Citation
New electrode materials and active energy harvesting for microbial electrochemical systems or MXCS

Material Information

Title:
New electrode materials and active energy harvesting for microbial electrochemical systems or MXCS
Creator:
Wang, Heming ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (141 pages). : ;

Subjects

Subjects / Keywords:
Fuel cells -- Research ( lcsh )
Microbial fuel cells ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Microbial electrochemical system of MXC is an emerging platform technology that integrates multiple disciplines and carries different functions, such as water treatment, environmental remediation, and desalination with simultaneous power or value-added chemical production. The versatility and high performance distinguishes MXC from traditional treatment-focused and energy-intensive environmental systems and makes it a potential transformative technology for the next generation of environmental biotechnology. The Chapter 1 of this dissertation summarizes nearly 50 corresponding reactor systems developed based on MXC platform, with the goal to introduce and summarize all the functions that have been developed so far and discusses the niche of this technology for environmental science and engineering. The MXC technology carries great potentials in transforming waste treatment process into energy and resource recovery systems, but the technology is still in early stage development, and there are many remaining challenges need to be addressed. In addition to comprehensive literature review, my doctoral study has been focused on the following three aspects on MXC scientific characterization and engineering development: 1. Development and characterization of high performing and low-cost electrode materials. The anode and cathode are two crucial elements in MXC structures and largely determine the performance and cost of a reactor system. The feasibility of using a recycled material tire crumb rubber, as a low-cost anode material in microbial fuel cells MFC was investigated (Chapter 2) in my first year of study. This is the first study that used recycled materials as an alternative to traditional graphite or carbon cloth anodes, the results show that tire crumbs produced comparable level of electricity after surface treatment. On the cathode side, three types of new carbon nanotube based air-cathode with very high surface area were developed, which created a 3-D structure and demonstrated superior performance as compared to traditional carbon cloth air-cathodes (Chapter 3). 2. Innovative active energy harvesting during organic removal and waste treatment. MXC can produce direct electricity from waste materials, but currently the current and voltage output are low, making the direct use of such power difficult. A new active harvesting approach using a maximum power points circuit (MPPS) was developed, which totally changed the traditional passive energy gaining process through resistors or non-controllable charge pumps. The system dramatically increased MFC power production by more than 70 times and improved energy recovery by 20 times (Chapter 4). Furthermore, the effects of inductance, duty ratio, and switching frequency on MFC energy harvesting were characterized to optimize operating conditions and direct further circuit development (Chapter 5). 3. Removal mechanisms of emerging trace organic compounds (TorCs) in microbial fuel cells. The MXC can theoretically degrade any biodegradable materials to produce energy, and many different kinds of waste streams have been tested. The removal of traditional organic matter and sulfur, and the recovery of nitrogen and phosphate have all been proven possible. The feasibility of removal and transformation of low level (e.g. ng/L) trace organic compounds (TorCs) in wastewater was studied in microbial fuel cells (Chapter 6), because it was hypothesized that MFC can be effective in this task by providing both reductive (anode) and oxidative (cathode) environments, and the biodegradation can be enhanced with electrochemical transformation. Results show that MFC was effective in removing most of the 34 TorCs tested with different efficiencies, and electricity was produced from all reactors.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver. Electrical engineering
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
Department of Electrical Engineering
Statement of Responsibility:
by Heming Wang.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
892558284 ( OCLC )
ocn892558284

Downloads

This item has the following downloads:


Full Text
NEW ELECTRODE MATERIALS AND ACTIVE ENERGY HARVESTING FOR
MICROBIAL ELECTROCHEMICAL SYSTEMS, OR MXCS
by
Heming Wang
B.S, Harbin Institute of Technology (China), 2006
M.S, Harbin Institute of Technology (China), 2008
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Civil Engineering
2013


This thesis for the Doctor of Philosophy degree by
Heming Wang
has been approved for the
Civil Engineering Program
by
Peter Jenkins, Chair
Zhiyong Jason Ren, Advisor
Jae-Do Park
JoAnn Silverstein
Angela Bielefeldt
April 18, 2013


Wang, Heming (Ph.D., Civil Engineering)
New Electrode Materials and Active Energy Harvesting for Microbial Electrochemical
Systems, or MXCs
Thesis directed by Assistant Professor Z. Jason Ren.
ABSTRACT
Microbial electrochemical system or MXC is an emerging platform technology
that integrates multiple disciplines and carries different functions, such as waste treatment,
environmental remediation, and desalination with simultaneous power or value-added
chemical production. The versatility and high performance distinguishes MXC from
traditional treatment-focused and energy-intensive environmental systems and makes it a
potential transformative technology for the next generation of environmental
biotechnology. The Chapter 1 of this dissertation summarizes nearly 50 corresponding
reactor systems developed based on the MXC platform, with the goal to introduce and
summarize all the functions that have been developed so far and discusses the niche of
this technology for environmental science and engineering.
The MXC technology carries great potentials in transforming waste treatment
process into energy and resource recovery systems, but the technology is still in early
stage development, and there are many remaining challenges need to be addressed. In
addition to comprehensive literature review, my doctoral study has been focused on the
following three aspects on MXC scientific characterization and engineering development:
(1) Development and characterization of high performing and low-cost electrode
materials. The anode and cathode are two crucial elements in MXC structures and largely
determine the performance and cost of a reactor system. The feasibility of using a
m


recycled material tire crumb rubber, as a low-cost anode material in microbial fuel cells
(MFC) was investigated (Chapter 2) in my first year of study. This is the first study that
used recycled materials as an alternative to traditional graphite or carbon cloth anodes,
the results show that tire crumbs produced comparable level of electricity after surface
treatment. On the cathode side, three types of new carbon nanotube based air-cathode
with very high surface area were developed, which created a 3-D structure and
demonstrated superior performance as compared to traditional carbon cloth air-cathodes
(Chapter 3).
(2) Innovative active energy harvesting during organic removal and waste
treatment. MXC can produce direct electricity from waste materials, but currently the
current and voltage output are low, making the direct use of such power difficult. A new
active energy harvesting approach using a maximum power points circuit (MPPC) was
developed, which totally changed the traditional passive energy gaining process through
resistors or non-controllable charge pumps. The system dramatically increased MFC
power production by more than 70 times and improved energy recovery by 20 times
(Chapter 4). Furthermore, the effects of inductance, duty ratio, and switching frequency
on MFC energy harvesting were characterized to optimize operating conditions and direct
further circuit development (Chapter 5).
(3) Removal mechanisms of emerging trace organic compounds (TOrCs) in
microbial fuel cells. The MXC can theoretically degrade any biodegradable materials to
produce energy, and many different kinds of waste streams have been tested. The
removal of traditional organic matter and sulfur, and the recovery of nitrogen and
phosphate have all been proved possible. The feasibility of removal and transformation of
IV


low level (e.g. ng/L) trace organic compounds (TOrCs) in wastewater was studied in
microbial fuel cells (Chapter 6), because it was hypothesized that MFC can be effective
in this task by providing both reductive (anode) and oxidative (cathode) environments,
and the biodegradation can be enhanced with electrochemical transformation. Results
show that MFC was effective in removing most of the 34 TOrCs tested with different
efficiencies, and electricity was produced from all reactors.
The form and content of this abstract are approved. I recommend its publication.
Approved: Z. Jason Ren
v


DEDICATION
I dedicate this doctoral dissertation to my husband, Tianhua Guo, and the Wang
and Guo families for their continuous love and support.
vi


ACKNOWLEDGMENTS
I would like to express my sincerest gratitude to my advisor, Dr. Zhiyong Jason Ren,
who offered me the opportunity to the Ph.D program and walked me through all the
difficulties in my study and research. I am deeply grateful of his support, guidance, and
encouragement, which not only helped me finish the doctoral study, but also will benefit
my professional development. I would also like to thank Dr. Jae-Do Park from Electrical
Engineering, for his great guidance and tremendous help in the energy harvesting project.
I am grateful to Dr. Zhuangchun Wu, Dr. Peter Jenkins, Dr. Atousa Plaseied, and Dr. Pei
Xu for their insightful and constructive discussions. I thank Dr. JoAnn Silverstein and Dr.
Angela Bielefeldt for serving in the committee and providing great advices for my
research and course work. My research was supported by the Office of Naval Research
(ONR) and US National Science Foundation (NSF).
I am grateful to all my committee members for their valuable participation and
helpful comments. My thanks would also go to all the members in the lab for their direct
and indirect assistance.
Finally, I also owe a special debt of gratitude to my family who are halfway
across the globe. I thank my parents for their unconditional support and love. I also owe
my deepest gratitude to my beloved husband who is always there waiting for me all these
years.
Vll


TABLE OF CONTENTS
CHAPTER
1. The X Factor: a Review of Bioelectrochemical System as a Platform Technology..1
1.1 Abstract......................................................................1
1.2 Introduction.................................................................2
1.3 The Shared Principle in the BES Anode Chamber................................3
1.4 The X Factor................................................................10
1.5 MFC-based Systems for Electricity Generation................................12
1.5.1 Wastewater microbial fuel cell (wastewater MFC)............................12
1.5.2 Benthic microbial fuel cell (benthic MFC)..................................14
1.5.3 Microbial remediation cell (MRC)...........................................15
1.5.4 Microbial solar cell (MSC).................................................17
1.6 MEC-based Systems for Chemical Production...................................19
1.7 MES-based Systems for Chemical Production...................................21
1.8 MDC-based Systems for Water Desalination and Beneficial Reuse...............22
1.9 Outlook.....................................................................24
1.10 Acknowledgement............................................................26
2. Recycled Tire Cmmb Rubber Anodes for Sustainable Power Production in Microbial
Fuel Cells.......................................................................27
2.1 Abstract....................................................................27
2.2 Introduction................................................................28
2.3 Materials and Methods.......................................................29
2.3.1 MFC construction and operation.............................................29
2.3.2 Statistical and electrochemical analyses...................................30
2.4 Results and Discussion......................................................32
viii


2.4.1 Resistance characterization of coated crumb rubber electrode...............32
2.4.2 Surface characterization of coated crumb rubber electrode...................33
2.4.3 Power production from tire rubber MFCs and graphite granule MFCs...........35
2.4.4 Cost-benefit outlook.......................................................36
2.5 Conclusions..................................................................37
2.6 Acknowledgements.............................................................37
3. Carbon Nanotube Modified Air-cathodes for Electricity Production in Microbial Fuel
Cells.............................................................................38
3.1 Abstract.....................................................................38
3.2 Introduction.................................................................39
3.3 Materials and Methods........................................................41
3.3.1 Cathode construction........................................................41
3.3.2 MFC construction and operation..............................................42
3.3.3 Electrochemical and Microscopy Analysis.....................................43
3.4 Results and Discussion.......................................................44
3.4.1 Electrochemical performance.................................................44
3.4.2 Performance of MFCs with nano-modified air-cathodes.........................46
3.4.3 FIB / SEM analysis..........................................................50
3.5 Conclusions..................................................................51
3.6 Acknowledgement..............................................................52
4. Active Energy Harvesting from Microbial Fuel Cells at the Maximum Power Point
without Using External Resistors..................................................53
4.1 Abstract.....................................................................53
4.2 Introduction.................................................................54
4.3 Materials and Methods........................................................56
4.3.1 MFC construction and operation..............................................56
IX


4.3.2 Maximum power point circuit (MPPC) design and operation....................57
4.3.3 Analyses....................................................................60
4.4 Results and Discussion........................................................61
4.4.1 MPPC can operate the MFC at the maximum power harvesting range.............61
4.4.2 MPPC harvests energy more actively and efficiently.........................63
4.4.3 The numbers of capacitors for energy storage...............................65
4.4.4 Conversion efficiency of the MPPC..........................................67
4.5 Outlook......................................................................70
4.6 Acknowledgement..............................................................71
5. Power Electronic Converters for Microbial Fuel Cell Energy Extraction: Effects of
Inductance, Duty Ratio, and Switching Frequency...................................72
5.1 Abstract.....................................................................72
5.2 Introduction.................................................................74
5.3 Materials and Methods........................................................76
5.3.1 MFC construction and operation..............................................76
5.3.2 Energy extraction circuit design............................................78
5.3.3 Tests.......................................................................79
5.4 Results and Discussion.......................................................81
5.4.1 Effects on MFC voltage and current..........................................81
5.4.2 Effects on MFC Energy and Efficiency........................................85
5.4.3 Discussion..................................................................88
5.5 Acknowledgement..............................................................89
6. Removal Mechanisms of Trace Organic Compounds in Microbial Fuel Cells.........90
6.1 Abstract.....................................................................90
6.2 Introduction.................................................................91
6.3 Materials and methods........................................................91
x


6.3.1 MFC construction and operation............................................92
6.3.2 Experimental Procedures...................................................93
6.3.3 Analysis..................................................................97
6.4 Results and Discussion......................................................97
6.4.1 Performance of Single-chamber and Two-chamber MFCs........................97
6.4.2 TOrCs removal in Single-chamber and Two-chamber MFCs......................100
References......................................................................105
xr


LIST OF TABLES
Table
1.1 Summary of All Types of BES/MXC.............................................4
3.1 List of Cathode Materials and Modifications Used in This Study and Their
Specifications..................................................................41
4.1 Analysis of Energy Extraction Efficiency by MPPC...........................69
6.1 34 TOrCs Detected by LC-MS/MS in ESI(+) and ESI(-) Methods and Selected
Physicochemical Properties......................................................94
xii


LIST OF FIGURES
Figure
1.1 Number of Published Journal Articles on MXCs Containing the Phrases Microbial
Fuel Cell, Microbial Electrolysis Cell, Microbial Desalination Cell or Microbial
Electrosynthesis (Source: Scopus on 3/8/2013; Document Type: Journal; Language:
English; Remove Duplicates from Searching
Results)............................................................................3
1.2 Basic Principles in Four Typical BESs (Left Chamber: Anode; Right Chamber:
Cathode). (A) Electricity Generation in Air-cathode Microbial Fuel Cell (MFC); (B)
Hydrogen Generation with External Power Supply in Microbial Electrolysis Cell (MEC);
(C) Chemical Production by Microbial Electrosynthesis (MES); (D) Middle Chamber
Desalination by Electric Drive in Microbial Desalination Cell
(MDC)..............................................................................11
1.3 MFC-based Systems for Electricity Generation: (A) Wastewater Microbial Fuel
Cell,5 (B) Benthic Microbial Fuel Cell,119 (C) Microbial Remediation Cell,126 and (D)
Microbial Solar Cell.111...........................................................13
1.4 Some Advanced MXC Systems: (A) Microbial Reverse-electrodialysis Electrolysis
Cell (MREC) for H2 Production,71 (B) Microbial Electrosynthesis (MES) for Organic
Synthesis,76 and (C) Microbial Capacitive Desalination Cell (MCDC) for
Desalination.86....................................................................20
2.1 Box Plot of Resistance Measurement and Statistics on Tire Crumb Particle Surface
with Different Coating Layers......................................................32
2.2 System Resistance of Single Chamber Bottle Reactor Filled with Graphite Granules
and Tire Particles with Different Coating
Layers.............................................................................33
2.3 Pore Size Distribution of (A) Rubber Particle with 4-layer Coating, and (B) Graphite
Granule as the MFC Anode...........................................................34
2.4 Voltage and Power Density as a Function of Current Density for Coated Tire Anode
MFCs and Graphite Granule Anode MFCs................................................
3.1 LSV Results (Current Density vs Potential) of Newly Modified Cathodes Before
Installing in MFCs. Current Density Range Was Marked Based On the Values Shown in
Figure 3.3................................................................................45
3.2 Voltage Generation as a Function of Time for the Different Cathodes
46
xiii


3.3 Power Density as a Function of Current Density (A) and Polarization Curves (B) for
MFCs Operated Using Different Air-cathodes
..............................................................................48
3.4 Comparison of LSV Electrochemical Test Results between New and Used Cathodes
of CC-Pt and SWNTn-Pt.........................................................49
3.5 SEM/FIB Images of New and Used Cathodes: (A) New CC-Pt, (B) Used CC-Pt
After MFC Operation, (C) New CNTM, (D) Used CNTM-Pt After MFC Operation, (E)
New SWNTn-Pt, and (F) Used SWNTn-Pt After MFC
Operation.................................................................51
4.1 Components in MPPC....................................................57
4.2 Block Diagram of the Maximum Power Point Circuit (MPPC): (A) Harvesting
Converter Controller. (B) Whole electric Circuit Diagram; (C) CHARGE Phase,
MOSFET is On While Diode is Off, Extracted Energy is Stored in the Inductor; (D)
DISCHARGE Phase, MOSFET is Off While Diode is On, Extracted Energy is Stored in
the Capacitors.....................................................................59
4.3 MFC Polarization Curve and Power Density Curve Obtained by Linear Sweep
Voltammetry (LSV). The Scan Rate of the Polarization was 0.1 mV/s.§: Operating Point
of the Charge Pump. A: Operating Range of the MPPC. Recirculating-flow MFC Open
Circuit Potential was 688 mV.......................................................61
4.4 Snapshot of On/Off Cycle of the MPPC During Active Energy Harvesting from
MFCs and the Voltage and Current Profiles. One division of X-Axis Represents 100
psec. The Figure Shows the Waveforms of 1 msec Duration in Terms of Current,
Voltage, and On/Off Switch Changes.................................................62
4.5 Comparison of MFC Voltage, Cathode Potential, and Anode Potential between the
MPPC Active Energy Harvesting Condition and 23 Ohm External Resistor Condition.
The Optimum External Resistance was Calculated to be 23 ohm Based on Polarization
Curve that Could Yield the Maximum Power
Density............................................................................64
4.6 (A) Comparison of Energy Harvesting by the MPPC and the Charge Pump and
Energy Stored in Capacitors. (B) Comparison of COD Removals in the MPPC and
Charge Pump Conditions. In the MPPC Test, 12 Capacitors were Connected in Parallel
for Energy Storage. In the Charge Pump Test, one Capacitor was Enough to Store All the
Harvested Energy from MFC..........................................................65
4.7 (A) Voltage Profile and (B) Energy Storage Differences by Using 3, 6, 9, and 12
Capacitors in Parallel During MPPC Active Energy Harvesting
66
4.8 Efficiencies through 18-hour Test.
68


