both voltage and ORP change were changed gradually. Near the end of the experiment, after the
nutrient in the substrate was exhausted, B. subtilis started to die, and organic matter was degraded
back into the environment. The voltage began to decrease while the ORP started to increase
correspondingly. In a nutshell, both no aeration treatment and 11.32 vvm aeration treatment on the
MFC anodic chamber provided suitable growth conditions for bacteria and supported the MFC-
based biosensor functions. For 11.32 vvm aeration rate treatment, a suitable oxygen supplement
supports the vigor growth of B.subtilis in the anodic chamber, no excessive oxygen to jeopardize
the voltage production, and exhibits fewer disturbances during the 72-hour experiment cycle.
Hence, the 11.32 vvm aeration rate in an anodic chamber was more feasible for utilizing B. subtilis
in an MFC-base biosensor.

Figure 4.3 The effect of three different aeration rate on voltage production and anodic ORP.
4.5 Correlation between growth curve and redox potential change in MFC
The growth pattern of B. subtilis in anodic chamber and batch condition with LB broth as
substrate and the redox potential behaviors of the anodic chamber were reported for the operating
condition at pH 8 and aeration rate of 11.32 vvm at 25°C. As seen in Figure 4.4, there were three
distinguished growth phases present, including lag phase, exponential phase, and stationary phase.
B. subtilis required a prolonged period for cell replication, and the lag phase extended to 7 h to

36
complete compared to the batch fermentation of 3-4 h. The exponential phase was observed from
8 h to 28 h after fermentation started with O.D. of 1.88-1.90. The stationary phase was observed
at 30 h, and O.D. of 1.91 was resulted by the end of MFC experiment. The death phase was not
observed. By comparing the O.D. between fermentation of B. subtilis in MFC and batch conditions,
the bacterial growth in an anodic chamber required a prolonged period for cell multiplication.
Eventually, the population size and cell mass of MFC fermentation were smaller than that obtained
from batch fermentation. B. subtilis was an obligate aerobe with ample metabolic repertoire but
limited anaerobic growth may occur when an electron acceptor was insufficient or specific electron
acceptors were present (Rödel & Lücke, 1990). The reason for inferior bacterial growth MFC
fermentation may be that the anodic environment only provided partial oxygen demand and a lack
of electron acceptor to complete the metabolic cycle required by bacterial cell development.
Over the entire course of the fermentation run, the redox potential level in the anodic
chamber was ranged from -415 to 35 mV. During the lag phase, the B. subtilis was adapted to the
environmental condition in the medium. Only a slight change of redox potential occurred (±10mV).
Most of the variation in redox potential change observed in this period was due to the experimental
error. When the growth of B. subtilis entered the exponential phase, the redox potential
dramatically decreased from 25.3 mV to -370 mV from 7 up to 20 h of fermentation. This change
indicated that the metabolic activity and microbial growth significantly increased in the
exponential phase. After 20 h of operation, the growth of B. subtilis reached the stationary phase.
The redox potential then kept descending, but the tendency to decrease turned slowly to the end of
the experiment. This observation inferred that the growth and the metabolic activity were slower
in the anodic chamber. The minor fluctuation of redox potential was also observed during this
phase of growth, possibly due to continuously sparging air and agitation in the anodic chamber,
causing the increase of oxygen content and redox potential. Still, oxygen was rapidly consumed
by B. subtilis for cell growth and metabolism and against the rise of redox potential. Near the end
of the fermentation, the lowest redox potential was observed at -418 mV. This redox potential and
the cell growth curve corresponded to the finding originally reported by Lin, Chien & Duan (2010).
However, the lowest trough and trend of increase of redox potential were not observed in this
experiment. This discrepancy could be attributed to the insufficient fermentation time in this
experiment, and B. subtilis still presented vigorous vitality and substrate not depleted at the end of

37
the experiment (72 h). The result was consistent with Hongo et al. (1972), who suggested that the
ORP was strictly related to cell growth.

Figure 4.4 Growth of Bacillus subtilis under batch and MFC condition and redox potential change
of anodic chamber in MFC device vs. time.
4.6 Correlation between growth curve and voltage output from MFC
Figure 4.5 showed the change in voltage vs. cell activity. The voltage increased steadily
over 72 h of the experiment. The stable voltage of 6.17 mV was observed at 3.5 h, which was in
B. subtilis’s lag phase, and the bacteria were adapting to the environment and trying to accumulate
more biomass. Afterward, the output of voltage was attributed to the significant growth of biomass;
the voltage reached more than 100 mV after 17 h of operation. Interestingly, the rise in voltage
stopped and decreased slightly for a short period from 18-20 h, indicating the transition from the
exponential growth phase to the stationary phase. From this point, the O.D. value reached 1.89,
and the output voltage increased linearly to 336.12 mV at 72 h.
The results revealed a strong correlation between B. subtilis growth and voltage output
from the MFC. The accumulation of the B. subtilis population in the lag phase provided extra
energy to the kick-start reactions and overcame the overpotential. The first voltage from the MFC

38
indicated that the B. subtilis population started to grow. At this time, the bacterial cells established
a connection with the electrode and began shuttling electrons through the external circuitry to
complete their metabolic cycle.
The linear voltage increases of 20 h was due to the continuous growth of the biofilm on the
electrode. Logan et al. (2007) and Malvankar et al. (2011) suggested that MFC voltage generation's
capacity mainly depended on the electrode surface area and projection area between the electrode
and bacterial biofilm.
From 20 to 72 h, the bacterial population was at its maximum – yet the expanding biofilm
generates an ever-increasing voltage. Theoretically, the voltage would eventually peak when the
maximum biofilm projection area was achieved, or the substrate was depleted. In our experiment,
the observed linear rise in voltage was inadequate to elucidate the underlying bacterial growth
dynamics.

Figure 4.5 The growth of Bacillus subtilis and voltage output from MFC system by bacteria cell
activity vs. time.

39
 ∆𝑶𝑹𝑷
4.7 Correlation analysis from potential parameter (𝑿 = 𝒆− ∆𝑽 )
The Nernst Equation evaluated the relation of reduction potential and the standard
electrode potential with given conditions. In Section 4.1, the original Nernst equation was
reformulated to adapt to the MFC condition, in which a unitless potential parameter X correlates
the fermentation redox potential and cell potential (voltage) was derived (Equation 3.5).
Figure 4.6 illustrated the intercorrelation between the potential parameter X and bacterial
growth by plotting the potential parament X, ORP, O.D., and d(O.D.)/dt vs. time. As seen in Figure
4.6, there were 2 peaks found along the potential parameter X curve. The potential parameter x
was 0 till 6 h of culturing. While B. subtilis growth left lag phase and entered the exponential phase,
the O.D. and potential parameter X increased dramatically, and the ORP decreased significantly.
The rise of potential parameters peaked around 12-13 h after inoculation, and the curve started to
decline rapidly afterward. When the potential parameter X curve peaked, the d(O.D.)/dt is also
peaked, which revealed the bacterial strain underwent the most rapid stage during growth. The
second peak was observed at approximately 19–20 h culturing time. While the second peak
reached in potential parameter X, the d(O.D.)/dt proceeded a sharped decrease. The change of O.D.
and ORP had been leveled off, indicating the deceleration of bacterial growth and the growth phase
transited from the exponential phase into stationary phase. After the second peak, the potential
parameter curve rapid declined between 22-24 h and then gradually decreased over time till the
end of the experiment.
The results suggested a correlation between the potential parameter and bacterial growth
under MFC conditions. The initial increase of the potential parameter curve indicated the microbial
strain thriving growth after the lag phase, and the population size increased significantly. While
the first peak was reached, the d(O.D.)/dt also reached its maximal value, which revealed that B.
subtilis underwent the most rapid cell propagation and population enlargement at the midpoint of
the exponential phase. As the second peak arrived, the growth of B. subtilis started to decline
simultaneously. The ORP and the growth curve indicated that the exponential phase was closed to
the end. Overall, the potential parameter X presented real-time, highly responsive, and high-
resolution information that revealed the start of the exponential phase, the midpoint of the
exponential phase, and the onset of the stationary phase. Compared to other microbial growth
monitoring parameters (O.D. and ORP), the O.D. can not operate in real-time; the measurement
process is time-consuming and labor-intensive. And the O.D. substantially is a turbidimetric

40
assessment that is prone to error, and the change of O.D. is delayed towards the actual microbial
growth dynamics. On the other hand, ORP is a versatile parameter, but it only presents the potential
parameter's partial information. The advantage of potential parameter X combined the ORP and
voltage measurement, which not only evaluated the external environmental condition change
(ΔORP) triggered by microbial growth but also took the internal metabolic activities and actual
microbial performance (ΔV) into account. Hence, the potential parameter X provided an insightful
understanding of bacterial growth dynamics and enabled precise fermentation monitoring and
controlling.
Therefore, based on our findings, the potential parameter derived from Nernst Equation
could be applied to the MFC and presented a strict correlation to the cell growth dynamics.

Figure 4.6 Profiles of potential parameter, ORP, optical density, and change of optical density of B. subtilis in 72 h.

41
 5 CONCLUSION AND RECOMMENDATIONS

This study set out to provide a step-by-step experimental design to prove the concept of
utilizing the MFC as the fermentation ORP sensor. MFC-based fermentation biosensor enabled in-
situ, real-time, and precise fermentation monitoring and provided a practical, economically
feasible alternative for traditional ORP probes. The modified Nernst equation offers a thoughtful
insight into integrating the MFC system's voltage and redox potential. A relationship among
voltage production, redox potential change, and bacterial growth dynamics could be correlated.
Compared to ORP monitoring, such a relationship elucidates the fermentation kinetics by
incorporating external chemical profiles and bacteria's internal metabolic performance.
This study lays the groundwork for future research into the novel biosensor research.
Despite these promising results, several interesting questions remained unaddressed. MFCs are
sensitive to factors such as temperature and pH – which also played an important role in the Nernst
equation (Kyazze et al., 2010). Other environmental conditions may impact the essential
correlations delineated in this work. Lastly, this research should employ other bacteria/bacterial-
consortia, other growth media, and other MFC architectures.
The first step in this study was to investigate the feasibility of different bacteria. Two common
bacteria with distinct physiological characters, B. subtilis and P. fluorescens were selected as
candidates. The B. subtilis present better adaptation, rapid growth, and strong power-producing
ability in the MFC condition. Different cathode and anode treatments have also been considered
and compared in this study. This study had tested other electron acceptors (oxygen, potassium
ferricyanide) and different cathode treatments (aerated, sparging), and the result revealed that
sparging is a better choice among different treatments by considering the functionality and
simplicity. The effect of anode aeration rate of 0, 11.32, 22.64 vvm had been investigated in this
study as well. The result suggested that aeration at 11.32 vvm would enhance microbial growth
and further support biosensor function. Finally, integrating all the above knowledge gained from
the experiments, voltage, ORP, and O.D. from MFC over time had been plotted. The voltage curve
presented a flat increase corresponding to the bacterial growth, and it is inadequate to elucidate the