4.9 Energy Conversion Efficiency and Distribution of Internal Energy Loss in the
MPPC. The Distribution was Quantified Based on an 18-hour, 12-capacitor
Operation..........................................................................69
5.1 Schematic Diagram of the Experimental Setup
...................................................................................77
5.2 Block Diagram of Energy Extraction Circuit: (a) Energy Harvesting Converter,
MOSFET is Controlled ON/OFF in Different Duty Ratios and Frequencies by Function
Generator; (b) CHARGE Mode, MOSFET is On and Switching Diode is Off, Energy
Extracted from MFC is Stored in the Inductor Temporally; (c) DISCHARGE Mode,
MOSFET is Off and Switching Diode is On, Energy Stored in the Inductor is Transferred
to the Capcitor....................................................................79
5.3 MFC Voltages (a)-(c) and MFC Current (d)-(f) During Energy Extraction under
Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle
Column: 14mH, and Right Column: 130mH................................................
5.4 Waveforms of MOSFET Gating Signal (Top), MFC Votlage (Middle), and MFC
Current (Bottom). MOSFET is Turned on When the Gating Signal is in High State, (a)
Inductance 0.45mH, Switching Frequency 100Hz, Duty Ratio 50%; (b) Inductance
14mH, Switching Frequency 1000 Hz, Duty Ratio
50%.........................................................................83
5.5 Anode Potentials (a)-(c) and Cathode Potentials (d)-(f) During Energy Extraction
under Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH,
Middle Column: 14mH, and Right Column:
130mH.......................................................................85
5.6 MFC Generated Energy (a)-(c), Harvested Energy (d)-(f), and Efficiencies (g)-(i)
During Energy Extraction Under Different Inductances, Duty Ratios and Frequencies.
Left Column: 0.45mH, Middle Column: 14mH, and Right Column:
130mH.......................................................................86
5.7 Waveforms of MFC Current (Top), MFC Votlage (Middle), and MOSFET Gating
Signal (Bottom). MOSFET is Turned on When the Gating Signal is in High State, (a)
Inductance 0.45 mH, Switching Frequency 1000 Hz, Duty Ratio 25%; (b) Inductance 130
mH, sSwitching Frequency 5000 Hz, Duty Ratio 25%; (c) Inductance 0.45mH, Switching
Frequency 100Hz, Duty Ratio 50%; (d) Inductance 14mH, Switching Frequency 500 Hz,
Duty Ratio 50%; (e) Inductance 0.45 mH, Switching Frequency 2000 Hz, Duty Ratio
75%; (f) Inductance 14 mH, Switching Frequency 5000 Hz, Duty Ratio
75%...........................................................................87
6.1 Polarization Curve and Power Density Curve Obtained by Linear Sweep
Voltammetry (LSV) in Single-chamber (A) and Two-chamber Reactors (B) Filled by
Sodium Acetate Spiked with TOrCs (Dark Blue) and without TOrCs (Orange). The Scan
Rate of the Polarization Was 0.1 mV/s.........................................98
xv


6.2 Total Organic Carbon (TOC) Removal in Single-chamber and Two-chamber
Reactors Filled by Sodium Acetate Spiked with TOrCs under 167 ohm and Open Circuit.
100
6.3 TOrCs Removal in Single-chamber Reactors and Cathode-chamber of Two-chamber
Reactors (A) and Anode-chamber of Two-chamber Reactors (B) Filled by Sodium
Acetate Spiked with TOrCs under 167 ohm and Open Circuit. TOrCs are Divided into
Three Categories Based on Charge. Biodegradibility Probability is Indicated in
Parentheses after the Name of Each Compound..................................103
xvi


1. The X Factor: a Review of Bioelectrochemical System as a Platform Technology1
1.1 Abstract
Bioelectrochemical system (BES) is an emerging technology that uses
microorganisms to covert the chemical energy stored in biodegradable materials to direct
electric current and chemicals. Compared to traditional treatment-focused, energy-
intensive environmental technologies, BES offers a new and transformative solution for
integrated waste treatment and energy and resource recovery, because it offers a flexible
platform for both oxidation and reduction reaction oriented processes. All BESs share
one common principle in the anode, in which biodegradable substrates, such as waste
materials, are oxidized and generate electrical current. In contrast, a great variety of
applications have been developed by utilizing this in situ current, such as direct power
generation (microbial fuel cell, MFC) or chemical production (microbial electrolysis cell,
MEC). This study provides a comprehensive review of the different functions developed
based on the BES platform to date and summarized nearly 50 corresponding systems as
MXCs with the X standing for the different functions and systems. It also discusses the
X factor the future development of this promising yet early-stage technology.
1 The work presented in this chapter is co-authored by Heming Wang and Zhiyong Ren
and in review by Energy Environ. Sci.
1


1.2 Introduction
Bioelectrochemical system (BES), microbial electrochemical system (MES), or
MXC are different collective names for an emerging environmental technology called
microbial electrochemical technology (MET).1'4 While this platform technology has only
been intensively studied and developed in the past decade, it opens up a new
interdisciplinary field for research and development which integrates microbiology,
electrochemistry, materials science, engineering, and many related areas together. BES
not only provides a unique environment to understand the largely unexplored microbial
electrochemistry, it also offers a flexible platform for many different engineering
functions to be developed. While many existing environmental technologies have only
one or two functions, the BES platform is so flexible that dozens of functions have been
discovered. Almost all BESs share one common principle in the anode, in which
biodegradable substrates, such as waste materials, are oxidized by microorganisms and
generate electrical current. The current can be captured directly for electricity generation
(microbial fuel cells, MFCs),5'7 or used to produce hydrogen, methane, and other value-
added chemicals (microbial electrolysis cells, MECs).8'10 The electrons can also be used
in the cathode chamber to synthesize organic compounds (microbial electrosynthesis,
MES) or remediate contaminants (microbial remediation cells, MRCs).11-15 The potential
across the electrodes can also drive desalination (microbial desalination cells, MDCs).16"
20 The production of current associated with microbial catabolism was first reported a
century ago by M. C. Potter,21 but research interests in this concept have only blossomed
in the past decade, resulting in an exponential growth in the number of journal articles
(Figure 1.1). There are several excellent reviews that provided information on the history
2


and development of BESs22-25 and the substrates, materials, and microbial communities in
BESs,26-30 but there has been no comprehensive review that directly addresses the factor
of X, where all the known functions were originated from and future functions will be
based upon. As shown in Table 1.1, this article aims to provide the first comprehensive
review to summarize all the X functions that have been developed using the BES
platform and shed lights on future system development for energy and environmental
science and engineering.
o l-H fN fO T ID VO 1"- CO ON O fN
o o O o O O o o o O l-H l-H
o o o o o o o o o O o o o
fN fN fN Year fN fN fN
Figure 1.1 Number of Published Journal Articles on MXCs Containing the Phrases
Microbial Fuel Cell, Microbial Electrolysis Cell, Microbial Desalination Cell
or Microbial Electrosynthesis (Source: Scopus on 3/8/2013; Document Type:
Journal; Language: English; Remove Duplicates from Searching Results).
1.3 The Shared Principle in the BES Anode Chamber
Compared to traditional chemical fuel cells, the BES platform uses low-cost and
self-sustaining microorganisms to oxidize organic and inorganic substrates, mainly waste
materials, and transfer electrons to the anode electrode. This microbial oxidation reaction
is a shared principle for almost all BES or MXC reactors, as shown in Table 1.1.
However, how to use these electrons on the cathode side gives the most beauty of this
3


platform technology, because any reduction-based reaction can be realized in the cathode
chamber which creates numerous possibilities. Based on the different functions, the BES
platform has been specified into many different names that researchers name them MXCs,
where X stands for the different applications.2,4 Table 1.1 summarizes all the Xs to date
and demonstrates the shared principle on the anode and the versatile functions on the
cathode.
Table 1.1 Summary of All Types of BES/MXC.
Types of Electron donor for Electron acceptor for Main Ref.
BES/MXC anode oxidization cathode reduction Products
MFC-based systems for electricity generation 31,32
Microbial Fuel Any biodegradable Oxygen, Potassium Electricity
Cell (MFC) in material ferricyanide, or other
general oxidants 33
Tubular Microbial Acetate, glucose, Potassium ferricyanide Electricity
Fuel Cells domestic wastewater,
(Tubular MFC) hospital wastewater, digester effluent from a potato processing plant 34,35
Upflow Microbial Sucrose Potassium Electricity
Fuel Cell (UMFC) ferricyanide, oxygen 36
Baffled Air- Glucose, liquid from Oxygen Electricity
cathode Microbial com stover steam
Fuel Cell explosion process
(BAFMFC) Stacked Microbial Sodium acetate Potassium ferricyanide Electricity 37
Fuel Cell (Stacked MFC) Submersible Microbial Fuel Domestic wastewater Oxygen Electricity 38
Cell (SBMFC) Benthic Microbial Fuel Cell (BMFC) Sediment Oxygen Electricity 39-41
4


Table 1.1 (cont)
Types of Electron donor for Electron acceptor for Main Ref.
BES/MXC anode oxidization cathode reduction Products
Sediment Acetate and other Oxygen Electricity 3U
Microbial Fuel fermentation products
Cell (AKA Benthic Unattended Generator or in the sediment
BUG) Self-stacked Submersible Microbial Fuel Sediment, acetate Oxygen Electricity 42
Cell (SSMFC) Microbial Diesel, ethanol, 1,2- Chlorinated solvents, Reduced/ 11-13,
Remediation Cell dichloroethane, perchlorate, non-toxic 43-48
(MRC) pyridine, phenol chromium, and chemicals
uranium
Photo-Microbial Water Potassium ferricyanide Electricity 49
Fuel cell (p-MFC) Microbial Marine sediment Oxygen Electricity, 50
Photoelectrochem glucose,
ical Solar Cell oxygen 51,52
Solar-powered Succinate, propionate Oxygen Electricity,
Microbial Fuel hydrogen
Cell Photobioelectroch Organic acids, Potassium ferricyanide Electricity, 53
emical Fuel Cell alcohols hydrogen 54
Photosynthetic Microbial Fuel Water Oxygen Electricity

Cells (PMFC) Photosynthetic Electrochemical Water, glucose Potassium ferricyanide Electricity 55
Cell Solar-driven Trypticase soy broth Proton Electricity 56
Microbial Photoelectrochem (TSB)
ical Cell (Solar MPC) Plant Microbial Plant-derived Oxygen, potassium Electricity 57
Fuel Cell (PMFC) organics (root exudates) ferricyanide 58
Phototrophic Sediment Oxygen Electricity
Microbial Fuel
Cells
(Phototrophic
MFC)
5


Table 1.1 (cont)
Types of Electron donor for Electron acceptor for Main Ref.
BES/MXC anode oxidization cathode reduction Products
Photosynthetic Algal Microbial Fuel Cell Algae Potassium ferricyanide Electricity
(PAMFC) Microbial Wastewater Oxygen Treated 60
Electrochemical wastewater,
Snorkel (MES, no
AKA short- circuited microbial fuel electricity
cell) Acid-mine Ferrous ion Oxygen Electricity, 61
drainage fuel cell removing
(AMD-FC) iron 62
Integrated Wastewater Oxygen Electricity,
photobioelectroch algal
emical system biomass
(IPB) Osmotic Sodium acetate Oxygen Diluted 63
Microbial Fuel draw
Cell (OsMFC) solution, electricity 64, 65
Microbial Reverse Electrodialysis Cell (MRC) Sodium acetate Oxygen Electricity

MEC-based systems for chemical production 8,66-
Microbial Any biodegradable Proton Hydrogen,
Electrolysis Cell material hydrogen 68
(MEC) in Peroxide,
general methane, sodium hydroxide 69
Bioelectro- Wastewater Proton Hydrogen
chemically assisted microbial
reactor (BEAMR) Solar-powered Microbial Electrolysis Fuel (Solar MEC) Acetate Proton Hydrogen 70


6


Table 1.1 tconf)
Types of Electron donor for Electron acceptor for Main Ref.
BES/MXC anode oxidization cathode reduction Products
Microbial Reverse- Acetate Proton Hydrogen Vi
electrodialysis Electrolysis Cells (MREC) Microbial Sodium acetate Proton Hydrogen, 72
Electrolysis Struvite- struvite
precipitation Cell (MESC) Submersible Microbial Electrolysis Cell (SMEC) Acetate Proton Hydrogen 73
MES-based systems for chemical production 14,
Microbial organic, poised Acetic acid or other Ethanol,
Electrosynthesis anode, hydrogen organics, carbon Acetate, 2- 74-78
(MES) in sulfide dioxide oxobutyrate,
general formate 79
Microbial Carbon Glucose Carbon dioxide Algal
Capture Cell biomass,
(MCC) electricity
MDC-based systems for water desalination and beneficial reuse 16
Microbial Any biodegradable Oxygen, potassium Desalinated
Desalination Cell material ferricyanide, organics, water
(MDC) in general or other oxidants 80
Microbial Sodium acetate Hydrogen Treated
Saline-wastewater saline
Electrolysis Cell wastewater,
(MSC) electricity 81,82
Osmotic MDC Sodium acetate, Oxygen, potassium Desalinated
(OsMDC, xylose, wastewater ferricyanide, proton water,
MODC) electricity 83
Microbial Sodium acetate Potassium ferricyanide Desalinated
Desalination Cell water
with capacitive
adsorption
capability
(cMDC)
7


Table 1.1 (cont)
Types of Electron donor for Electron acceptor for Main Ref.
BES/MXC anode oxidization cathode reduction Products
Microbial Sodium acetate Oxygen Desalinated u
Desalination Cell water,
packed with ion- exchange resin electricity
(R-MDC) Microbial Sodium acetate Proton Hydrogen, 19
Electrolysis desalinated
Desalination Cell water
(MEDC) Microbial Sodium acetate Oxygen Desalinated 85
Electrolysis water,
Desalination and sodium
Chemical- hydroxide,
production Cell hydrochlori
(MEDCC) c acid 86
Microbial Sodium acetate Oxygen Desalinated
Capacitive Desalination Cell water
(MCDC) Capacitive Sodium acetate Potassium ferricyanide Desalinated 87
Deionization water
coupled with Microbial Fuel
Cells (CDI-MFC) Upflow Microbial Sodium acetate Oxygen Desalinated 18
Desalination Cell water,
(UMDC) electricity 88
Stacked Microbial Sodium acetate Oxygen Desalinated
Desalination Cells water,
(SMDC) electricity 89
Recirculation Xylose Oxygen Desalinated
Microbial water,
Desalination Cell (rMDC) electricity
Ideal anodic reactions in BESs or MXCs generally include dynamic and effective
microbial activity and community, higher substrate conversion rate and electron transfer
efficiency, and lower material and system cost. The conversion of chemical energy to
electrical energy in the BES anode chamber requires the respiration of the insoluble
8


anode, where a unique group of microbes called electrochemically active bacteria (EAB),
exoelectrogen, electricigen, or anode respiring bacteria have been used.29,30,90,91 Such
microorganisms are able to transfer electrons out of cell membranes to the electrode
either directly through immobilized stmctures or using mobile electron shuttles. For
example, recent studies showed that Geobacter sulfurreducens requires conductive pili as
nanowires for cell-to-cell electron conduction and c-type cytochrome OmcZ to promote
electron transfer onto the electrode.92,93 In contrast, Shewanella species were reported to
make both direct electrode contact through conductive filaments and indirect electron
transfer via mediators, such as riboflavin or flavin ademine mononucleotide (FMN).94-96
Many other bacteria can produce and use soluble redox mediators or electron shuttles,
which transport the electrons from the cell to the electrode. For example, Pseudomonas
species can produce phenazines as extracellular electron shuttles, and other bacteria can
use externally provided mediators, such as neutral red, anthraquinone-2,6-disulfonate
(AQDS), thionine, methyl viologen, methyl blue, and some humics.13,97-101
Using microorganisms as biocatalysts, BESs or MXCs can theoretically be used
to convert any biodegradable substrate into energy and chemicals. Besides simple sugars
and derivatives used in most lab scale studies, many complex waste materials have also
been utilized, such as different wastewaters from municipal and industrial sources,
biomass waste, and inorganic substrates such as ammonia, sulfide, and acid mine
drainage.27,61,102-104 The utilization of complex waste materials requires the cooperation
of polymer-degrading bacteria and electrochemically active bacteria, with the first group
breaks down the complex organic matters into monomers or solvents, and the second
group oxidizes the fermentation products with the anode serving as the electron
9


acceptor.6,105-107 in terms of waste treatment in the anode chamber, BES represents a new
generation of technology, because it carries the potential to transform traditional energy-
intensive, treatment-focused processes into integrated systems that recover energy,
nutrient, water, and other value-added products.
1.4 The X Factor
As shown in Table 1.1, there have been nearly 50 systems with different functions
developed using the BES platform, and people used MXCs to represent the different
functions and systems. Though no specific rales have been established to name the Xs,
this article attempts for the first time to summarize and categorize all the Xs that have
been reported so far and provides some insights on the unknown X factor regarding to
technology development.
In general, an X simply presents the main function and benefit of a specific cell.
For example, microbial fuel cell (MFC) is the very original type BES or MXC, whose
main function is direct electricity generation (Figure 1.2A).108 When an external power
source is added in an MFC reactor to reduce cathode potential, the system becomes a
microbial electrolysis cell (MEC), where hydrogen gas, methane gas, and other products
can be generated through electrolysis (Figure 1.2B).8,9,67-69 If the main function of the
system is to use the cathode to reduce oxidized contaminants, such as uranium,
perchlorate or chlorinated solvents, the cell can be named microbial remediation cell
(MRC),11-13 and if the main goal of the system is to synthesize value-added bio-chemicals
through cathodic reduction, the system can be named microbial electrosynthesis (MES)
(Figure 1.2C).14,15 Another system called microbial desalination cell (MDC) (Figure
10