42
underlying bacterial growth dynamics. Although the ORP curve presented a strict relationship to
the O.D. change, only one peak represented the shifting from lag to the exponential phase. By
utilizing the potential parameter derived from the modified Nernst equation, a tight relationship
was portrayed among the MFC output, fermentation ORP, and bacterial growth dynamics. The
potential parameter presented a detailed and high-resolution curve, with higher sensitivity to
monitoring the bacterial growth dynamic than ORP and voltage alone. Such a method provides
insight for understanding the fermentation bioprocess.
A limitation in this study is that investigation only emphasis the electrochemical performance
and the relationship between bacteria and fermentation dynamics. The metabolic flux, multi-omics
analysis, and specific electrochemical assay are not researched. Whilst this study only constructs
a laboratory-scale simplified MFC under optimal conditions, it did substantiate the concept of an
alternative real-time MFC biosensor. Disparate from the traditional BOD/COD MFC-based
biosensor, this novel MFC biosensor monitored internal ORP change, and integrated with the
voltage generation to achieve the fermentation monitoring function. Possible future work could be
done in two different directions:
1. Apply the result and experience gained from this study and develop a versatile, miniature
MFC-based biosensor. Such biosensors could apply not only for industrial fermentation
but also for soil, wastewater, and oenological research.
2. Investigate the specific physiological adaption of bacteria in the MFC-base biosensor and
utilize high-throughput technology to enable metabolomics, transcriptomic, and proteinic
research. Construct specific strain that specialized for the MFC-based biosensor.

43
 6 REFERENCES

Allen, J. F. (1993). control of gene expression by redox potential and the requirement for
chloroplast and mitochondrial genomes. Journal of Theoretical Biology, 165(4), 609-631.
doi:10.1006/jtbi.1993.1210
Allen, R. M., & Bennetto, H. P. (1993). Microbial fuel-cells: Electricity production from
carbohydrates. Appl Biochem Biotechnol, 39(27), 27–40.
Asveld, L., & Van Est, R. (2011). Getting to the core of the bio-economy A perspective on the
sustainable promise of biomass. In. Netherlands.
Bonan, C. I. D. G., Biazi, L. E., Dionísio, S. R., Soares, L. B., Tramontina, R., Sousa, A. S., . . .
Ienczak, J. L. (2020). Redox potential as a key parameter for monitoring and optimization
of xylose fermentation with yeast Spathaspora passalidarum under limited-oxygen
conditions. Bioprocess and Biosystems Engineering, 43(8), 1509-1519.
doi:10.1007/s00449-020-02344-2
Cantó, C., Menzies, J., Keir, & Auwerx, J. (2015). nad+ metabolism and the control of energy
homeostasis: a balancing act between mitochondria and the nucleus. Cell Metabolism,
22(1), 31-53. doi:10.1016/j.cmet.2015.05.023
Chang, Cheng, H.-B., & Chao, A. C. (2004). applying the nernst equation to simulate redox
potential variations for biological nitrification and denitrification processes. Environmental
Science & Technology, 38(6), 1807-1812. doi:10.1021/es021088e
Chang, S.-H., Wu, C.-H., Wang, R.-C., & Lin, C.-W. (2017). Electricity production and benzene
removal from groundwater using low-cost mini tubular microbial fuel cells in a monitoring
well. Journal of Environmental Management, 193, 551-557.
Chaudhuri, S. K., & Lovely, D. R. (2003). Electricity generation by direct oxidation of glucose in
mediatorless microbial fuel cells. Nat Biotechnol., 1229-1232.