1.2D),16,71 includes additional chambers between the anode and cathode and uses the
internal potential to drive water desalination.
J Anode Bacteria
| JCathode Bacteria
V
oh2 oo2
OC02
AOrganics
Figure 1.2 Basic Principles in Four Typical BESs (Left Chamber: Anode; Right
Chamber: Cathode). (A) Electricity Generation in Air-cathode Microbial Fuel Cell
(MFC); (B) Hydrogen Generation with External Power Supply in Microbial
Electrolysis Cell (MEC); (C) Chemical Production by Microbial Electrosynthesis
(MES); (D) Middle Chamber Desalination by Electric Drive in Microbial
Desalination Cell (MDC).
There are also many different sub-systems within each main category. Take
MFCs as an example, based on different substrates used in MFC reactors, there are
wastewater MFCs, sediment or benthic MFCs, etc.109,110 By utilizing different
photosynthetic organisms for solar energy capturing, people have developed plant-MFC,
phototrophic-MFC, and algae-MFC.57,58,111 By integrating other technologies with the
BES platform, new systems with superior performance can be developed. For instance,
11


by incorporating reverse-electrodialysis (RED) with MEC, the microbial reverse-
electrodialysis electrolysis cell (MREC) can produce H2 without any external power
supply.71 By integrating capacitive deionization (CDI) with MDC, the microbial
capacitive desalination cell (MCDC) could improve desalination efficiency by 7-25
times.86 Other names may come from the combination of multiple functions in one
system, and they are generally straightforward, such as microbial electrolysis desalination
cell (MEDC),19 microbial electrolysis desalination and chemical-production cell
(MEDCC),85 osmotic microbial fuel cell (OsMFC),63 and microbial electrolysis struvite-
precipitation cell (MESC),72 etc.
1.5 MFC-based Systems for Electricity Generation
1.5.1 Wastewater microbial fuel cell (wastewater MFC)
MFC refers to the reactor systems that focus on electricity production from
biodegradable materials. Table 1.1 provides a complete list of different MFCs to date.
Early lab scale MFC studies mostly used acetate, glucose, or other simple substrates to
characterize the performance of materials, reactor configurations, or microbial
activities.112,113 The first MFC study that used real wastewater as substrate was reported
in 2004,109 and since then hundreds of studies have been published to report power
production from different substrates, including both organic and inorganic waste streams
using various electrode or separator materials and reactor configurations. Several review
articles have provided comprehensive information on the substrates,27 electrode
materials,28 separator materials,114 and reactor configurations108 used in different MFC
studies.
12


Classic MFC designs include the single-chamber air-cathode MFCs (SCMFCs)
developed by Liu et al, which in its cubic designs for the first time eliminated the
membrane and therefore significantly reduced system internal resistance and cost (Figure
1.3A).5,112 Tubular designs (Tubular MFC) with different flow patterns simplified
construction process and optimized systems with increased electrode surface area and
reduced system resistance.33,34 Baffled air-cathode microbial fuel cell (BAFMFC) was
designed to increase organic loading rate,36 and stacked MFCs were able to increase
direct voltage or current output while also enhance substrate oxidation.37 Other MFC
systems used in wastewater applications include submersible MFCs (SBMFC),38 which
may convert the information of substrate concentration, toxicity, or dissolved oxygen
concentration into electronic signals as MFC sensors.
Figure 1.3 MFC-based Systems for Electricity Generation: (A) Wastewater
Microbial Fuel Cell,5 (B) Benthic Microbial Fuel Cell,119 (C) Microbial Remediation
Cell,126 and (D) Microbial Solar Cell.111
13


The main advantages of using MFCs in wastewater treatment come from the
savings of aeration energy and sludge disposal.62,115,116 For traditional activated sludge
system, aeration can amount to 45-75% of plant energy costs, so the conversion of
aeration tank to MFC units is very beneficial because it not only eliminates aeration
energy consumption, studies also showed that the MFC can produce 10-20% more energy
that can be used for other processes.27,117 The reported maximum power density from lab
scale air-cathode MFCs has reached to 2.87 kW/m3, making it possible for
commercialization development.118 Another main benefit of MFC system is the low
biomass production. Because MFC is a biofilm based system, the cell yield of
electrochemically active bacteria (0.07-0.16 gVSS/gCOD) was much less than the
activated sludge (0.35-0.45gVSS/gCOD), so it can reduce sludge production by 50-
70%,117,118 which in turn may reduce 20-30% of the plant operation cost. Other benefits
may include nutrient removal and the production of value-added products, such as caustic
solutions for disinfection, or FF or biogas for energy, which will be discussed more
extensively in the following sections.
1.5.2 Benthic microbial fuel cell (benthic MFC)
Benthic MFC (BMFC), also known as sediment MFC (SMFC) is a system that
utilizes the naturally occurred potential difference between the anoxic sediment and oxic
seawater to produce electricity.30 Microorganisms oxidize the substrates in the sediment
and transfer electrons to the anode either embedded in or rested on top of the sediment,
and then the electrons are transferred to the cathode suspended in the overlying seawater,
where dissolved oxygen is reduced to water (Figure 1.3B).119 The abundant availability of
substrates in the sediment makes BMFC a very promising power source for autonomous
14


marine sensors and underwater vehicles, because they can provide consistent and
maintenance-free power supply for a long period of time without using batteries. This is a
huge advantage compared to batteries, because batteries are limited in service life for
about 2-4 years, and the replacement can be very expensive, especially in deep water. It
was estimated that the initial organic matters in 1 L marine sediment could generate an
average current of 0.3 mA continuously for 22 years.50 While the concept of BMFC was
only introduced in 2001 by Reimers et al.,110 it is a type of BES device that is closest
toward commercialization. The first demonstration of BMFC as a viable power source
was reported by Tender et al. in 2008, where an 18 mW meteorological buoy was
powered for nearly 7 months.40 Another study showed a chambered BMFC was used to
power an acoustic modem interfaced with an oceanographic sensor for over 50 days with
an average power density of 44 mW/m2.41 Different configurations of BMFCs have been
developed and deployed. Initial designs include simple graphite plates buried in the
sediment with suspended cathode in water, but such designs are fragile and the power
output is very low.120 Nielsen et al. developed a chamber-based BMFC that incorporates
a suspended and semi-enclosed anode, which reduced system footprint and increased
power output to a range of 380 mW/m2 (3.8 W/m3).39 A Self-stacked submersible
microbial fuel cell (SSMFC) showed an open circuit voltage (OCV) of 1.12 V and a
maximum power density of 294 mW/m2.42
1.5.3 Microbial remediation cell (MRC)
Another emerging application of the BES/MXC platform is using the electrodes
to serve as inexhaustible electron acceptors (anode) or donors (cathode) for underground
contaminant remediation.121'123 Fike sediment MFCs, the MRCs used in groundwater or
15


soil remediation can be a single or an array of electrodes without using enclosed
containers. Such bioelectrochemically enhanced approach can stimulate microbes to
concurrently degrade underground pollutants and produce additional electricity. Such
process is considered sustainable because it eliminates the injection of expensive
chemicals and reduces operational energy cost as compared to other technologies.
Microbial electrochemical remediation of petroleum contaminants was
demonstrated by using electrode as a channel linking underground hydrocarbon oxidation
and upground O2 reduction. One study showed that the active MRC increased the
degradation of diesel range organics (DRO) by 164% as compared to open circuit
potential,43 and another study using U-tube MFC showed crude oil degradation can be
increased by 120% at the location near the electrode.124 Similar remediation studies on
other reductive pollutants including diesel, ethanol, 1,2-dichloroethane, pyridine, and
other contaminants were also reported.45-47 Conversely, oxidative contaminants, such as
chlorinated solvents, perchlorate, chromium, and uranium, can be reduced using the
electrode as the electron donor.11-13,125 For instance, studies showed that a negatively
polarized electrode could act as an electron donor for the reductive dechlorination of
trichloroethene (TCE) to ethene by a mixed culture of microorganisms.13 Similar
approach was also used in both lab and field tests for U(VI) reduction, where the
horizontally distributed anodes and cathodes enabled direct correlation between acetate
injection and uranium reduction, and current production may be an effective proxy for
monitoring in situ microbial activity and remediation performance (Figure 1.3C).126
16


1.5.4 Microbial solar cell (MSC)
Microbial solar cell is a collective name for different BES/MXC systems that
integrate the photo synthetic reaction with microbial electricity (or chemical) production
using synergistic relationships between photosynthetic organisms and EAB.111 While the
EABs are generally the same bacterial groups in other MXC systems, the organisms that
are responsible to convert solar energy to organic matters may include higher plants,
photoautotrophic bacteria, and algae. A very wide variety of names and systems related
to MSCs have appeared in literature, such as photo-microbial fuel cell (p-MFC),49
microbial photoelectrochemical solar cell,50 solar-powered microbial fuel cell,52
photobioelectrochemical fuel cell,53 photosynthetic microbial fuel cells (PMFC),54
photosynthetic electrochemical cell,55 and solar-driven microbial photoelectrochemical
cell (solar MPC).56 Despite the variations in system designs, the basic principle of MSCs
usually include 4 steps, as described by Strik et al., and illustrated in Figure 1.3D, (i)
photosynthesis of organic matter; (ii) transport of organic matter to the anode
compartment; (iii) anodic oxidation of organic matter by EAB; and (iv) cathodic
reduction of oxygen or other electron acceptors.111 Here we categorize and discuss the
MSCs into 3 groups based on the organisms responsible for photosynthesis plant MSC,
phototrophic MSC, and algae MSC. More detailed information can be found in other
reviews.5758111
The most popular MSCs are plant MSCs, which use the organic rhizodeposits
excreted from living higher plants to feed EAB for electricity production. Reed
mannagrass and rice plants were used first to demonstrate the syntrophic relations, with a
maximal power output of 67 mW/m2 and 26 mW/m2, respectively.127128 Other plants
17


such Spartina anglica, Arundinella anomala and Arundo donax were also investigated for
concurrent electricity and biomass production. The A. donax failed,129 but S. anglica was
able to generate current for up to 119 days.130 Despite the low power output at current
stage, an European research consortium estimated that the power production from plant-
MSCs could reach 1000 GJ/ha/year (3.2 W/m2).111 Unlike plant MSCs, the phototrophic
MSCs do not require the cooperation between the two groups of microbes, because
studies showed that strains of photosynthetic bacteria such as Rhodobacter sphaeroides
can generate electricity through the metabolic activity of in situ oxidation of
photobiological hydrogen,53 and the power density can be comparable with
nonphotosynthetic MFCs.131 A self-assembling self-repairing marine sediment system
with photosynthetic microbes was reported to generate electricity from sunlight without
providing constant flux of glucose and oxygen.50 The algae MSC is an emerging system,
because the functions of algae and EAB are complementary. The consortium not only can
convert solar energy to electric energy, it can also remove nutrients and produce value-
added chemicals, such as protein and biodiesel. Both microalgae (Chlorella vulgaris) and
macroalgae (Ulva lactuca) have been used in algae MSCs to provide substrates for
EAB.102 In addition to traditional batch reactors, Strik et al. developed a flow through
photosynthetic algal microbial fuel cell (PAMFC) to automatically feed algae to MFCs.59
Another study integrated photobioreactor, anaerobic digester, and MFC reactors together
to recover both biogas and electricity.132 Other systems include recycling anode off gas
(CO2) into an algae grown cathode for additional carbon capture,79 and an integrated
photobioelectrochemical system with an MFC enclosed inside an algal bioreactor.62
Utilizing the algae, cyanobacteria and protozoa, Strik et al. reported an MSC with a
18


reversible bioelectrode that the electrode can function as a biocathode during illumination
for photosynthesis reaction and also can switch as the anode in the dark for organic
degradation.52 MSCs are the only BESs that do not rely on external electron donors but
convert inexhaustible solar energy into electrical energy and chemicals, so they carry
great potentials if current challenges such as low power output are addressed.
1.6 MEC-based Systems for Chemical Production
The concept of microbial electrolysis cell was originated in 2005, with the key
feature of using an external voltage on top of the MFC potential to enable hydrogen gas
evolution at the cathode through the reduction of protons.66,133 Early studies used
external power supplies ranged from 0.6-1.0 V to catalyze H2 evolution, which was much
lower than the 1.8-2.0 V used in traditional water electrolysis.9,66 Another advantage was
that the substrates can be from renewable and waste materials rather than fossil fuels, and
the H2 production rate can be more than 1 m3/d/m3 reactor with a yield upto 11 mol
H2/mol glucose, more than 3 times higher than dark fermentation.9,10 Several excellent
reviews summarized the material and system development of the MECs for H2
production.9,10,134
The elimination of membranes or separators converted dual chamber MECs to
single chamber reactors and significantly increased H2 generation rate, but the produced
H2 was more likely consumed by methanogenesis to generate CH4.9,10 Researchers have
tried different inhibition approaches such as adding expensive methanogen inhibitors,
periodically expose solution in aerobic environment, and control the pH and redox
potentials, but the CH4 contamination of H2 in single chamber MECs still remains a major
obstacle.9,10,135 The small external voltage can be supplied by MFC stacks or other
19


renewable power sources such as solar and wind.70136 Recently, reverse electrodialysis
(RED) was added into MEC generating a new system called microbial reverse-
electrodialysis electrolysis cells (MREC) with spontaneous EE production by combing
together the driving forces from anode organic oxidation and salinity gradient energy
(Figure 1.4A), and salt solutions could be continuously regenerated with waste heat (5?40
65, 71
Salt I
Water I
Anode Desalination Cathode
Chamber Chamber Chamber
Catiuii Cation" 4 Cation
Anion cn Anion g Anion
n
71 z
Anode
Chamber
Rinse
Water
T^7
n
Desalination
Chamber
Cation
4
Anion
Cathode
Chamber
Treated Desalinated Salt
Waste Water
Water (DESALINATION)
Collected Salts
(REGENERATION)
Figure 1.4 Some Advanced MXC Systems: (A) Microbial Reverse-electrodialysis
Electrolysis Cell (MREC) for H2 Production,71 (B) Microbial Electrosynthesis (MES)
for Organic Synthesis,76 and (C) Microbial Capacitive Desalination Cell (MCDC)
0/
for Desalination.
20


By using similar strategies in MECs, other inorganic chemicals have been
produced in the cathode chamber. Cusick and Logan discovered that phosphate can be
recovered as struvite (MgNfEPCE-OfEO) in a modified microbial electrolysis struvite-
precipitation cell (MESC).72 Rozendal et al. reported that hydrogen peroxide can be
produced by reducing oxygen through the two electron reduction, and the proof-of-
concept study showed at an applied voltage of 0.5 V, H2O2 can be generated at a rate of
1.9 0.2 kg H202/m3/day1 at a concentration of 0.13 0.01 wt% with an overall
efficiency of 83.1 4.8%.67 The same group later used a similar approach to produce
alkaline solutions, as they found that by using acetate as the electron donor in the anode,
the MEC generated up to 1.05A in current at 1.77 V applied voltage, which allowed for
the production of caustic to 3.4 wt%.68 Such chemicals can be produced during
wastewater treatment process and then used as low-cost disinfectants for many industries.
1.7 MES-based Systems for Chemical Production
Microbial electrosynthesis is an emerging area in BES research and development,
and it uses the electrons derived from the cathode to reduce carbon dioxide and other
chemicals into a variety of organic compounds, especially those with multiple carbons
that are precursors for desirable value-added chemicals or liquid transportation fuels.14,15,
74 The potential of MES not only comes from the double benefits of carbon sequestration
and organic production, it may also address the harvesting, storage, and distribution
problems associated with energy crops, solar and wind farms, and natural gas exploration,
because the electrons can be from any renewable source, and microbes may harvest solar
energy in a 100-fold higher efficiency than biomass-based chemical production.
21


The concept of microbial electrosynthesis was only introduced in 2009-2010, with
the initial findings associated with methane generation from a reactor with an abiotic
anode and a biocathode acclimated with Methanobacterium palustre,8 Another early
study demonstrated that biofilms of Sporomusa ovata could use the electrons supplied by
the cathode to reduce carbon dioxide into acetate and small amounts of 2-oxobutyrate.
Electrons appearing in these products accounted for over 85% of the electrons consumed
(Figure 1.4B).76 In general, acetogenic bacteria use hydrogen as the electron donor for
carbon dioxide reduction, but it was found that many acetogenic bacteria, such as
Clostridium ljungdahlii, Clostridium aceticum, Sporomusa sphaeroides, and Moorella
thermoacetica were all able to consume electrical current and produce organic acids.77
Studies also show that ethanol can be produced by reducing acetate at the cathode, but
some processes required the addition of mediators, such as methyl viologen (MV).75
Mixed culture originated from brewery wastewater were reported to generate methane,
acetate, and hydrogen gas from a biocathode poised at -590 mV (vs SHE) with CO2 as
the only carbon source,137 and research on genetically modified microorganisms may
significantly facilitate electron uptake and organic synthesis. As discussed in several
conceptual review articles, the microbial electrosynthesis carries great potentials, but
there are also many technological and economic challenges to be solved before it can be
implemented in large scale.14,15,74
1.8 MDC-based Systems for Water Desalination and Beneficial Reuse
Water desalination using the MDC process was first introduced in 2009 by Cao, et
al, and the proof-of-concept study was selected as the top technology paper by
Environmental Science & Technology.16 The basic principle of MDC is to utilize the
22


electric potential generated across the anode and cathode to drive desalination in situ.
Compare to other MXCs, MDC has a third chamber for desalination by inserting an anion
exchange membrane (AEM) and a cation exchange membrane (CEM) in between the
anode and cathode chambers. When bacteria in the anode chamber oxidize biodegradable
substrates and produce current and protons, the anions (e.g., Cl ) in the middle chamber
migrate to the anode and the cations (e.g., Na+) are drawn to the cathode for charge
balance, thus the middle chamber solution is desalinated.16,20 Recently, other approaches
were developed to achieve desalination as well. For example, by switching the CEM to
the anode side and AEM to the cathode side, microbial saline-wastewater electrolysis cell
(MSC) desalinates anolyte and catholyte by driving salts into the middle chamber.80
Osmotic microbial fuel cell (OsMFC) or osmotic MDC (OsMDC, MODC) uses a
forward osmosis membrane to replace the AEM and withdraw pure water from
wastewater to the draw solution, and then water can be recovered during draw solution
regeneration.63,82 Capacitive microbial desalination cell (cMDC) incorporates capacitive
deionization into MDC to improve desalination efficiency.83,86,87 In addition to
desalination, acid (HC1) and base (NaOH) solutions can be produced if a bipolar
membrane is placed into the MDC next to the anode chamber, creating a four-chamber
system called microbial electrolysis desalination and chemical-production cell
(MEDCC).85
The MDC can be used either stand-alone for simultaneous organic and salt
removal with energy production or serve as a pretreatment for conventional desalination
processes such as reverse osmosis (RO) to reduce the feed solution salt concentration,
and minimize energy consumption and membrane fouling. Compared with current
23


technologies that use 6-68 kWh to desalinate 1 m3 of seawater, MDC studies showed that
180-231% more energy can be recovered as H2 than the reactor energy input when
desalinating 5-20 g/L1 NaCl solutions,1719 and it was estimated that MDC may produce
upto 58% of the electrical energy required by downstream RO systems.18 Higher
desalination efficiency and current output can be achieved through membrane stacks,88
138 and electrolyte recirculation was shown effective in stabilizing electrolyte pH.89,139
Traditional MDC designs accomplish desalination by transporting ions from the middle
chamber to the anode and cathode chambers, which increases the conductivity of the
anolyte and catholyte. This change has been shown beneficial to electricity generation
due to improved mass transfer, but the increased salinity may also affect effluent water
quality and prevent subsequent beneficial use of treated wastewater.20 One solution for
complete salt removal from all solutions may involve the physical and electrical
adsorption of ions onto high surface area membrane electrode assemblies, such as
microbial capacitive desalination cell (MCDC), which showed upto 25 times of increase
in salt removal and complete salt recovery (Figure 1.4C).86 Similar as many membrane
based technologies, one challenge for MDC may come from membrane fouling due to
biofilm growth and scaling due to the deposition of hardness-causing cations, but studies
on understanding and addressing such problems are just getting started, and solutions
remain to be found.139140
1.9 Outlook
In one decade of research and development, the functionality of BES/MXC has
expanded dramatically and the performance has improved exponentially. Taking MFC as
an example, the power density has increased by orders of magnitude, from less than 1
24


mW/m3 to 2.87 kW/m3 (or 10.9 kA/m3),118 and the projected wastewater treatment
capacity of MFC can reach to 7.1 kg chemical oxygen demand (COD)/m3 reactor
volume/day, which is even higher than conventional activated sludge systems (-0.5-2 kg
COD/m3 reactor volume/day).3 However, there are still remaining challenges that need to
be addressed before the technology can be applied in commercial scale. Despite the
elimination of expensive metal catalysts and membranes, the overall cost of MXCs is still
considered expensive for wastewater treatment, unless an estimated threshold of internal
resistance <40 mO m2 in combination of a current density around 25 A/m2 can be
reached.25 Most studies are still limited in lab scale, and several pilot scale plants with
capacities between 20-1000 liters have yet shown stable and high enough performance
due to the problems of water leaking, low power output, influent fluctuation, and
unfavorable products.141143 To achieve practical implementation, BESs will need to be
scaled-up to at least cubic meter scale, the reactor configurations have to be easily
integrated with current infrastructure, and effectively harvesting systems instead of
resistors have to be developed to deliver usable power.144,145 Multiple reviews have
summarized the progresses of MFC system development and provided insights in further
directions.3,28,92,142
Compared to electricity generation in MFCs, chemical production and
desalination from BESs have been considered technically and economically more feasible
due to the higher price of chemical and relatively simple collection process. But such
processes are relatively new and mainly in lab scale, and there has been few reports in
scale-ups.141,143 Among the many different functions developed using this BES/MXC
platform technology, as discussed across this article, it is not clear where the BES can
25


contribute the most to the current environmental infrastructure and chemical industry.
There have been very limited evaluations of BESs/MXCs regarding to their life cycle in
terms of function selections or comparisons with established technologies which they can
complement.146,147 It has been assumed that the most environmental benefits from BESs
come from the displacement of fossil fuel dependent resources (i.e. grid electricity, or
chemical manufacture) through co-product production (i.e. electricity, chemicals) from
renewable sources, but the energy and environmental footprints of BESs have to be
clearly quantified before implementing large scale applications. Despite the remaining
challenges, BES/MXC has been widely considered the next generation of platform
technologies that will provide the multidisciplinary X factor for energy and
environmental sustainability.
1.10 Acknowledgement
This work was supported by the US National Science Foundation under Award
CBET-1235848.
26


2. Recycled Tire Crumb Rubber Anodes for Sustainable Power Production in
Microbial Fuel Cells2
2.1 Abstract
One of the greatest challenges facing microbial fuel cells (MFCs) in large scale
applications is the high cost of electrode material. We demonstrate here that recycled tire
crumb rubber coated with graphite paint can be used instead of fine carbon materials as
the MFC anode. The tire particles showed satisfactory conductivity after 2 to 4 layers of
coating. The specific surface area of the coated rubber was over an order of magnitude
greater than similar sized graphite granules. Power production in single chamber tire-
anode air-cathode MFCs reached a maximum power density of 421 mW/m2, with a
coulombic efficiency (CE) of 25.1%. The control graphite granule MFC achieved higher
power density (528 mW/m2) but lower CE (15.6%). The light weight of tire particle could
reduce clogging and maintenance cost but posts challenges in conductive connection. The
use of recycled material as the MFC anodes brings a new perspective to MFC design and
application and carries significant economic and environmental benefit potentials.
Keywords: tire crumb rubber, microbial fuel cell, anode, electricity
2 The work presented in this chapter has been published by Heming Wang, Matthew
Davidson, Yi Zuo, and Zhiyong Ren in J. Power Sources, 2011, 196, 5863-5866.
27


2.2 Introduction
Microbial fuel cells (MFCs) are an emerging bioelectrochemical technology that
produces electrical energy from organic matter catalyzed by exoelectrogenic bacteria on
the anode.148-150 In less than a decade, researchers have increased power densities by
several orders of magnitude, from mW/m3 to kW/m3.143,150 These increases have come
primarily from addressing the physical and chemical constraints on MFC performance by
exploring new materials and optimizing reactor architectures. A remaining challenge for
MFCs as they become more technically feasible for full-scale applications such as
simultaneous wastewater treatment and bioenergy recovery is the high cost of electrode
material currently used in lab scale studies. Many anode materials have been tested to
improve biofilm attachment and conductivity. The popular materials include graphite
granules,33 carbon paper,44 carbon cloth,151 carbon mesh,148 and activated carbon.35 The
recent development of graphite brush anodes with high specific surface area and an open
structure to prevent fouling problems provides a solution for scaling up.152 However, the
cost of most electrode materials, from ~$50/m2 to over $l,000/m2, is prohibitive to use in
large scale.143,148
We investigated the performance of a recycled material crumb rubber (granular
particles produced by grinding waste tires) as a potential inexpensive and abundant
alternative material for MFCs. The Rubber Manufactures Association (RMA) estimated
that 303.2 million scrapped tires were produced in the U.S. in 2007; approximately one
discarded tire per person per year.153 In addition, more than 300 million tires are currently
stockpiled throughout the country due to the lack of end-use markets. These stockpiles
pose great environmental, safety, and health concerns. The materials are fire hazards,
28


non-biodegradable, and occupy significant landfill space. Current disposal solutions
include incineration for tire derived fuels, reuse of crumb rubber as surfaces for
playgrounds and sports fields, and reuse of tire rubber in asphaltic concrete mixtures.154
Recently Tang et al. developed a new crumb rubber filtration system to treat wastewater
and ship ballast water.155,156 They found that crumb rubber filters significantly reduced
clogging compared to sand filters without compromising pollutant removal. Tire derived
rubber particles also showed better organic adsorption capacity than sand particles, as
well as exhibited good performance as high surface area, non-toxic media for biofilm
attachment in bioreactors.157
In this study, we tested for the first time the feasibility of using crumb rubber with
a conductive graphite coating as the anode material for electricity production in MFCs,
and compared its performance with graphite granule anodes. Statistical and
electrochemical analyses were conducted to evaluate the performance of the coated
material in terms of conductivity, resistances, and specific surface area. The potential
benefits of using crumb rubber as MFC electrode was also discussed.
2.3 Materials and Methods
2.3.1 MFC construction and operation
Single-chamber MFCs were constructed from Wheaton graduated media bottles
(250 mL, Wheaton, NJ) by adding one glass extension tube on the side.152,158 Rubber top
caps were used to provide an air-tight condition. Air cathodes (projected area of 4.5 cm2,
one side) were made by applying Pt/C (0.5 mg/cm2) and four PTFE diffusion layers on 30%
wet-proofed carbon cloth (Fuel Cell Earth, MA, USA) as previously described.159
Recycled tire crumb rubber was donated by AcuGreen Inc. (CO, USA). The crumb
29


rubber was pre-shredded from recycled tires and sieved to collect particles with 4-8 mm
diameter. The rubber particles were washed with deionized water and air dried before the
application of the conductive coating (E-34, Superior Graphite Co. OH, USA) to the
particle surface.160 To determine the optimal coating condition, additional coatings were
applied to some of the particles after the previous coating was completely dried in air.
Coated crumb rubber was packed into the MFC anode chamber to a volume of 140 mL
(71.5 g). A twisted titanium wire was inserted into the anode pack as a current collector
and connected to the external circuit. The same volume (140 mL, 133.6 g) of graphite
granules (D=2-6 mm, Graphite Sales Inc. OH) were used as the control anode material in
a separate MFC.
MFCs were inoculated with anaerobic sludge obtained from the Englewood-
Littleton Wastewater Treatment Plant (Englewood, CO). The reactors were fed with 190
mL medium containing: 1.25 g/L of sodium acetate, 0.31g/L of NH4CI, 0.13 g/L of KC1,
3.321 g/L of NaH2P04-2H20, 10.317 g/L of Na2HP04- 12H20, 12.5 mL/L of mineral
solution and 5 mL/L of vitamin solution.158 All MFCs were operated in fed-batch mode at
room temperature. Growth media was replaced with fresh media when the voltage
dropped below 50 mV (1000 Q resistance).
2.3.2 Statistical and electrochemical analyses
The optimal number of coatings on the crumb rubber was determined by
resistance measurement and statistical analysis. After each coat was finished and
completely dried in air, 35 coated rubber samples were randomly selected and the ohmic
resistance across a 4 mm distance was measured repeatedly by a programmable
multimeter. The t distribution was used to calculate confidence intervals (CIs) for the
30


mean changes between the resistances on two adjacent coatings. The 95% CIs were
calculated by X 2.032 s/4n where X = ^ Xt n was sample mean, S =
i=i /
was sample standard variation, and n was the sample size.
MFC cell voltage was continuously monitored using a data acquisition system
(Keithley Instruments, OH). The circuits were operated under a fixed load of 1000 Q.
During the stable power production stage of each batch experiment polarization
measurements were made using a variable resistor box (50 to 50 kQ). Current (I = V/R),
power (P = IV), and coulombic efficiency (CE, based on COD) were calculated as
previously described.6 Electrochemical impedance spectroscopy (EIS) tests were
conducted using a Potentiostat (PC 4/300, Gamry Instruments, NJ, USA) to measure the
internal resistances with the anode as the working electrode, and the cathode as the
counter electrode and reference electrode. The scan range was from 105 Hz to 0.005 Hz
with a small sinusoidal perturbation of 10 mV.
Specific surface area and pore size distribution of the particles were estimated by
Brunauer-Emmett-Teller (BET) method using a five-point N2 gas adsorption technique
(ASAP 2020; Micromeritics, Norcross, GA).161 The average pore size and pore size
distribution were determined from desorption of N2 according to the method developed
by Barrett, Joyner, and Halenda (BJH).162
2><-x)2
31


2.4 Results and Discussion
2.4.1 Resistance characterization of coated crumb rubber electrode
Ohmic resistance (Q/mm) K> ON OO O K> 20000000 > 8 6 4 2 0 t
X , , .. ^ ¥ 1 x 1 1 M
2-layer 3-layer 4-layer 5 -layer Z
u 1-layer 2-layer 3-layer 4-layer 5-layer
Maximum (Q/mm) 131 6 6.2 4.8 4.9
Minimum (Q/mm) 6.5 2.0 1.7 1.4 1.3
Median (Q/mm) 16 3.5 2.8 2.1 2.2
95%CI (Q/mm) 22.0+8.1 3.5+0.3 3.0+0.4 2.5+0.3 2.3+0.2
n=35, tO.025,34=2. 332
Figure 2.1 Box Plot of Resistance Measurement and Statistics on Tire Crumb
Particle Surface with Different Coating Layers.
To convert the almost non-conductive crumb rubber into electrically conductive
electrode, multiple layers of graphite paint were applied to the rubber particle surface.
Such approach has been successfully applied on ultrafiltration membrane MFC cathodes,
as described previously.160 Figure 2.1 is the box plot showing the statistics of ohmic
resistance variations of the 35 randomly selected particles coated with different layers of
graphite paint. It appears that additional coatings reduced the particle surface ohmic
resistance and heterogeneity. After the first coating, a particle has an average ohmic
resistance of 22.0 Q/mm, but the numbers across the 35 samples varied significantly,
from 6.5 to 131 Q/mm, resulting in a huge standard deviation. The second layer of
coating reduced the ohmic resistance by a factor of 6, to an average of 3.5 Q/mm.The
32


variation was also considerably reduced. Additional coatings showed only minor
improvements: the difference between 4 coatings and 5 coatings was not statistically
significant at the 5% level (Figure 2.1). In order to conserve coating material and reduce
cost, the rubber particles with 4 layers of coating was used in further MFC and
electrochemical characterizations. The average ohmic resistance of the particles was 2.5
Q/mm, ~10 times greater than graphite granules. EIS shows a similar trend in reactor
ohmic resistance in coated crumb rubber or graphite granules MFC reactors (Figure 2.2).
The system resistance decreased along with additional coatings. The average system
resistance of the MFC with 4-layer coated mbber was 574 Q, and the resistance using
same volume of graphite granule was 210 Q.
0 200 400 600 800 1000 1200
Zreai (ohm)
Figure 2.2 System Resistance of Single Chamber Bottle Reactor Filled with
Graphite Granules and Tire Particles with Different Coating Layers.
2.4.2 Surface characterization of coated crumb rubber electrode
High specific surface area is a crucial parameter of the MFC anode. It allows
higher biofilm density and thus making higher current output possible. Surface
characterization shows that the crumb mbber particle with 4-layer coating has an average
33


BET surface area of 4.5 m2/g (32,143 m2/m3), more than one order of magnitude greater
than the graphite granule (0.3 m2/g, or 2,143 m2/m3). The BJH desorption cumulative
pore volume of the coated rubber particle was 0.013 cm3/g and BJH desorption average
pore diameter was 88 A, also significantly higher than the graphite granule in terms of the
same parameters. The granule has a BJH desorption cumulative pore volume of 0.0006
cm3/g and BJH desorption average pore diameter of 54 A. Figure 2.3 compares the pore
size distribution of the 4-layer coated tire particle and graphite granule. The incremental
pore area and cumulative pore area of the tire particle are each one order of magnitude
greater than those of the graphite granule. Specifically, the desorption cumulative pore
area of tire particle and graphite granule were 5.78 cm2/g and 0.45 cm2/g, respectively.
Figure 2.3 Pore Size Distribution of (A) Rubber Particle with 4-layer Coating, and
(B) Graphite Granule as the MFC Anode.
34