44
Chen, B.-Y., Ma, C.-M., Han, K., Yueh, P.-L., Qin, L.-J., & Hsueh, C.-C. (2016). Influence of
textile dye and decolorized metabolites on microbial fuel cell-assisted bioremediation.
Bioresource Technology, 200, 1033-1038.
Chouler, J., & Di Lorenzo, M. (2015). water quality monitoring in developing countries; can
microbial fuel cells be the answer? Biosensors (Basel), 5(3), 450-470.
doi:10.3390/bios5030450Cohen, B. (1931). The bacterial culture as an electrical half-cell.
Bacteriol, 21, 18-19.
Cohen, B., Chambers, R., & Reznikoff, P. (1928). intracellular oxidation-reduction studies : i.
reduction potentials of amoeba dubia by micro injection of indicators. J Gen Physiol, 11(5),
585-612. doi:10.1085/jgp.11.5.585
Coton, M., Pawtowski, A., Taminiau, B., Burgaud, G., Deniel, F., Coulloumme-Labarthe, L.,
Coton, E. (2017). Unraveling microbial ecology of industrial-scale Kombucha
fermentations by metabarcoding and culture-based methods. FEMS Microbiology Ecology,
93(5). doi:10.1093/femsec/fix048
Creasey, R., Mostert, B., Nguyen, T., Virdis, B., Freguia, S., & Laycock, B. (2018). Microbial
nanowires – Electron transport and the role of synthetic analogues. Acta Biomaterialia, 69,
1-30.
Cui, Y., Lai, B., & Tang, X. (2019). Microbial Fuel Cell-Based Biosensors. Biosensors (Basel),
9(3). doi:10.3390/bios9030092
Davis, J. B., & Yarbrough, H. F. (1962). preliminary experiments on a microbial fuel cell. Science,
137(3530), 615-616.
Delord, B., Neri, W., Bertaux, K., Derre, A., Ly, I., Mano, N., & Poulin, P. (2017). Carbon
nanotube fiber mats for microbial fuel cell electrodes. Bioresour Technol, 243, 1227-1231.
doi:10.1016/j.biortech.2017.06.170
Di Lorenzo, M., Thomson, A. R., Schneider, K., Cameron, P. J., & Ieropoulos, I. (2014). A small-
scale air-cathode microbial fuel cell for on-line monitoring of water quality. Biosensors
and Bioelectronics, 62, 182-188. doi:https://doi.org/10.1016/j.bios.2014.06.050
Do, M. H., Ngo, H. H., Guo, W., Chang, S. W., Nguyen, D. D., Liu, Y., . . . Kumar, M. (2020).
Microbial fuel cell-based biosensor for online monitoring wastewater quality: A critical
review. Science of The Total Environment, 712, 135612.
doi:https://doi.org/10.1016/j.scitotenv.2019.135612

45
Du, Z., Li, H., & Gu, T. (2007). A state of the art review on microbial fuel cells: A promising
technology for wastewater treatment and bioenergy. Biotechnology Advances, 25, 464-482.
Emde, R., Swain, A., & Schink, B. (1989). Anaerobic oxidation of glycerol by Escherichia coli in
an amperometric poised-potential culture system. Applied Microbiology and
Biotechnology, 32(2), 170-175.
Evelyn, Li, Y., Marshall, A., & Gostomski, P. A. (2014). Gaseous pollutant treatment and
electricity generation in microbial fuel cells (MFCs) utilising redox mediators. Reviews in
Environmental Science and Biotechnology, 35-51.
Feiner, A.-S., & McEvoy, A. (1994). The nernst equation. Journal of Chemical Education, 71(6),
493.
Ferguson, J. B. (1923). The oxides of iron. Journal of the Washington Academy of Sciences, 13(13),
275-281.
Foyer, C. H., & Noctor, G. (2005). redox homeostasis and antioxidant signaling: a metabolic
interface between stress perception and physiological responses. The Plant Cell, 17(7),
1866-1875. doi:10.1105/tpc.105.033589
Ghasemi, M., Wan Daud, W. R., Ismail, M., Rahimnejad, M., Ismail, A. F., Leong, J. X., Ben
Liew, K. (2013). Effect of pre-treatment and biofouling of proton exchange membrane on
microbial fuel cell performance. International Journal of Hydrogen Energy, 38(13), 5480-
5484. doi:10.1016/j.ijhydene.2012.09.148
Glaser, P., Danchin, A., Kunst, F., Zuber, P., & Nakano, M. M. (1995). Identification and isolation
of a gene required for nitrate assimilation and anaerobic growth of Bacillus subtilis.
Journal of Bacteriology, 177(4), 1112-1115. doi:10.1128/jb.177.4.1112-1115.1995
Gorby, Y. A., Yania, S., McLean, J. S., Rosso, K. M., Moyles, D., Dohnalkova, A., . . . Fredrickson,
J. K. (2006). Electrically conductive bacterial nanowires produced by Shewanella
oneidensis strain MR-1 and other microorganisms. Proc Natl Acad Sci U S A, 103(30),
11358–11363.
GrandViewResearch. (2019). Fermentation Chemicals Market Size, Share & Trends Analysis
Report By Product (Alcohols, Enzymes), By Application (Industrial, Plastic,
Pharmaceutical), By Region, And Segment Forecasts, 2019 - 2025. Retrieved from