2.4.3 Power production from tire rubber MFCs and graphite granule MFCs
Repeatable cycles of power production were obtained from both tire rubber and
graphite granule MFCs after 3-4 feeding cycles with fresh media. The stable voltages
over a 1000 Q external resistor in tire and graphite MFCs were around 390 mV and 430
mV, respectively. The tire reactor generally showed longer batch durations than the
graphite MFC. A regular batch cycle for graphite MFCs took around 20 days before the
media change, while a batch cycle for tire MFCs took about 30 days. Polarization and
power density curves obtained by varying the external circuit resistances from 50 to
50,000 Q showed that the crumb rubber MFC produced less power than graphite granule
MFC. Figure 2.4 shows that the maximum power density of the 4-layer coating tire MFC
was 421 mW/m2 (cathode projected area), -20% less than the power density of the
graphite MFC (528 mW/m2). The COD removal of the tire reactor (85.0 %) within a
batch was less than that from the graphite MFC (92.8%), but the columbic efficiency
obtained from the tire reactor (25.1%) was nearly one and half times higher than that
calculated from the graphite granule MFC (15.6%).
-- 750
-- 600
-- 450
-- 300
s
tu
Q
&
150 £
Figure 2.4 Voltage and Power Density as a Function of Current Density for Coated
Tire Anode MFCs and Graphite Granule Anode MFCs.
35


The difference in power production from the two types of reactors is a result of
several factors. It was noted that the specific area of the coated rubber particle was much
higher than the graphite granule, which could result in higher attachment and current
output, but the high ohmic resistance of the coated tire particle outweighed the benefit of
surface area and caused lower power generation. The conductive coating only allows the
electrons transfer across the surface of the tire particle rather than through the diameter of
the granule, which increased the length of the transfer route. Additionally, the density of
the crumb rubber is about 1.1 g/cm3, only a little greater than water but much less than
the density of the graphite (2.2 g/cm3). The low density of crumb rubber media has the
benefits with reduced maintenance cost and clogging potential,155 but it results in a loose
packing that hinders conductive connection. The integration of metal current collectors
into the anode pack could alleviate the problem in larger scale systems by compacting tire
particles and generating a highly conductive network for more efficient electron
transfers.163
2.4.4 Cost-benefit outlook
The use of recycled tire crumb rubber instead of expensive carbon products as
MFC electrode material is believed to carry economical and environmental benefits.
Compared to the high cost of refined carbon electrode ($50 to over $l,000/m2), the crumb
rubber is free except for the minimal cost of the coating material. Our preliminary cost
calculation shows the cost of coated tire electrode is $0.71/m2 and $1.42/m2 for 2-layer
coating and 4-layer coating, respectively, which is comparable to graphite granules
($1.29/m2). Moreover, many countries currently have tire disposal tax ($1 $3 per tire)
and reuse subsidy programs ($0.1-$0.5 per tire) to encourage tire recycle and reuse.
36


These policies make the use of crumb rubber more economically competitive and can
reduce the cost of rubber electrode by another $0.20-0.40/m2.164 In addition, government
regulations and higher public expectation on waste recycle and renewable energy
production make the technology more attractive to industries, as the use of recycled tire
rubber in MFCs will reduce the cost of tire disposal and bring more environmental
benefits by increasing tire reuse, treating wastewater, and generate alternative energy.
2.5 Conclusions
Recycled tire crumb rubber was tested for the first time as an alternative electrode
material in microbial fuel cells. The tire particles showed good conductivity after 2-4
layers of graphite coating. The specific surface area of the coated tire particle was more
than 10 times greater than similarly sized graphite granules and provides improved
attachment surface for microbes. The single chamber air-cathode tire MFC produced the
same level of power, COD removal, and coulombic efficiency as the graphite granule
MFCs did. The concept of using recycled material as MFC electrodes opens up a whole
new approach toward MFC design and application that carries significant economic and
environmental benefits.
2.6 Acknowledgements
This work was supported by the Office of Naval Research (ONR) under Awards
N000140910944. We thank Dr. Peter Jenkins and Dr. Pei Xu for valuable discussion and
Superior Graphite Co. for donating graphite coating materials.
37


3. Carbon Nanotube Modified Air-cathodes for Electricity Production in Microbial
Fuel Cells3
3.1 Abstract
The use of air-cathodes in microbial fuel cells (MFCs) has been considered
sustainable for large scale applications, but the performance of most current designs is
limited by the low efficiency of the three-phase oxygen reduction on the cathode surface.
In this study we developed carbon nanotube (CNT) modified air-cathodes to create a 3-D
electrode network for increasing surface area, supporting more efficient catalytic reaction,
and reducing the kinetic resistance. Compared with traditional carbon cloth cathodes, all
nanotube modified cathodes showed higher performance in electrochemical response and
power generation in MFCs. Reactors using carbon nanotube mat cathodes showed the
maximum power density of 329 mW/m2; more than twice that of the peak power obtained
with carbon cloth cathodes (151 mW/m2). The addition of Pt catalysts significantly
increased the current densities of all cathodes, with the maximum power density obtained
using the Pt/carbon nanotube mat cathode at 1118 mW/m2. The stable maximum power
density obtained from other nanotube coated cathodes varied from 174 mW/m2 to 522
mW/m2. Scanning electron micrographs showed the presence of conductive carbon
nanotube networks on the CNT modified cathodes that provide more efficient oxygen
reduction.
Keywords: microbial fuel cell, carbon nanotube, cathode, electricity
3 The work presented in this chapter has been published by Heming Wang, Zhuangchun
Wu, Atousa Plaseied, Peter Jenkins, Lin Simpson, Chaiwat Engtrakul, and Zhiyong Ren
in J. Power Sources, 2011, 196, 7465-7469.
38


3.2 Introduction
Microbial fuel cells (MFCs) are renewable energy systems that employ bacteria to
convert chemical energy stored in biodegradable materials to electrical energy. The MFC
technology carries great potential as it concurrently removes pollutants from the
environment and produces energy. An MFC reactor generally consists of an anode, a
cathode, and sometimes a separator between the two electrodes.149 In order to provide
good access for bacteria and improve power generation, the electrodes in MFCs need to
have high surface area, high conductivity, and be resistant to physical and chemical
corrosion. Many anode materials have been tested, including woven graphite mat,33
carbon paper,158 carbon cloth,151 and activated carbon.35 A recent development of
graphite brash anodes provides a solution for scaling up because it has very high specific
surface area (7170 18200 m2/m3) and an open structure to prevent fouling problems.152
Moreover, this makes the anode no longer the main limitation on power production, but
instead brings up the challenge of effective cathode development.165,166
Even though ferricyanide or permanganate can provide a higher cathode open
circuit potential, oxygen is considered as the electron acceptor for eventual MFC
applications due to its availability and high redox potential (E=1.23 V).166,167 However,
the tri-phase reaction among oxygen gas, solid catalyst, and liquid electrolyte does not
make satisfactory reaction kinetics. Moreover, the specific surface area (-100 m2/m3) of
current popular air-cathode designs are orders of magnitude lower compared with the
brash anode, making the cathode the primary limiting factor for improved power
production in MFCs. Previous results showed that MFC power output improved along
with the increase of cathode specific surface area. For example, Deng et al., found the
39


power density increased from 67 mW/m2 to 315 mW/m2 by replacing carbon paper
cathodes with high specific surface area activated carbon felt cathodes.168 Cheng and
Logan recently demonstrated that the volumetric power density has a linear relationship
with the cathode specific surface area.169
Carbon nanotubes (CNTs) have exhibited great potential as electrode materials in
fuel cell applications due to their high surface-to-volume ratio and unique electrical and
mechanical properties. Several studies investigated the performance of CNT-modified
anodes in MFCs and found that the anode biofilm activity was not affected by the carbon
nanotubes but the power density was improved.170-172 Though the CNT-modified anodes
do not provide direct large enough macro-scale porosity for more microbial colonization,
the highly conductive nanotube network serves as nanowires to facilitate the electron
transfer between the microbes to the electrodes. The reported power densities obtained
from CNT-modified anodes varied from 22 mW/m2 (multi-walled CNT)173 to 1098
mW/m2 (3-D CNT-texile).171 However, to date little effort has been reported on CNT-
modified cathodes, even though the benefits can be more significant because the CNT-
modified cathodes provide higher conductivity for improved electron transfer efficiency
and allow more nano-sized catalyst particles to be deposited in a 3D texture to facilitate
the tri-phase reaction kinetics. Therefore, in this study we used carbon nanotubes to
modify the air-cathode using several different methods and characterized the MFC
performance under each condition.
40


3.3 Materials and Methods
3.3.1 Cathode construction
Different coating methods were used to modify MFC air-cathodes using carbon
nanotubes. In addition, carbon nanotube mat and carbon cloth modified with Pt
nanoparticles were tested for comparison. Table 3.1 shows the list of 8 different
electrodes that were characterized and the specifications of their modification.
Table 3.1 List of Cathode Materials and Modifications Used in This Study and
Their Specifications.
Cathode Name Material Type Pt Coating Method
CC Carbon cloth 10%Pt/carbon black
CC-Pt Carbon cloth with Pt mixture (brush)
CNTM Carbon nanotube mat 10%Pt/carbon black
CNTM-Pt Carbon nanotube mat with Pt mixture (brush)
^WNTn ^WNTn-Pt SWNT coated electrode SWNT coated electrode with Pt
2SWNTC-Pt 2MWNTC-Pt SWNT coated electrode with Pt MWNT coated electrode with Pt H2PtCl6 (Microwave)
1 Nanotube synthesized in National Renewable Energy Laboratory
2 Nanotube purchased from Cheap Tubes Inc.
Carbon cloth cathodes (CC) were made by applying one layer of carbon black
nanoparticles and four PTFE diffusion layers on the air side of the carbon cloth according
to Cheng et al.159 In some cases, a catalyst layer containing 0.5mg/cm2 platinum
nanoparticles was applied on the water side of the carbon cloth to improve reaction
kinetics (CC-Pt). Carbon nanotube mat (CNTM) that contains more than 90% carbon
nanotubes was donated by Nanocomp Technologies Inc. (OH, USA) and was used
directly as the MFC cathode. The same amount of Pt catalyst was applied in some of the
CNTM cathodes for comparison (CNTM-Pt).
41


Single-walled carbon nanotubes (SWNTs) were synthesized by Laser
Vaporization methods174,175 at the National Renewable Energy Laboratory (NREL) (CO,
USA). The as-prepared nanotubes were purified using a method introduced by Dillon,
which consisted of treatment in 3 molar nitric acid flux for 16 hours, followed by an
acetone wash and air burn at 525C.174 Commercially available SWNTs and multi-walled
carbon nanotubes (MWNTs) were purchased from CheapTubes Inc., (NJ, USA) and
purified by the same procedure. The purified carbon nanotubes were suspended in 1%
sodium dodecyl sulfate (SDS) surfactant water solution to prevent agglomeration. The
suspended carbon nanotubes were then deposited onto a PTFE membrane (Advantec
MFS, Inc, CA, USA) by vacuum filtration to form a compact cathode. Sufficient DI
water washing was applied after to remove the extra surfactant. In some cases, Pt
nanoparticles (0.5mg/cm2) were uniformly coated on the surface of carbon nanotubes by
using a controlled temperature microwave method (250 W, 140 C) for 90 seconds.176
The cathodes containing nanotubes synthesized at NREL with and without Pt coatings
were labeled as SWNTn-Pt and SWNTn, respectively. Similarly, the cathodes using
commercial SWNTs and MWNTs from CheapTubes Inc. with Pt coating were labeled as
SWNTC-Pt MWNTC-Pt, respectively (Table 3.1).
3.3.2 MFC construction and operation
Single chamber cubic-shaped reactors were constructed as previously described.5,
152 The total volume of each reactor was 28 mL. Graphite fiber brushes were used as the
anodes in all of the experiments. Prior to the tests, the brashes were treated by soaking in
acetone overnight and then heated at 450C for 30 minutes.148 All reactors were initially
equipped with CC-Pt cathodes and inoculated using effluent from an air-cathode MFC
42


operated for nearly one year. When exoelectrogenic bacteria were acclimated and all
reactors showed repeatable and comparable voltages (510+10 mV), nanotube modified
air-cathodes were transferred to MFC reactors for performance characterization. Medium
solution was prepared containing 1.25 g/L CfFCOONa, 0.31g/L NH4CI, 0.13g/L KC1,
3.321g/L NaH2P04-2H20, 10.317g/L Na2HP04- 12H20, 12.5 mL/L mineral solution, and
5 mL/L vitamin.19 Reactors were operated in fed-batch mode at room temperature and
refilled with new medium solution when voltages reduced below 30 mV forming one
cycle of operation.
3.3.3 Electrochemical and Microscopy Analysis
Voltage across an external resistor (1000 Q) was recorded at 10-minute intervals
using a data acquisition system (Keithley Instrument, OH, USA) connected to a computer.
Polarization power density curves were obtained by altering external resistances from
50000 Q to 50 Q during the stable power production stage of each batch. The calculations
of power density was performed according to Ren et al.177
Linear Sweep Voltammetry (LSV) was applied using a potentiostat (PC4/300,
Gamry Instruments, NJ, USA) to characterize the abiotic electrochemical oxygen
reduction behavior on different cathodes. LSV tests were conducted in the same reactor
filled with same media solution but without carbon source and bacterial inoculums.178
The working electrode was the cathode (4.9 cm2 projection area), the counter electrode
was a titanium wire (diameter: 0.081 cm, length: 40 cm), and the reference electrode was
an Ag/AgCl electrode (RE-5B, BASi, IN, USA). The potential was scanned from open
circuit potential to -0.2 V at a rate of 1.0 mV/s. Reactor internal resistances were
measured by Electrochemical Impedance Spectroscopy (EIS) with the anode as the
43


working electrode, and the cathode as the counter electrode and reference electrode. The
scan range was from 105 Hz to 0.005 Hz with a small sinusoidal perturbation of 10
mV.158 All tests were repeated at least three times to verify consistency.
Selected cathode samples were examined using a dual beam focused ion beam
scanning electron microscope (FIB/SEM, NOVA 600i, FEI Company). Samples were
fixed overnight at 4C by Kamovskys Fixative (Electron Microscopy Sciences, CA,
USA), washed three times in phosphate buffer (0.2 M, pH 7.2), and then dehydrated
stepwise in a series of water/ethanol solutions with increasing ethanol concentration (50,
70, 80, 90, 100 %). Samples were then kept in a desiccator prior to Pd/Pt sputtering and
SEM observation.
3.4 Results and Discussion
3.4.1 Electrochemical performance
The electrochemical performance of the cathodes with respect to current density
was evaluated using LSV tests in the absence of bacteria (Figure 3.1). Among the
cathodes without Pt catalyst coating, those electrodes consisting of or modified with
carbon nanotubes (e.g. CNTM and SWNTn) showed much larger current responses than
the carbon cloth cathode (CC). SWNTn cathodes showed the most positive onset potential,
followed by the CNTM cathode, while the CC showed minimal current response across
the potential scan range. The larger current response from the nanotube cathodes
indicates a higher limiting current density and better electrochemical performance,
presumably due to the higher specific surface area generated by nanotube modification.179
In comparison, the flat current response of CC cathodes over the range of voltages
examined indicates that the carbon cloth itself did not catalyze oxygen reduction.
44


E
<
co
C
u
Q
4>
c
s
u
2
0
2
4
-6
-8
10
-12
-14
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1
Potential (V, vs Ag/AgCl)
0.2 0.3
0.4
Figure 3.1 LSV Results (Current Density vs Potential) of Newly Modified Cathodes
Before Installing in MFCs. Current Density Range Was Marked Based On the
Values Shown in Figure 3.3.
The addition of Pt catalyst on the cathodes significantly increased the current
densities of all cathodes (Figure 3.1). The current densities of CNTM-Pt, SWNTn-Pt, and
CC-Pt cathodes all increased substantially compared with their non-Pt counterparts, and
the SWNTn-Pt showed the best performance across the potential range. Considering the
same amount of catalyst was applied on the electrodes, the LSV results suggest that the 3-
D structure created by the carbon nanotubes on SWNTn-Pt allowed more Pt nanoparticles
to be deposited inside the electrode space rather than on the surface, leading to increased
reaction kinetics.180 The electrochemical performance of cathodes modified by
commercial single-walled nanotubes (SWNTC-Pt) and multi-walled nanotubes (MWNTC-
Pt) with Pt coatings was also evaluated using the same LSV tests. In general, both
cathodes showed lower current densities compared to the other three electrodes described
previously. The reason could be attributed to the different nanotube synthesis approaches.
The laser vaporization method used for SWNTn-Pt has been known to produce higher
45


quality nanotubes than the chemical vaporization method used for SWNTC-Pt. The higher
crystallization and electric conductivity of SWNTn resulted in higher electron transfer
efficiency thus higher performance. The SWNTC-Pt showed larger current response than
the MWNTc-Pt across the scanned potentials. This is presumably due to the high
resistance thus more electrochemical loss of MWNTs compared with SWNTs. The
resistance of the reactor with the MWNTC-Pt cathode was 94 Q, which is nearly 3 times
higher than with the SWNTC-Pt cathode (35 Q).
3.4.2 Performance of MFCs with nano-modified air-cathodes
100
200
Time (h)
300
400
100
200
Time (h)
300
50 100 150
Time (h)
200
50 100 150 200 250 300
Time,(h)
Figure 3.2 Voltage Generation as a Function of Time for the Different Cathodes.
The nano-modified air-cathodes were transferred to pre-acclimated single
chamber MFCs after LSV tests. Rapid voltage generation was observed in all reactors.
46