46
He, Z., Minteer, S. D., & Angenent, L. T. (2005). Electricity generation from artificial wastewater
using an upflow microbial fuel cell. Environmental Science and Technology, 39(14), 5262-
5267.
He, Z., Wagner, N., Minteer, S. D., & Angenent, L. T. (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.
Hoffmann, T., Troup, B., Szabo, A., Hungerer, C., & Jahn, D. (1995). The anaerobic life of
Bacillus subtilis: Cloning of the genes encoding the respiratory nitrate reductase system.
FEMS Microbiology Letters, 131(2), 219-225. doi:10.1111/j.1574-6968.1995.tb07780.x
Hongo, M., Ishizaki, A., & Uyeda, M. (1972). studies on oxidation-reduction potentials (orp) of
microbial cultures. Agricultural and Biological Chemistry, 36(1), 141-145.
doi:10.1080/00021369.1972.10860221
Huang, X., & Lin, Y. H. (2020). Reconstruction and analysis of a three‐compartment genome‐
scale metabolic model for Pseudomonas fluorescens. Biotechnology and Applied
Biochemistry, 67(1), 133-139. doi:10.1002/bab.1852
Humphrey, A. E., & Lee, S. E. (1992). Industrial Fermentation: Principles, Processes, and Products.
In J. A. Kent (Ed.), Riegel’s Handbook of Industrial Chemistry (pp. 916-986). Dordrecht:
Springer Netherlands.
Ishizaki, A., Shibai, H., & Hirose, Y. (1974). basic aspects of electrode potential change in
submerged fermentation. Agricultural and Biological Chemistry, 38(12), 2399-2406.
doi:10.1080/00021369.1974.10861537
Ismail, Z. Z., & Jaeel, A. J. (2013). Sustainable power generation in continuous flow microbial
fuel cell treating actual wastewater: influence of biocatalyst type on electricity production.
The Scientific World Journal, 2013, 1-7. doi:10.1155/2013/713515
Jensen, W. B. (1996). electronegativity from avogadro to pauling: part 1: origins of the
electronegativity concept. Journal of Chemical Education, 73(1), 11.
doi:10.1021/ed073p11
Jiang, H., Halverson, L. J., & Dong, L. (2015). A miniature microbial fuel cell with conducting
nanofibers-based 3D porous biofilm. Journal of Micromechanics and Microengineering,
25(12), 1-15.
Jones, D. P., & Sies, H. (2015). The redox code. Antioxidants & redox signaling, 23(9), 734-746.

47
Kakarla, R., Kim, J. R., Jeon, B.-H., & Min, B. (2015). Enhanced performance of an air–cathode
microbial fuel cell with oxygen supply from an externally connected algal bioreactor.
Bioresource Technology, 195, 210-216. doi:10.1016/j.biortech.2015.06.062
Kalathil, S., Patil, S. A., & Pant, D. (2018). microbial fuel cells: electrode materials. In K. Wandelt
(Ed.), Encyclopedia of Interfacial Chemistry (pp. 309-318). Oxford: Elsevier.
Karamanou, M., & Androutsos, G. (2013). Antoine-Laurent de Lavoisier (1743–1794) and the
birth of respiratory physiology. Thorax, 68(10), 978-979. doi:10.1136/thoraxjnl-2013-
203840
Killeen, D. J., Boulton, R., & Knoesen, A. (2018). advanced monitoring and control of redox
potential in wine fermentation. American Journal of Enology and Viticulture,
ajev.2018.17063. doi:10.5344/ajev.2018.17063
Kim, B. H., Kim, H. J., Hyun, M. S., & Park, D. H. (1999). Direct electrode reaction of Fe(III)-
reducing bacterium, Shewanella putrifaciens. J Microbiol Biotechnol, 127-131.
Kolodkin-Gal, I., Elsholz, A. K. W., Muth, C., Girguis, P. R., Kolter, R., & Losick, R. (2013).
Respiration control of multicellularity in Bacillus subtilis by a complex of the cytochrome
chain with a membrane-embedded histidine kinase. Genes & Development, 27(8), 887-899.
doi:10.1101/gad.215244.113
Kovárová-Kovar, K., & Egli, T. (1998). Growth kinetics of suspended microbial cells: from single-
substrate-controlled growth to mixed-substrate kinetics. Microbiology and molecular
biology reviews : MMBR, 62(3), 646-666. doi:10.1128/MMBR.62.3.646-666.1998
Kyazze, G., Popov, A., Dinsdale, R., Esteves, S., Hawkes, F., Premier, G., & Guwy, A. (2010).
Influence of catholyte pH and temperature on hydrogen production from acetate using a
two chamber concentric tubular microbial electrolysis cell. International Journal of
Hydrogen Energy, 35(15), 7716-7722. doi:https://doi.org/10.1016/j.ijhydene.2010.05.036
Lande, A. D. L., Babcock, N. S., Řezáč, J., Lévy, B., Sanders, B. C., & Salahub, D. R. (2012).
Quantum effects in biological electron transfer. Physical Chemistry Chemical Physics,
14(17), 5902. doi:10.1039/c2cp21823b
Lawson, K., Rossi, R., Regan, J. M., & Logan, B. E. (2020). Impact of cathodic electron acceptor
on microbial fuel cell internal resistance. Bioresource Technology, 316, 123919.
doi:https://doi.org/10.1016/j.biortech.2020.123919

48
Li, W.-W., Sheng, G.-P., Liu, X.-W., & Yu, H.-Q. (2011). Recent advances in the separators for
microbial fuel cells. Bioresource Technology, 244-252.
Li, X., Zheng, R., Zhang, X., Liu, Z., Zhu, R., & Zhang, X. (2019). A novel exoelectrogen from
microbial fuel cell: Bioremediation of marinepetroleum hydrocarbon pollutants. Journal
of Environmental Management, 235, 70-76.
Lin, Y.-H., Chien, W.-S., & Duan, K.-J. (2010). Correlations between reduction–oxidation
potential profiles and growth patterns of Saccharomyces cerevisiae during very-high-
gravity fermentation. Process Biochemistry, 45(5), 765-770.
doi:10.1016/j.procbio.2010.01.018
Liu, Xue, C., Lin, Y., & Bai, F. (2013). Redox potential control and applications in microaerobic
and anaerobic fermentations. Biotechnol Adv, 31(2), 257-265.
doi:10.1016/j.biotechadv.2012.11.005
Liu, C., Qin, J.-C., & Lin, Y.-H. (2017). Fermentation and redox potential. Fermentation Processes.
IntechOpen, 23-42.
Liu, H., & Logan, B. E. (2004). electricity generation using an air-cathode single chamber
microbial fuel cell in the presence and absence of a proton exchange membrane.
Environmental Science Technology, 38, 4040-4046.
Liu, Z., Liu, J., Zhang, S., Xing, X.-H., & Su, Z. (2011). Microbial fuel cell based biosensor for in
situ monitoring of anaerobic digestion process. Bioresource Technology, 102(22), 10221-
10229. doi:https://doi.org/10.1016/j.biortech.2011.08.053
Logan, B., Cheng, S., Watson, V., & Estadt, G. (2007). graphite fiber brush anodes for increased
power production in air-cathode microbial fuel cells. Environmental Science & Technology,
41(9), 3341-3346. doi:10.1021/es062644y
Logan, B., Hamelers, B., Rozendal, R., Schroder, U., Keller, J., Freguia, S., . . . Rabaey, K. (2006).
microbial fuel cells: methology and technology. Environmental Science & Technology,
40(17), 5181-5192.
Logan, B. E. (2009). Exoelectrogenic bacteria that power microbial fuel cells. Nature Reviews
Microbiology, 7, 375-381.
Lohar, S. A., Patil, V. D., & Patil, D. B. (2015). role of mediators in microbial fuel cell for
generation of electricity and waste water treatment. International Journal of Chemical
Sciences, 6, 7-11.

49
Lovley, D. (2006). Bug juice: harvesting electricity with microorganisms. Nature Reviews
Microbiology, 4, 497-508.
Lovley, D. R. (2008). The microbe electric: conversion of organic matter to electricity. Current
Opinion in Biotechnology, 19(6), 564-571. doi:10.1016/j.copbio.2008.10.005
Mahadevan, A., Gunawardena, D. A., & Fernando, S. (2014). biochemical and electrochemical
perspectives of the anode of a microbial fuel cell. In Technology and Application of
Microbial Fuel Cells (pp. 13–32): Intech Open.
Malvankar, N. S., Vargas, M., Nevin, K. P., Franks, A. E., Leang, C., Kim, B.-C., . . . Lovley, D.
R. (2011). Tunable metallic-like conductivity in microbial nanowire networks. Nature
Nanotechnology, 6(9), 573-579. doi:10.1038/nnano.2011.119
Marcus, R. A. (1956). On the theory of oxidation‐reduction reactions involving electron transfer.
I. The Journal of Chemical Physics, 24(5), 966-978. doi:10.1063/1.1742723
Marcus, R. A., & Sutin, N. (1985). Electron transfers in chemistry and biology. Biochimica et
Biophysica Acta (BBA) - Reviews on Bioenergetics, 811(3), 265-322. doi:10.1016/0304-
4173(85)90014-x
María Martínez-Espinosa, R. (2020). Introductory Chapter: A brief overview on fermentation and
challenges for the next future. In New Advances on Fermentation Processes: IntechOpen.
Mckenney, P. T., Driks, A., & Eichenberger, P. (2013). The Bacillus subtilis endospore: assembly
and functions of the multilayered coat. Nature Reviews Microbiology, 11(1), 33-44.
doi:10.1038/nrmicro2921
Mohan, S. V., Raghavulu, S. V., Srikanth, S., & Sarma, P. N. (2007). Bioelectricity production by
mediatorless microbial fuel cell under acidophilic condition using wastewater as substrate:
Influence of substrate loading rate. Current Science, 92(12), 1720-1726. Retrieved from
http://www.jstor.org/stable/24107621
Morris, J. M., & Jin, S. (2008). Feasibility of using microbial fuel cell technology for
bioremediation of hydrocarbons in groundwater. Journal of Environmental Science and
Health, Part A, 18–23.
Mulchandani, A. (1998). Principles of enzyme biosensors. In A. Mulchandani & K. R. Rogers
(Eds.), Enzyme and Microbial Biosensors: Techniques and Protocols (pp. 3-14). Totowa,
NJ: Humana Press.