Figure 3.3 shows the power density and polarization curves obtained from each MFC
reactor during steady power production stages. Figure 3.2 shows the voltage profiles as a
function of time. For MFC cathodes without Pt catalyst coating, CNTM reactor showed
the maximum power density of 329 mW/m2, more than twice that of the peak power
obtained from CC reactor (151 mW/m2). The SWNTn reactor showed a lower power
density (117 mW/m2) than the other two reactors without Pt. Similar to the results
observed during LSV tests, the reactor performance was greatly improved by the
application of Pt catalyst in the cathodes. During the first several batches, the power
densities of SWNTn-Pt and CNTM-Pt reactors, calculated from voltages at 1000 Q, were
735 mW/m2 and 723 mW/m2, respectively, which were higher than that of CC-Pt reactor
(672 mW/m2). The voltages of CNTM-Pt and CC-Pt MFCs kept stable during multiple
batches, but the voltage of the SWNTn-Pt reactor decreased gradually from the first batch
and stabilized after the fourth batch, resulting in a voltage drop from more than 600 mV
to around 300 mV. As shown in Figure 3.3, the power density curves obtained at the
steady state operation of each reactor show the maximum power densities from CNTM-Pt
and CC-Pt MFCs of 1118 mW/m2 and 1071 mW/m2, respectively, while the maximum
power density of SWNTn-Pt after stabilizing was 302 mW/m2. A similar voltage decline
was observed in SWNTC-Pt and MWNTC-Pt MFCs in the first batches. The maximum
power density produced by SWNTC-Pt was 522 mW/m2, which was about two times
higher than that from MWNTC-Pt (174 mW/m2), confirming that SWNT material
performs better as MFC cathodes due to its lower ohmic resistance compared with
MWNT material.
47


cc
CC-Pt
CNTM
CNTM-Pt
SWNTn
SWNTn-Pt
MWNTc-Pt
SWNTc-Pt
CC
CC-Pt
CNTM
CNTM-Pt
SWNTn
SWNTn-Pt
MWNTc-Pt
SWNTc-Pt
Current Density (A/m )
Figure 3.3 Power Density as a Function of Current Density (A) and Polarization
Curves (B) for MFCs Operated Using Different Air-cathodes.
The electrochemical performance of CC-Pt and SWNTn-Pt cathodes were
analyzed again by LSV tests after MFC operation in order to understand the chemical and
microbial effects on the cathode performance. Figure 3.4 shows the electrochemical
performance of the used CC-Pt cathode decreased slightly compared with the new
cathode, while the current response of the used SWNTn-Pt cathode declined significantly
compared with the new SWNTn-Pt. This drop may explain why the steady state power
density of the SWNTn-Pt reactor was lower than the CNTM-Pt and CC-Pt reactors
despite a better LSV performance with new SWNTn-Pt cathodes. Such finding could also
48


be confirmed by the changes of open circuit potential (OCP). The OCP of the SWNTn-Pt
cathode dropped by 35%, from 357 mV in batch 1 to 233 mV in batch 4, while the OCP
of CC-Pt only dropped by 9%, from 345 mV to 310 mV during the same period.
Potential (V, vs Ag/AgCl)
Figure 3.4 Comparison of LSV Electrochemical Test Results between New and
Used Cathodes of CC-Pt and SWNTn-Pt.
Similar findings in air-cathode performance decline were reported by several
other studies, and the reasons were believed to be due to the adherence of biofilm and
chemical deposits on the cathode surfaces that reduced charge transfer and catalyst
activity.181,182 For the SWNTn-Pt cathode, the large amount of pores, cavities, and
curving paths among carbon nanotubes that increase the specific surface area also enables
adsorption of impurities that may cover the surfaces of the cathode and reduce charge
transfer. Adding a separator on the nanotube modified electrode to block adsorption or
switching the nanotube layer to the air-face side could be possible solutions.183,184 It may
also be possible that the SWNTn-Pt cathode performance decreased due to the loss of the
Pt catalyst overtime since the microwave deposition method has not been optimized for
49


MFC operation. However, this was not confirmed using energy dispersive spectroscopy
element percentage test during FIB/SEM characterization.
3.4.3 FIB / SEM analysis
Scanning electron micrographs were taken for the CC-Pt, CNTM-Pt, and SWNTn-
Pt cathode materials before and after MFC operation. Figure 3.5 shows a distinct
difference in morphology between the new CC-Pt cathode and new CNTM-Pt and
SWNTn-Pt cathodes. The CC-Pt cathode surface was covered by aggregated particles or
short fibers (Figure 3.5A), while the new CNTM- Pt and SWNTn-Pt electrode surfaces
clearly showed the carbon nanotube network (Figure 3.5C, 5E). The deposited Pt catalyst
shows up in the images as bright nanoparticles deposited across the nanotube network.
Denser microbial biofilms were formed on the surface of the CC-Pt cathode
(Figure 3.5B) compared with the CNTM- Pt and SWNTn-Pt cathodes (Figure 3.5D, 5F),
but more chemical deposit covered a large portion of the surface of the two CNT-
modified electrodes. Studies showed that the excess accumulation of biofilm and
chemical scales could adversely affect the system performance due to the decrease in
active cathode specific surface area and increase in diffusion resistance in oxygen.181,182
Such observations may also explain the reduced electrochemical activities in certain
MFCs after a period of operation, where the presence of cavities and the adsorption of
impurities played a major role as compared to cathode biofilms.
50


Figure 3.5 SEM/FIB Images of New and Used Cathodes: (A) New CC-Pt, (B) Used
CC-Pt After MFC Operation, (C) New CNTM, (D) Used CNTM-Pt After MFC
Operation, (E) New SWNTn-Pt, and (F) Used SWNTn-Pt After MFC Operation.
3.5 Conclusions
Microbial fuel cells provide direct and efficient electricity generation from
renewable sources, but the current 2-D air-cathode configuration limits system
51


performance due to the low kinetics of oxygen reduction. Carbon nanutubes were used in
this study to modify air-cathodes in single chamber MFCs to create a 3-D structure for
improved surface area and reaction kinetics. Compared with traditional carbon cloth
cathodes, all nanotube modified cathodes showed greater electrochemical performance as
well as higher power density in MFCs. The maximum power density obtained from
carbon nanotube mat cathodes (329 mW/m2) was more than double the power output
from traditional carbon cloth cathodes (151 mW/m2). The addition of Pt catalyst on the
cathodes increased the current densities of all cathodes, with the maximum power density
achieved by a CNTM-Pt of 1118 mW/m2. Electrodes made from commercial single wall
carbon nanotubes (SWNTC-Pt) have much lower ohmic resistance than those made from
multiwall nanotubes (MWNTC-Pt) and showed larger current response and higher power.
The customized SWNTn electrode showed great electrochemical responses, but its
performance declined gradually due to the deposition of chemical and microbial
impurities which blocked reaction surfaces. Scanning electron micrographs demonstrated
different electrode surface morphologies, with the CNT modified cathodes showing
carbon nanotube networks and carbon cloth cathodes showing aggregated particles on the
electrode surface.
3.6 Acknowledgement
This research was supported by the Office of Naval Research (ONR) Grant
N000140910944. The authors thank Nanocomp Technologies Inc. for donating carbon
nanotube mat, and Dr. Paul Rice at University of Colorado Nanomaterials
Characterization Facility (NCF) for helping with FIB/SEM operation.
52


4. Active Energy Harvesting from Microbial Fuel Cells at the Maximum Power
4
Point without Using External Resistors
4.1 Abstract
Microbial fuel cell (MFC) technology offers a sustainable approach to harvest
electricity from biodegradable materials. Energy production from MFCs have been
demonstrated using external resistors or charge pumps, but such methods can only
dissipate energy through heat or receive electrons passively from the MFC without any
controllability. This study developed a new approach and system that can actively extract
energy from MFC reactors at any operating point without using any resistors, especially
at the peak power point to maximize energy production. Results show that power
harvesting from a recirculating-flow MFC can be well maintained by the maximum
power point circuit (MPPC) at its peak power point, while a charge pump was not able to
change operating point due to current limitation. Within 18-hour test, the energy gained
from the MPPC was 76.8 J, 76 times higher than the charge pump (1.0 J) that was
commonly used in MFC studies. Both conditions resulted in similar organic removal, but
the Coulombic efficiency obtained from the MPPC was 21 times higher than that of the
charge pump. Different numbers of capacitors could be used in the MPPC for various
energy storage requirements and power supply, and the energy conversion efficiency of
the MPPC was further characterized to identify key factors for system improvement. This
active energy harvesting approach provides a new perspective for energy harvesting that
can maximize MFC energy generation and system controllability.
4 The work presented in this chapter has been published by Heming Wang, Jae-Do Park,
and Zhiyong Ren in Environ. Sci. Technol., 2012, 46, 5247-5252.
53


4.2 Introduction
A microbial fuel cell (MFC) is a bioelectrochemical system (BES) that employs
exoelectrogenic bacteria to oxidize organic matter and produce direct electrical current.
Because MFC offers a sustainable solution for remote sensing and simultaneous pollution
control and energy production, it has been intensively researched in recent years, and the
improvements in reactor configurations, materials, and operations have led to orders of
magnitude increase in power density, from less than 1 mW/m2 to the level of 6.9
0 1 A"3 1ZC
W/m . However, most studies operate the MFC with a static external resistance or
applied potential and report the power density using a polarization curve, which assumed
that the maximum power density is achieved when the applied external resistance is equal
to the MFC internal resistance.177,185,186 Such characterizations do represent the
theoretical potential of MFC power output, but no usable energy could be captured,
because the electricity generated in such systems is actually dissipated into heat instead
of being utilized by electronics. Moreover, the fixed external resistance cannot always
match the system internal resistance and recover the maximum power output during MFC
operation, because the internal resistance of an MFC varies constantly with changes in
microbial activities and operational parameters, such as substrate concentration, pH, and
temperature.112,187489 Studies showed that MFCs may lose more than 50% of produced
power across the internal resistance if the operating voltage is not at the maximum power
point voltage.190
To effectively and efficiently harvest MFC energy, unnecessary resistors need to
be eliminated, and technologies need to be developed to track and harvest energy at the
peak level with sufficient controllability. Progresses have been made in maximum power
54


point tracking (MPPT) and harvesting systems, such as using perturbation and
observation or gradient method to track and optimize external resistance.186,190 For
example, Pinto et al. find that MFC power output can be significantly improved when
real-time resistance optimization was implemented during long term operation.186
However, traditional MPPT techniques still use external resistances and cannot capture
and utilize the energy directly. Another harvesting approach is using capacitor-based
circuits such as super capacitors and charge pumps, which capture MFC energy passively
and transfer it to a boost converter.191,192 For example, a recently study by Liang et al.,
showed that current production from a BES reactor can be increased by 22-32% if an
alternative charging and discharging method is used. In such operation, a capacitor is
firstly charged by the reactor but then discharges the electrons back to the reactor. This is
different from traditional intermittent charging, where a capacitor discharged the
collected electrons to a resistor.191 Another study by Kim et al., showed that parallel
charging of multiple capacitors can avoid potential voltage reversal while series
discharging could increase MFC output voltage.193 The problem of directly using
capacitors or charge pumps is that such devices can only passively receive MFC energy
at a fixed operating point without any control on the MFC reactor, and the operating
points cannot be adjusted to capture energy at the maximum power density point.
In this study, we developed a new energy harvesting approach and system that not
only can capture the maximum power from the MFC, but also harvests energy actively
without using any resistance. Instead of passively receiving electrons from the MFC
reactor, this controller can actively extract energy from the MFC at any operating point,
especially at the peak power point to maximize energy production. The energy harvesting
55


efficiency, organic removal, and Coulombic efficiency of the MFC operated by this
maximum power point circuit (MPPC) was characterized and compared with a common
charge pump operation. The energy storage capacity using different numbers of
capacitors and system energy conversion efficiency was also investigated for system
optimization.
4.3 Materials and Methods
4.3.1 MFC construction and operation
Each MFC reactor consisted of two polycarbonate cube-shaped chambers that
were separated by a cation exchange membrane (38 cm2, CMI-7000, Membranes
International, NJ).20 The empty volume of either anode or cathode chamber was 150 mL.
Heat treated graphite brashes were used as the anodes, and carbon cloth (projected
surface area 38 cm2) was selected as the cathode material.148,194 MFCs were inoculated
with anaerobic sludge obtained from Fongmont Wastewater Treatment Plant (Fongmont,
CO). The anode chamber was fed with growth medium containing (per liter) 1.25g
CH3COONa, 0.3lg NH4C1, 0.13g KC1, 3.32g NaH2P04-2H20, 10.32g Na2HP0412H20,
12.5mF mineral solution, and 5mF vitamin solution.177 Phosphate buffered potassium
ferricyanide solution (50 mM) was used as the catholyte to minimize the cathode effects
on system performance.20 MFCs were operated in fed-batch mode at the acclimation
stage until repeatable voltage profiles were obtained. Reactors were then operated by
recirculating anolyte with a 1000 mF reservoir at a flow rate of 45 mF/min and
recirculating catholyte with another reservoir at a flow rate of 114 mF/min, respectively.
Such operation was aimed to maintain stable substrate and pH conditions so energy
harvesting characterization could be focused.19,195
56


4.3.2 Maximum power point circuit (MPPC) design and operation
NO. Component Function
(D Capacitor Inductor Stores energy extracted from MFC Intermediate energy storage before it is transferred to capacitor
Diode Transfers energy from inductor to capacitor and blocks reverse power flow
MOSFET Main switch of energy harvesting converter
Comparator Generates hysteresis voltage band
Potentiometer Adjusts MFCs working voltage
Potentiometer Adjusts width of hysteresis voltage band
Connector Monitors MFC voltage
Connector Reference voltage for comparator
Manufacture
Taiyo Yuden
Triad Magnetics
Micro
Commercial
Components
Vishay
National
Semiconductor
General
General
TE Connectivity
TE Connectivity
Figure 4.1 Components in MPPC.
The MPPC consisted of a metal-oxide-semiconductor field-effect transistor
(MOSFET), a comparator, an inductor, a diode, capacitors, potentiometers, and
connectors. The detailed information of each MPPC component is listed in Figure 4.1,
and the circuit design details is described by Park and Ren.196 Figure 4.2 shows the
principles of the energy harvesting MPPC. The MPPC is able to operate the MFC reactor
in the vicinity of the maximum power operating point, which is regulated by a hysteresis
57


controller (Figure 4.2A). The hysteresis controller confines the MFC voltage in a pre-
defined range to avoid voltage collapse and ensure enough recovery time of the MFC
reactor, and the upper (VthH) and lower (VthO voltage thresholds can be defined by
equation (l).196
VthH =Va
R,
r2+(rj/r3)
R, HR,
vthL=v 2 3
r2 + (r2//r3)
(1)
Where Vcc is the external voltage for MPPC circuit, Rl, R2, and R3 are internal
resistors to set the harvesting hysteresis voltage band, and the double slash means parallel
connections of resistors (Figure 4.2). For comparison with traditional passive energy
harvesting approaches, a charge pump (S-882Z24, Seiko Instruments) was used in a
control experiment with the same reactor configuration and operation.
The operation of the MPPC consists of two modes, CHARGE and DISCHARGE,
according to the energy flow on the inductor connected with the MFC (Figure 4.2B).
During CHARGE mode, the MOSFET switch is on and diode is off, and the energy is
extracted from the MFC and charged to the inductor (Figure 4.2C). Due to energy
extraction, the voltage of the MFC decreases in this mode. During DISCHARGE mode,
the MOSFET switch is off and diode is on, and the energy stored in the inductor is
discharged to the capacitor (Figure 4.2D). MFC voltage increases in this mode as it
recovers from energy extraction. The controller turns off the MOSFET automatically
when the MFC voltage reaches lower threshold in CHARGE mode and turns it back on
when the MFC voltage gets to the upper threshold in DISCHARGE mode. The duty ratio
and switching frequency can vary depending on the generating capacity and recovery
time of the operating MFC. The comparator generated hysteresis voltage band according
58


to the MFC voltage, and the voltage band can be easily tracked and adjusted by
. 14*
potentiometers.
vcc vcc
B
Capacitor
D
+
MFC

Inductor
Capacitor
Figure 4.2 Block Diagram of the Maximum Power Point Circuit (MPPC): (A)
Harvesting Converter Controller. (B) Whole electric Circuit Diagram; (C)
CHARGE Phase, MOSFET is On While Diode is Off, Extracted Energy is Stored in
the Inductor; (D) DISCHARGE Phase, MOSFET is Off While Diode is On,
Extracted Energy is Stored in the Capacitors.
59