50
Nakano, M. M., Dailly, Y. P., Zuber, P., & Clark, D. P. (1997). Characterization of anaerobic
fermentative growth of Bacillus subtilis: identification of fermentation end products and
genes required for growth. Journal of Bacteriology, 179(21), 6749-6755.
doi:10.1128/jb.179.21.6749-6755.1997
Nakano, M. M., & Zuber, P. (1998). Anaerobic growth of a “strict aerobe” (Bacillus subtilis).
Annual Review of Microbiology, 52(1), 165-190. doi:10.1146/annurev.micro.52.1.165
Needham, J., & Needham, D. M. (1926). The oxidation-reduction potential of protoplasm: A
review. Protoplasma, 1(1), 255-294. doi:10.1007/bf01602996
Nimje, V. R., Chen, C.-Y., Chen, C.-C., Jean, J.-S., Reddy, A. S., Fan, C.-W., . . . Chen, J.-L.
(2009). Stable and high energy generation by a strain of Bacillus subtilis in a microbial fuel
cell. Journal of Power Sources, 190(2), 258-263. doi:10.1016/j.jpowsour.2009.01.019
Oh, S.-E., & Logan, B. E. (2006). Proton exchange membrane and electrode surface areas as
factors that affect power generation in microbial fuel cells. Applied Microbiology and
Biotechnology, 70(2), 162-169. doi:10.1007/s00253-005-0066-y
Oliveira, V. B., Simoes, M., Melo, L. F., & Pinto, A. M. F. R. (2013). Overview on the
developments of microbial fuel cells. Biochemical Engineering Journal, 73, 53-64.
Park, S. H., Kim, E. H., Chang, H. J., Yoon, S. Z., Yoon, J. W., Cho, S.-J., & Ryu, Y.-H. (2013).
history of bioelectrical study and the electrophysiology of the primo vascular system.
Evidence-Based Complementary and Alternative Medicine.
Paul, A., Laurila, T., Vuorinen, V., & Divinski, S. V. (2014). Fick’s laws of diffusion. In
Thermodynamics, Diffusion and the Kirkendall Effect in Solids (pp. 115-139): Springer
International Publishing.
Peixoto, L., Min, B., Martins, G., Brito, A. G., Kroff, P., Parpot, P., . . . Nogueira, R. (2011). In
situ microbial fuel cell-based biosensor for organic carbon. Bioelectrochemistry, 81(2), 99-
103. doi:https://doi.org/10.1016/j.bioelechem.2011.02.002
Pepè Sciarria, T., Arioli, S., Gargari, G., Mora, D., & Adani, F. (2019). Monitoring microbial
communities’ dynamics during the start-up of microbial fuel cells by high-throughput
screening techniques. Biotechnology Reports, 21, e00310.
doi:https://doi.org/10.1016/j.btre.2019.e00310
Potter, M. C. (1912). electrical effects accompanying the decomposition of organic compound.
Proceedings of the Royal Society of London, 84(260).

51
Pous, N., Puig, S., Coma, M., & Balaguer, M. D. (2013). feasibility of using bioelectrochemical
systems for bioremediation. Bioremediation of nitrate-pollutedgroundwater in a microbial
fuel cell, 88, 1690–1696.
Rabaey, K., Boon, N., Siciliano, S., Verhaege, M., & Verstraete, W. (2004). biofuel cells select
for microbial consortia that self-mediate electron transfer. Applied and Environmental
Microbiology, 70(9), 5373-5382.
Rahimnejad, M., Adhami, A., Darvari, S., Zirepour, A., & Oh, S. E. (2015). Microbial fuel cell as
new technol ogy for bioelectricity generation: A review. Alexandria Engineering Journal,
54(3), 745-756.
Reguera, G., McCarthy, K. D., Mehta, T., Nicoll, J. S., Tuominen, M. T., & Lovley, D. R. (2005).
Extracellular electron transfer via microbial nanowires. Nature, 435(7045), 1098-1101.
Ringeisen, B. R., Henderson, E., Wu, P. K., Pietron, J., Ray, R., Little, B., . . . Jones-Meehan, J.
M. (2006). High power density from a miniature microbial fuel cell using Shewanella
oneidensis DSP10. Environmental Science and Technology, 40(8), 2629-2634.
Rismani-Yazdi, H., Carver, S. M., Christy, A. D., & Tuovinen, O. H. (2008). Cathodic limitations
in microbial fuel cells: An overview. Journal of Power Sources, 180(2), 683-694.
doi:10.1016/j.jpowsour.2008.02.074
Rödel, W., & Lücke, F.-K. (1990). Effect of redox potential on Bacillus subtilis and Bacillus
licheniformis in broth and in pasteurized sausage mixtures. International Journal of Food
Microbiology, 10(3-4), 291-301. doi:10.1016/0168-1605(90)90076-h
Rohrback, G. H., Scott, W. R., & Canfield, J. H. (1962). Biochemical fuel cells. Power sources
conference., 16, 18-21.
Sangeetha, T., & Muthukumar, M. (2012). influence of electrode material and electrode distance
on bioelectricity production from sago-processing wastewater using microbial fuel cell.
Environmental Progress & Sustainable Energy, 32(2), 390-395.
Schafer, F. Q., & Buettner, G. R. (2001). Redox environment of the cell as viewed through the
redox state of the glutathione disulfide/glutathione couple. Free Radical Biology and
Medicine, 30(11), 1191-1212. doi:10.1016/s0891-5849(01)00480-4
Scholz, F. (2017). Wilhelm Ostwald’s role in the genesis and evolution of the Nernst equation.
Journal of Solid State Electrochemistry, 21(7), 1847-1859. doi:10.1007/s10008-017-3619-
y

52
Schroder, U. (2007). Anodic electron transfer mechanisms in microbial fuel cells and their energy
efficiency. Physical Chemistry Chemical Physics, 9(21), 2619-2629. Retrieved from
https://pubs.rsc.org/en/content/articlepdf/2007/cp/b703627m
Shapiro, H. M. (1972). Redox balance in the body: An approach to quantitation. Journal of
Surgical Research, 13(3), 138-152. doi:10.1016/0022-4804(72)90057-1
Shi, L., Squier, T. C., Zachara, J. M., & Fredrickson, J. K. (2007). Respiration of metal
(hydr)oxides by Shewanella and Geobacter: A key role for multihaem c-type cytochromes.
Molecular Microbiology, 65(1), 12-20.
Singh, A., & Yakhmi, J. (2014). Microbial fuel cells – Applications for generation of electrical
power and beyond. Critical Reviews in Microbiology, 1-17.
Song, H.-L., Zhu, Y., & Li, J. (in press). Electron transfer mechanisms, characteristicsand
applications of biological cathode microbialfuel cells – A mini review. Arabian Journal of
Chemistry.
Su, L., Jia, W., Hou, C., & Lei, Y. (2011). Microbial biosensors: A review. Biosensors and
Bioelectronics, 26(5), 1788-1799. doi:https://doi.org/10.1016/j.bios.2010.09.005
Suslow, T. V. (2004). Oxidation-reduction potential (orp) for water disinfection monitoring,
control, and documentation. doi:10.3733/ucanr.8149
Torres, C. I., Marcus, A. K., Lee, H. S., Parameswaran, P., Krajmalnik-Brown, R., & Rittmann, B.
E. (2010). A kinetic perspective on extracellular electron transfer by anode-respiring
bacteria. FEMS Microbiology Reviews, 34(1), 3-17.
Ucar, D., Zhang, Y., & Angelidaki, I. (2017). An overview of electron acceptors in microbial fuel
cells. Frontiers in Microbiology, 8(643). doi:10.3389/fmicb.2017.00643
Uria, N., Ferrera, I., & Mas, J. (2017). Electrochemical performance and microbial community
profiles in microbial fuel cells in relation to electron transfer mechanisms. BMC
Microbiology, 17(1), 1-12.
Vargas, M., Malvankar, N. S., Tremblay, P.-L., Leang, C., Smith, J. A., Patel, P., . . . Lovley, D.
R. (2013). aromatic amino acids required for pili conductivity and long- range extracellular
electron transport in geobacter sulfurreducens. microbiology, 106(23), e00105-00113.
Vejarano, F., Benítez-Campo, N., Bravo, E., Loaiza, O. A., & Lizcano-Valbuena, W. H. (2018).
Electrochemical Monitoring and Microbial Characterization of a Domestic Wastewater-

53
 Fed Microbial Fuel Cell Inoculated with Anaerobic Sludge. Revista de Ciencias, 22, 13-
32. doi:https://doi.org/10.25100/rc.v22i2.7910
Vellingiri, A., Song, Y. E., Munussami, G., Kim, C., Park, C., Jeon, B. H., . . . Kim, J. R. (2019).
Overexpression of c-type cytochrome, CymA in Shewanella oneidensis MR-1 for
enhanced bioelectricity generation and cell growth in a microbial fuel cell. Journal of
Chemical Technology and Biotechnology, 94(7), 2115-2122.
Wang, R.-S., Oldham, W. M., Maron, B. A., & Loscalzo, J. (2018). systems biology approaches
to redox metabolism in stress and disease states. Antioxidants & redox signaling, 29(10),
953-972. doi:10.1089/ars.2017.7256
Weibel, M. K., & Dodge, C. (1975). Biochemical fuel cells. Demonstration of an obligatory
pathway involving an external circuit for the enzymatically catalyzed aerobic oxidation of
glucose. Arch Biochem Biophys., 169(1), 146-151.
Yaropolov, A. I., Varfolomeev, S. D., & Berezin, I. V. (1976). Bioelectrocatalysis. activation of a
cathode oxygen reduction in the peroxidase-mediator carbon electrode system. Febs Letter,
71(2), 306-308.
Yong, Y.-C., Yu, Y.-Y., Li, C.-M., Zhong, J.-J., & Song, H. (2011). Bioelectricity enhancement
via overexpression of quorum sensing system in Pseudomonas aeruginosa-inoculated
microbial fuel cells. Biosensors and Bioelectronics, 30(1), 87-92.
doi:https://doi.org/10.1016/j.bios.2011.08.032
Zhang, X. C., & Halme, A. (1995). Modelling of a microbial fuel cell process. Biotechnology
Letters, 17(8), 809-814. doi:10.1007/BF00129009
Zhao, L., Brouwer, J., Naviaux, J., & Hochbaum, A. (2014, 2014-06-30). Modeling of Polarization
Losses of a Microbial Fuel Cell. Paper presented at the ASME 2014 12th International
Conference on Fuel Cell Science, Engineering and Technology.
Zhou, S., Huang, S., Li, Y., Zhao, N., Li, H., Angelidaki, I., & Zhang, Y. (2018). Microbial fuel
cell-based biosensor for toxic carbon monoxide monitoring. Talanta, 186, 368-371.
doi:https://doi.org/10.1016/j.talanta.2018.04.084
Zhu, J.-G., Ji, X.-J., Huang, H., Du, J., Li, S., & Ding, Y.-Y. (2009). Production of 3-
hydroxypropionic acid by recombinant Klebsiella pneumoniae based on aeration and ORP
controlled strategy. Korean Journal of Chemical Engineering, 26(6), 1679-1685.

54
Zilberman, D., & Kim, E. (2011). The lessons of fermentation for the new bio-economy.
Agbioforum (Columbia, Mo.), 14(3), 97-103.

55
 7 APPENDICES

7.1 Appendix A MFC apparatus and instrumentation

Figure 7.1 Actual MFC-based biosensor laboratory setup

Figure 7.2 Spectrophotometer for bacterial growth monitoring

56
7.2 Appendix B pH profile during fermentation

Figure 7.3 pH profile during the MFC fermentation, the potential parameter presents
random dispersion of change without introduces pH regulator

57