4.3.3 Analyses
The MFC voltage, capacitor voltage, and the output voltage across a current probe
(K110, AEMC Instruments) were recorded at 66-sec intervals using a data acquisition
system (Keithley Instrument, OH). The anode potential and cathode potential were
measured against a Ag/AgCl reference electrode (RE-5B, Bioanalysis) inserted in the
anode chamber. An oscilloscope (TPS2014B, Tektronics) was used to continuously
monitor MFC voltage, output current and the main switch on/off signal. Chemical oxygen
demand (COD) was measured using a standard colorimetric method (Hach Company,
CO). Polarization curves were obtained by linear sweep voltammetry (LSV) using a
potentiostat (PC4/300, Gamry Instruments, NJ). The scan rate of LSV was 0.1 mV/s with
the anode as working electrode and the cathode as counter and reference electrode.178,194
The output power (P) of MFC was calculated by P = UI, where U is the voltage
across the MFC anode and cathode, and I is the MFC output current monitored by the
current meter. Power density and current density were normalized by the projected area
of the cathode (38 cm2). Energy (Wc) consumed by an external resistor (R) was calculated
by Wr = j U2 / Rdi, and the energy (Wp) supplied by the MFC during harvesting by the
MPPC or charge pump was expressed as Wp=J Pdt, where dt is 66 sec. The Energy (E)
stored in the capacitors was calculated by E = 0.5CV2, where C is the capacitance, and V
is the capacitor voltage. Energy conversion efficiency (ECE) was calculated by
ECE = EjWp xl00% .The energy (J) consumption in each MPPC component was
calculated based on E = Vlt during the harvesting period. Coulombic efficiency (CE) was
presented as CE = 8000j Idt jFVACOD, where F is Faraday constant, V is total volume,
60


and A COD is COD concentration change. Duty ratio (D) was defined as the ratio of
turn-on time to the total switching time, D = tonl(ton +toff), where ton and toff is the on
and off time of the MOSFET, respectively.
4.4 Results and Discussion
4.4.1 MPPC can operate the MFC at the maximum power harvesting range
800
700
600
g 500
m 400
£ 300
200
100
0
Current Density (A/m2)
Figure 4.3 MFC Polarization Curve and Power Density Curve Obtained by Linear
Sweep Voltammetry (LSV). The Scan Rate of the Polarization was 0.1 mV/s.#:
Operating Point of the Charge Pump. A: Operating Range of the MPPC.
Recirculating-flow MFC Open Circuit Potential was 688 mV.
Figure 4.3 shows the polarization and power density curves obtained in the
steady-state recirculating flow MFC reactor. The maximum power density produced by
the MFC was around 1370 mW/m2 when the reactor voltage was between 372 mV and
316 mV, with an average of 344 mV. The corresponding external resistor at the peak
power density was 23 Q. In order to harvest the maximum power identified by the MFC
c
qj
Q
u
qj
*
o
61


power density curve, the upper voltage threshold at 372 mV and lower voltage threshold
at 316 mV were determined to form an energy extraction band for the hysteresis
controller in MPPC. In contrast, the charge pump was only able to harvest the MFC
energy at the 633 mV (317 mW/m2) due to the current limitation of the charge pump
(Figure 4.3). The difference in the operating points on the power density curve
demonstrates that the MPPC could be modulated to harvest energy at the range of the
peak point while the power harvesting by the charge pump was limited at lower points
due to the lack of controllability.
(f-harge j j Discharge
1 \ i y
Voltage
On/off time
11
11
11
CURSOR
Type
|Amplitude|
Source
SG.OnnV
Cursor 1
372mV
Sir
. u
100m'v
CH2
CM
1.00 V
SOOrmV
M 100 jus
19Dec1117:06
Cursor 2
316mV
CH2 I 236mV
5.45763kHz
Figure 4.4 Snapshot of On/Off Cycle of the MPPC During Active Energy
Harvesting from MFCs and the Voltage and Current Profiles. One division of X-
Axis Represents 100 psec. The Figure Shows the Waveforms of 1 msec Duration in
Terms of Current, Voltage, and On/Off Switch Changes.
Typical operation cycles of the MPPC harvesting are shown in Figure 4.4. When
the MFC voltage reaches to 372 mV, the MPPC actively extracts energy from the MFC
and charge the inductor (CHARGE mode). The extraction stops when the voltage drops
to 316 mV. While waiting for the MFC voltage to recover, the controller discharges the
energy from the inductor to the capacitor (DISCHARGE mode). Once the MFC voltage
62


recovers back to 372 mV, the controller charges the inductor again. The switching
frequency between CHARGE and DISCHARGE phase was very fast and in the order of
kHz. The duration of CHARGE and DISCHARGE phases depend on the MFC condition,
and the DISCHARGE phase was also affected by the target of the capacitor voltage. The
range of operation can be tracked and controlled by the Hysteresis controller.145 Anolyte
and catholyte recirculation operation was used in this study because such system could
maintain a relatively stable substrate concentration, pH, and other operating conditions as
compared to fed-batch operation and thus reduces the effects of environmental factors.
4.4.2 MPPC harvests energy more actively and efficiently
MFC power density curves demonstrate that when the applied external resistance
is equal to the MFC internal resistance, the maximum power can be achieved. Figure 4.3
shows that the peak power of the MFC used in this study could be obtained at 23 Q, and
Figure 4.5 shows the MPPC-controlled MFC was operated nearly as the same condition
as the reactor operated under a 23 Q resistor. The operating curves of the anode potential,
cathode potential, and reactor voltage in both conditions basically overlapped each other,
indicating very similar operating conditions, where the maximum power could be
generated. However, instead of dissipating the energy into heat as resistors do, the MPPC
captured the energy and stored energy into capacitors.
Energy harvesting results show that the MPPC-controlled MFC was able to
charge multiple capacitors (Taiyo Yuden, PAS1016FR2R3205). After 18 hours of
operation, the voltage of the 12 capacitors connected to the MPPC controller increased
from 0 V to 2.5 V, and the MPPC extracted 214.1 J of energy from the MFC, in which
76.8 J were stored in the capacitors (Figure 4.6A). In comparison, the charge pump was
63


able to charge 1 capacitor to 1.0V during the same period, and the total extracted and
stored energy was 23.2 J and 1.0 J, respectively (Figure 4.6A). The results show that by
actively extracting energy at the maximum power point, the MPPC harvested 76 times
more energy than the charge pump.
700
500
>
B
~ 300
is
'S
1 100
e-
o
6Jj
2 -100
"o
E*
-300
-500
0 4 8 12 16 20
Time (h)
Figure 4.5 Comparison of MFC Voltage, Cathode Potential, and Anode Potential
between the MPPC Active Energy Harvesting Condition and 23 Ohm External
Resistor Condition. The Optimum External Resistance was Calculated to be 23
Ohm Based on Polarization Curve that Could Yield the Maximum Power Density.
--MFC_MPPC -*-Cathode_MPPC Anode_MPPC
MFC_23ohm * Cathode_23ohm Anode_23ohm
Comparable substrate degradation was observed in both energy harvesting
operations, as the COD removals were 49.8% and 47.1% for the MPPC and charge pump,
respectively (Figure 4.6B). However, the Columbic efficiency of the MPPC operation
was 15.6%, 21 times higher than that of the charge pump (0.7%). This finding is
consistent with previous studies that higher Columbic efficiency can be achieved by
operating MFCs at an optimal external resistance.186 Moreover, compared to other studies
that selected the optimal resistance to demonstrate the power generation potential, the
MPPC actually captured the energy at the maximum power point that is available for
64


electronic utilization. Ferricyanide solution was used as the catholyte in this 2-chamber
MFC study, so no cathode biofilm was observed to consume substrate and affect
Columbic efficiency.
Energy extracted Energy stored in Start_COD End_COD
from MFC capacitors
Figure 4.6 (A) Comparison of Energy Harvesting by the MPPC and the Charge
Pump and Energy Stored in Capacitors. (B) Comparison of COD Removals in the
MPPC and Charge Pump Conditions. In the MPPC Test, 12 Capacitors were
Connected in Parallel for Energy Storage. In the Charge Pump Test, one Capacitor
was Enough to Store All the Harvested Energy from MFC.
4.4.3 The numbers of capacitors for energy storage
Different numbers of capacitors were tested in the study for energy storage and to
provide stable electricity for electronic devices. While resistors dont capture any energy,
and the charge pump was only able to charge 1 capacitor, the MPPC harvested so much
energy that multiple capacitors had to be used for energy storage. The charging behavior
of using 3, 6, 9, and 12 capacitors during 18-hour harvesting was characterized. The
capacitors were connected in parallel in order to maintain charging efficiency.
65


n 4 8 12 16 20
Figure 4.7 (A) Voltage Profile and (B) Energy Storage Differences by Using 3, 6, 9,
and 12 Capacitors in Parallel During MPPC Active Energy Harvesting.
Figure 4.7 shows that with the increase of charging time, the voltage of the 3-
capactior condition increased faster than other conditions and reached the saturated level
of 2.9 V in 4 hours. The stored energy in the 3 capacitors was 25.2 J within this period.
After 4 hours, the 3-capacitors could not store more energy due to saturation, but the
MPPC kept harvesting energy from the MFC, as evidenced by the 6, 9, and 12-capacitor
conditions. With increasing numbers of capacitors, the voltage increase rate declined, but
66


the total amount of energy stored in capacitors increased. The energy storage in the 6, 9,
and 12-capacitor condition was 47.0 J, 65.6 J, and 76.8 J, respectively, and the
corresponded voltage after 18 hours was 2.8 V, 2.7 V, and 2.5 V, respectively. Because
the larger the capacitance, the longer the charging time is required, which resulted in
differences in capacitor voltages and energy storage. More energy can be stored in higher
capacitance conditions if the harvesting continues. These results indicate that different
numbers of capacitors or capacitors with different capacitance can be used as energy
storage for operating electronic devices. The required number of capacitors, charging
time and energy storage capacity are determined by the characteristics of end users, and
the MPPC-controlled MFC should not be a limiting factor under stable operating
condition.
4.4.4 Conversion efficiency of the MPPC
The MPPC can actively harvest energy at the maximum power point thus
significantly increased energy generation from MFCs. However, like any electronic
device, the MPPC consists multiple electronic components that consume energy, which
reduce MPPCs energy conversion efficiency. When the MFC was operated under 23 Q,
the optimal external resistance that lead to the maximum power density, the MFC could
provide 215.7 J (prorated by duty ratio) of energy in 18-hour, even though all such
energy will be dissipated through heat by the resistor. In comparison, the MPPC
harvested 214.1 J of energy from the MFC without external resistors and transferred 76.8
J to the capacitors. This further confirmed that the MPPC was able to harvest 99.2% of
the energy from the MFC, but it also shows that only 35.9% of the harvested energy was
transferred to the capacitors. Even though the MPPC-transferred energy is still 76 times
67


higher than that from the charge pump (1.0 J), it is important to identify the limiting
factors within the MPPC circuit and improve the conversion efficiency. Figure 4.8 shows
the MPPC efficiency through an 18-hour test, and it can be seen that the efficiency
increased sharply at the beginning and then stabilized till a decline was observed due to
the saturation of capacitance. The highest energy extraction efficiencies occured between
8.3 to 11.5 hour, with an efficiency of 42.1%.
a
(M3
S
"3
Figure 4.8 Efficiencies through 18-hour Test.
The power consumption of each component in the MPPC circuit was calculated in
Table 4.1. Figure 4.9 illustrates the percentage of energy loss in each component during
MFC energy extraction by the MPPC during an 18-hour period. While most MPPC
components consumed minimum amount of energy, the diode contributed to 58.8% of the
energy loss within the MPPC, indicating that it is the single element that needs to be
replaced or improved. The diode is used in the MPPC to transfer the energy from the
68


inductor to capacitors and blocks reverse flow. Advanced converters are currently being
developed to replace the diode and improve MPPC conversion efficiency.
Table 4.1 Analysis of Energy Extraction Efficiency by MPPC.
Components Information from datasheet Calculation
Energy stored in 12 capacitors after 18 h Capacitance: 2F (1/2) (2F 12) (2.53E)2 = 76.8/
Inductor DC Resistance Max: 0.58ohm (0.58 ohm) (12 mA/1000)2 (3600 sec/h) (18 h) = 5.4 J
Capacitor ESR (Equivalent Series Resistance): 50.0 mohm for each 12 (1/50 mohm) = 1/R R = 4.17 mohm (4.17 mohm/1000) (12 mA/1000)2 (3600 sec/h) (18 h)
* (40/178) = 0.009 J
Diode Forward voltage drop: 0.72V at 12mA (12 mA/1000) (0.72 V) (3600 sec/h) (18 h) (40/178) = 125.8 J
MOSFET Drain-Source On-State Resistance: 0.033 ohm at Vgs=1.8V (0.033 ohm) (12 mA/1000)2 (3600 sec/h) (18 h) (138/178) = 0.24 J
Total extracted energy (76.8 J) + (5.4 J) + (0.009 J) + (125.8 J) + (0.24 J) = 208.2 J
Total energy provided by MFC (276.1 J) (138/178) = 214.1 J
Error (214.1 J) (208.2 J) = 5.9 J
Energy extraction efficiency (76.8 J)/(214.1 J) = 36%
Average current in the circuit is 12 mA;
Duty ratio is 138ps:40ps.
Others
2.6%
Figure 4.9 Energy Conversion Efficiency and Distribution of Internal Energy Loss
in the MPPC. The Distribution was Quantified Based on an 18-hour, 12-capacitor
Operation.
69


4.5 Outlook
MFC technology has been considered as a sustainable method to directly produce
energy from biodegradable substrates, but the improvement of power density has been
stagnant for several years after significant advancements in reactor configuration and
material development. Compared with traditional approaches that use external resistors
and charge pumps, this study demonstrates a new active approach to harvest energy from
MFCs. Instead of passively receiving electrons from the MFC, the MPPC actively
extracts energy from the MFC at the peak power point. The remarkable increase in
energy generation by the MPPC compared to the common charge pump shows this
approach is much more efficient and effective to capture MFC energy. There are very
few charge pumps available for MFC systems, and the charge pump used in this study is
representative, because it has been used by many other studies in different conditions.119,
197,198
The active energy harvesting approach is new to MFC operation, and there are
many questions remain to be answered. For example, one unique feature of MFCs is the
variable biocatalyst density on the electrodes. Exoelectrogenic bacteria transfer electrons
to the anode electrode and gain energy during anaerobic respiration. Within the capability
of bacterial extracellular electron transfer, the more electrons get extracted from the
external circuit, the less electrons and energy become available for microbial growth.
Therefore, it is important to understand how the active harvesting affects microbial
activity, community, and metabolisms, so a balanced and sustainable reactor performance
can be maintained. We did not find active harvesting negatively affect MFC performance
in terms of power density and Coulombic efficiency in recirculation operation. In
70


addition, further optimization of the MPPC circuit needs to be conducted to reduce
internal energy loss. Systems with precise tracking capability will allow the circuit to
adjust and maintain the maximum energy extraction based on real-time changes of MFC
condition due to the variations of environmental conditions such as pH, temperature, and
substrate concentration.
4.6 Acknowledgement
This work was supported by the Office of Naval Research (ONR) under Award
N000140910944. We thank Drs. Bmce Logan and Peter Jenkins for constructive
discussions.
71


5. Power Electronic Converters for Microbial Fuel Cell Energy Extraction: Effects
of Inductance, Duty Ratio, and Switching Frequency5
5.1 Abstract
Power converter based microbial fuel cell (MFC) energy harvesting has been
recently researched to replace the external resistors that have been utilized to show MFC
output in many studies. The electronic circuit can operate as an equivalent external
resistor, but the energy generated from MFC can be harvested in storage instead of being
dissipated. However, there is limited information in the literature about the effects of
operating configuration of power electronic circuits on MFC energy harvesting. In this
study, a boost-converter based energy harvester circuit was examined in terms of
inductance, duty ratio, and switching frequency. The results showed that all of these
factors play important roles for the performance of MFC and energy harvesting, and their
effects can be cross linked. Current and voltage is generally proportional and inversely
proportional to the inductance, respectively. The total harvested energy and efficiency
vary significantly by combinations of duty ratio and switching frequency. For the MFC
reactor tested in the study, the highest energy harvested was 3.48 J which was under the
combination of 14 mH inductance, 75% duty ratio and 5000 Hz frequency, comparing to
the highest efficiency of 67.7% happened at 130 mH inductance, 25% duty ratio and
4000 Hz frequency. When using the smallest inductance of 0.45 mH the highest energy
and efficiency were only 1.38 J (50% duty ratio and 5000Hz frequency) and 19.9% (25%
duty ratio and 5000Hz frequency), respectively. Regardless of the voltages and currents
5 The work presented in this chapter has been published by Heming Wang, Zhiyong Ren,
and Jae-Do Park in J. Power Sources 2012, 220, 89-94
72


produced in various operating configurations, anode potentials were stable, suggesting
that there were enough electrons available to be utilized for current generation. An
optimal operating configuration that provides ideal system performance can be found for
different reactors and applications.
Keywords: microbial Fuel Cell, energy extraction, DC/DC converter, inductance, duty
ratio, switching frequency
73


5.2 Introduction
Microbial fuel cells (MFCs) are considered new green energy sources due to their
wastewater treatment and simultaneous energy producing capability. In most studies so
far, an external resistor was inserted between the anode and cathode of an MFC, so the
voltage across the resistor can be monitored to demonstrate the power production.
However, this method does not harvest any energy because the energy is dissipated on
the resistor as heat. Furthermore, the relatively low voltage (<1.0 V) and low power
(~W/m3) output from an MFC cannot directly support majority of commercial electrical
devices, which is one of the biggest obstacles for practical application of MFC. Recently
power electronics based harvester circuits for MFCs has been researched,145193,199-201
aiming significantly improved MFC energy harvest and output voltage boost, which can
be a crucial step to make MFC technology commercially viable. Different from
conventional operations using external resistors, energy generated from MFCs will be
collected and stored, which in turn will be utilized to power electrical devices, for
example, wireless sensors to monitor environment.119,198
To design boost-converter based circuits for more versatile and efficient MFC
energy harvesting, there are three fundamental factors to consider: the inductance of an
inductor, extraction duty ratio (also known as duty cycle), and extraction frequency (i.e.
converter switching frequency). The inductor is the intermediate energy storage for an
MFC and determines the rate of current change and level of energy extraction. Large
inductance makes the current changing slowly while small inductance makes it faster,
and this contributes to determine the MFC terminal voltage. The duty ratio governs the
relative duration of energy extration in a certain switching period. In other words, it
74


determines the time period of energy extraction and MFC recovery. In MFC recovery
period, the energy harnessed in inductor is transfered to the capacitor storage. Higher
duty ratios lead to longer energy extraction and shorter time for MFC recovery, while low
duty ratios allow more time for MFC to recover and energy transfer. The switching
frequency determines the number of energy extraction and transfer within a given time
period. Higher switching frequency means the energy extraction happens more frequently,
but each extraction is shorter for a given duty ratio. The total energy extraction time is
determined by duty ratio regardless of switching frequency, which can be given as
Text = DxTprd. For example, the energy extraction with 50% duty ratio at 1000Hz
switching frequency has the same energy extraction time of 500 msec per 1 second period
as the 50% duty ratio at 2000Hz frequency, but the number of energy extraction cycle for
the first case is 500 times, only half of the latter case. However, it has been revealed that
different combinations of duty ratio and switching frequency affect MFC energy
harvesting results, even if the energy is harnessed for the same amount of time.
There are very few studies that investigated how to capture MFC energy more
effectively through the design and optimization of electronic harvesting circuits,
especially by using high-speed switching converters. Dewan et al.202 concluded that
intermittent energy harvesting (IEH) by alternatively collecting energy in the capacitor
and dispensing it through a resistor was more effective than continuous energy harvesting
(CEH) with constant energy extraction. The capacitor was charged for hours but
discharged for only less than a minute, which indicated that electroactive species around
the electrode was replenished while the capacitor was being discharged. Gardel et al.203
obtained similar results with duty cycling based energy harvesting from a multi-anode
75


MFC, which suggests that it was necessary to replenish depleted electron donors within
the biofilm and surrounding diffusion layer to maximum charge transfer. Grondin et al.204
also investigated the power output as a function of duty cycle, but the effect of extraction
frequency was not studied.
We recently developed a boost converter based energy harvesting circuit for
MFCs,144 and our results showed that the new active harvesting approach was much more
efficient than passive charge pump method, as the energy output increased by 76 times.144
In the active harvesting circuit, energy extraction was controlled within a voltage band at
the MFCs maximum power points. The selection of inductance was based on a fixed
condition of the MFC, but the duty ratio and switching frequency were flexibly controlled
based on MFC condition144 In this study, we investigated the energy extraction with
different inductances, duty ratios, and switching frequencies to characterize how these
parameters affect MFC energy output performance. The energy harvesting frequency or
switching frequency of the power converter ranges from 100 to 5000 Hz, which means
that our switching periods (10 msec 200 psec) were orders of magnitude shorter than
previous studies, which were in the range of hours,202 minutes191 and seconds.203,204
5.3 Materials and Methods
5.3.1 MFC construction and operation
As shown in Figure 5.2, a two-chamber MFC reactor with anode and cathode
chamber separated by cation exchange membrane (38 cm2, CMI-7000, Membranes
International) was used in this study.144 The reactor was originally inoculated by
anaerobic sludge from Longmont Wastewater treatment Plant (Longmont, CO) and has
been operated stably for nearly one year. The empty volume of anode or cathode chamber
76


each was 150 mL. Anode was a treated graphite fiber brush (Gordon Brush) and cathode
was a 38 cm2 plain carbon cloth (Fuel Cell Earth). To maintain stable conditions of both
anode and cathode during tests, anolyte and catholyte was separately recirculated from a
1000 mL reservoir. The flow rates of recirculation were 45mL/min and 114 mL/min for
anolyte and catholyte, respectively. The anolyte was sodium acetate dissolved in 50 mM
phosphate buffer containing 1.25g of CFLCOONa, 0.31g of NH4CI, 0.13g of KC1, 3.32g
of NaFEPCE^FEO, 10.32g of Na2HP04- 12FEO, 12.5 mL of mineral solution, and 5 mL
of vitamin solution.177 The catholyte was potassium ferricyanide dissolved in 50 mM
phosphate buffer contains 16.5g of CeLeKsNe, 3.32g of NaFLPCL^FLO and 10.32g of
Na2HP04- I2H2O. All of the tests were conducted at room temperature.
Catholyte
reservoir
£llB*nt Oscilloscope
ogggg
o gggg
Waveform
generator
P
f Electric circuit
Figure 5.1 Schematic Diagram of the Experimental Setup.
77


5.3.2 Energy extraction circuit design
Energy extraction circuit based on boost converter was composed of a metal-
oxide-semiconductor field-effect transistor (MOSFET, Si3460BDV, VISHAY), an
inductor (Triad Magnetics), a Schottky diode (1N5711, Micro Commercial Components)
and a 2 F supercapacitor (PAS1016LR2R3205, Taiyo Yuden). The MOSFET is the main
switch of the circuit and controlled by a 15 MHz function/arbitrary waveform generator
(33120A, Agilent Techonologies). The function/arbitrary waveform generator can
generate square waves in various duty ratios and frequencies that turn MOSFET on and
off to extract energy from MFC in different conditions. The inductor is a temporary
energy storage while the MOSFET is on, and the stored energy is transfered to capacitor
when the MOSFET is off. The Schottky diode blocks reverse power flow from capacitor
to inductor and automatically turns on when the MOSFET is off due to the induced
voltage across the inductor. The capacitor is the terminal energy storage in this study.
Figure 5.2(a) shows block diagram of the energy extraction circuit controlled by
the function generator. The circuit was operated under two modes: CHARGE (Figure
5.2(b)) and DISCHARGE (Figure 5.2(c)). Under CHARGE mode, MOSFET is on and
switching diode is off, energy harvested from MFC is stored temporarily in the inductor;
Under DISCHARGE mode, MOSFET is off and switching diode is on, energy stored in
the inductor is transferred to the capacitor. After alternative operation under CHARGE
and DISCHARGE modes, energy can be cumulated in the capacitor. Detailed operation
of the energy harvester can be found in authors' previous work.144,201
78


(b)
Inductor
(c)
+
MFC

Inductor
Capacitor
Figure 5.2 Block Diagram of Energy Extraction Circuit: (a) Energy Harvesting
Converter, MOSFET is Controlled ON/OFF in Different Duty Ratios and
Frequencies by Function Generator; (b) CHARGE Mode, MOSFET is On and
Switching Diode is Off, Energy Extracted from MFC is Stored in the Inductor
Temporally; (c) DISCHARGE Mode, MOSFET is Off and Switching Diode is On,
Energy Stored in the Inductor is Transferred to the Capcitor.
5.3.3 Tests
In this study, three different inductors (RC-7 (0.45mH), CST206-1A (14mH), and
CST206-3A (130mH), Triad Magnetics), three duty ratios (25%, 50% and 75%) and
seven switching frequencies (100Hz, 500Hz, 1000Hz, 2000Hz, 3000Hz, 4000Hz and
5000Hz) were examined. In each set of the tests, switching frequency was changed with
one fixed inductor and duty ratio. So there were total nine sets and each set includes
79


seven tests. Each test lasts for 30 min to examine one specific configuration of inductor,
duty ratio and switching frequency. Before each test batch, MFC reactor was fed with
new anolyte and catholyte and operated under a 23 Q resistor to facilitate the recovery of
electrochemical active bacteria. A 23Q resistor was used because it is correlated to the
highest power density based on the systems polarization curve. If MFC reactor output
voltage was stable at 3555mV with 23Q, meaning the MFC was maintained at the
maximum power point, the reactor was assumed to be ready for the test. For each test in a
different condition, the reactor was initially kept at open circuit condition until it reached
an open circuit voltage of 7055mV. Then the characterization was conducted from this
open circuit condition.
The MFC voltage, anode potential, cathode potential, capacitor voltage, and
voltage across the current probe were recorded every 66 seconds by data acquisition
system (Model 2700, Keithley Instrument). Anode potential and cathode potential were
measured against an Ag/AgCl reference electrode (RE-5B, Bioanalysis) inserted in anode
chamber and cathode chamber, respectively. A digital storage oscilloscope (Tektronix
TPS2014) was used to continuously monitor MFC voltage, output current and duty ratio.
The energy stored in the storage capacitor (£) was calculated by E = 0.5j CV2dt,
where C is the capacitance, and V is storage capacitor voltage. Energy supplied by the
MFC during harvesting (W) was expressed as W = J VMFCIMFCdt, where Vmfc is the
voltage across the MFC anode and cathode, Imfc is the MFC output current and the
sampling time dt is 66 seconds. Energy harvesting efficiency (.EHE) was calculated by
EHE = E/WXl00%. Duty ratio (D) was defined as D = ton / (ton + lojr), where ton and
80


toff is the on and off time of the MOSFET, respectively 144. Arithmetic mean values of
MFC voltage, MFC current, anode potential, and cathode potential are computed for each
30-minute test.
5.4 Results and Discussion
5.4.1 Effects on MFC voltage and current
RC7: Average MFC Vottage
C5T206-1A: Average MFC Vottage
CST206-3A: Average MFC Vottage
RC7: Average MFC Current CST206-1A: Average MFC Current CST206-3A: Average MFC Current
Figure 5.3 MFC Voltages (a)-(c) and MFC Current (d)-(f) During Energy
Extraction under Different Inductances, Duty Ratios and Frequencies. Left Column:
0.45mH, Middle Column: 14mH, and Right Column: 130mH.
The MFC voltages showed the same trend with different inductances, duty ratios
and frequencies (Figure 5.3(a) 3(c)). MFC voltages decreased with increasing duty
ratios and decreasing frequencies. Higher duty ratios and low frequencies mean that more
time was used for energy extraction from the MFC, so MFC voltage decreases from open
circuit voltage in response to the energy extraction. Comparing different inductances,
small inductances led to lower MFC voltages than larger inductances, because the smaller
81


inductance will introduce larger current. The two larger inductances (14 mH and 130 mH)
showed similar MFC voltages especially with high switching frequencies because of the
similar current amplitudes. The current does not show much difference between 14 mH
and 130 mH when its amplitude is very low.
The relationship between MFC output current and the inductor can be given as
Imfc = 0 / L) J VMFCdt, where L is the inductance, Vmfc is the MFC voltage, and dt is
sampling time. As shown in Figure 5.2(b), the voltage across the inductor Vl is identical
with MFC output voltage when the MOSFET is on in CHARGE mode. Hence, average
MFC output current is inversely proportional to the inductance for a given MFC voltage
output, which can be seen in Figure 5.3(a) 3(f). Although smaller inductance extracts
more current, the corresponding voltage is low therefore resulting in low power output.
The rate of current change also depends on the inductance. As can be seen in the
voltage and current relationship in the differential form, VMFC = L(dIMFC / dt), the current
is increasing fast if the inductance is small for a given voltage. This causes a fast decrease
of MFC voltage as well in DISCHARGE period, which generates fluctuations in MFC
output voltage, current, and power in the given switching frequency. On the contrary,
larger inductance will make them less fluctuating and closer to a constant value due to
smaller dl/dt. Typical instantaneous MFC current and voltage waveforms of small and
large inductance are shown in Figure 5.4(a) and (b). It can be seen that the voltage and
current are smoothed and the current level is low with the high inductance.
82


(a) (b)
Figure 5.4 Waveforms of MOSFET Gating Signal (Top), MFC Votlage (Middle),
and MFC Current (Bottom). MOSFET is Turned on When the Gating Signal is in
High State, (a) Inductance 0.45mH, Switching Frequency 100Hz, Duty Ratio 50%;
(b) Inductance 14mH, Switching Frequency 1000 Hz, Duty Ratio 50%.
The MFC voltage and current are also a function of duty ratio. As the duty ratio
represents the conduction time of MOSFET, which is the time duration for MFC output
terminals connected to the inductor, the average current increases and voltage decreases
with increasing duty ratio. Under a low duty ratio, ON time is short and fewer electrons
are extracted, so the produced current is low although there is a long OFF time to
replenish electron donors; if ON time is long and more electrons are extracted with a high
duty ratio, so the produced current becomes high but the short OFF time may reduce time
for electron donor replenishment. Hence, there should be an optimum duty ratio to
balance ON and OFF time for a given MFC condition. The experiment with lower
inductance case shows more linear relationship between voltage and current in terms of
duty ratio. The generated energy surface in low inductance experiment shows clear peak
point similar to the polarization curve as can be seen in Figure 5.6(a). For higher
inductances, the current and voltage do not change as much (Figure 5.3(a) 3(f)).
83


The switching frequency also affects the MFC voltage and current. Similar as the
duty ratio, it has higher effects when inductance is smaller. For a given duty ratio, MFC
generates more current with lower switching frequency because more electrons were
extracted at low switching frequency and the reactor can have more time for recovery.
This should be very important factor for switching converter-based energy harvesting
system designs, because MFC voltage may collapse without proper amount of recovery
time. However, due to the high switching frequencies (100Hz 5000Hz) used in this
study, the cycle times were very short ranging from 200 psec to 10 msec. Therefore,
when energy was extracted within this short period, there were enough electrons
available around the electrode for next extraction because only a small portion of the
electrons was pulled out. At each duty ratio, current decreased as switching frequency
increased because fewer electrons were extracted due to short energy extraction time. The
MFC current and voltage showed a tendency to be stabilized after a certain switching
frequency (Figure 5.3(a) 3(f)). In the experiments in this paper, recirculation of anolyte
helped improve the mass diffusion and replenish electrons at the electrode.
As can be seen in Figure 5.3(a) (f), cathode potentials showed a significant
differences for a given duty ratio and switching frequency between low (0.5mH) and high
inductances (14mH and 130mH) even with enough catholyte provided during the tests,
compared to the anode potentials that were relatively stable. This result suggested that the
different duty ratios and frequencies in energy harvesting would not affect the activity of
mature anode biofilm (developed on the anode for nearly one year). This confirms that
the duty cycling itself had little or no effect on gross community composition on the
anode. The stable anode potential also suggested that enough electrons which could be
84


Full Text

PAGE 1

NEW ELECTRODE MATERIALS AND ACTIVE ENERGY HARVESTIN G FOR MICROBIAL ELECTROCHEMICAL SYSTEMS, OR MXCS by Heming Wang B.S, Harbin Institute of Technology (China), 2006 M.S, Harbin Institute of Technology (China), 2008 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Civil Engineering 2013

PAGE 2

ii This thesis for the Doctor of Philosophy degree by Heming Wang has been approved for the Civil Engineering Program by Peter Jenkins, Chair Zhiyong “Jason” Ren, Advisor Jae-Do Park JoAnn Silverstein Angela Bielefeldt April 18, 2013

PAGE 3

iii Wang, Heming (Ph.D., Civil Engineering) New Electrode Materials and Active Energy Harvestin g for Microbial Electrochemical Systems, or MXCs Thesis directed by Assistant Professor Z. Jason Ren ABSTRACT Microbial electrochemical system or MXC is an emerg ing platform technology that integrates multiple disciplines and carries di fferent functions, such as waste treatment, environmental remediation, and desalination with si multaneous power or value-added chemical production. The versatility and high perfo rmance distinguishes MXC from traditional treatment-focused and energy-intensive environmental systems and makes it a potential transformative technology for the next ge neration of environmental biotechnology. The Chapter 1 of this dissertation s ummarizes nearly 50 corresponding reactor systems developed based on the MXC platform with the goal to introduce and summarize all the functions that have been develope d so far and discusses the niche of this technology for environmental science and engin eering. The MXC technology carries great potentials in tran sforming waste treatment process into energy and resource recovery systems, but the technology is still in early stage development, and there are many remaining cha llenges need to be addressed. In addition to comprehensive literature review, my doc toral study has been focused on the following three aspects on MXC scientific character ization and engineering development: (1) Development and characterization of high perfor ming and low-cost electrode materials. The anode and cathode are two crucial el ements in MXC structures and largely determine the performance and cost of a reactor sys tem. The feasibility of using a

PAGE 4

iv recycled material – tire crumb rubber, as a low-cos t anode material in microbial fuel cells (MFC) was investigated (Chapter 2) in my first year of study. This is the first study that used recycled materials as an alternative to tradit ional graphite or carbon cloth anodes, the results show that tire crumbs produced comparab le level of electricity after surface treatment. On the cathode side, three types of new carbon nanotube based air-cathode with very high surface area were developed, which c reated a 3-D structure and demonstrated superior performance as compared to tr aditional carbon cloth air-cathodes (Chapter 3). (2) Innovative active energy harvesting during orga nic removal and waste treatment. MXC can produce direct electricity from waste materials, but currently the current and voltage output are low, making the dire ct use of such power difficult. A new active energy harvesting approach using a maximum p ower points circuit (MPPC) was developed, which totally changed the traditional pa ssive energy gaining process through resistors or non-controllable charge pumps. The sys tem dramatically increased MFC power production by more than 70 times and improved energy recovery by 20 times (Chapter 4). Furthermore, the effects of inductance duty ratio, and switching frequency on MFC energy harvesting were characterized to opti mize operating conditions and direct further circuit development (Chapter 5). (3) Removal mechanisms of emerging trace organic co mpounds (TOrCs) in microbial fuel cells. The MXC can theoretically deg rade any biodegradable materials to produce energy, and many different kinds of waste s treams have been tested. The removal of traditional organic matter and sulfur, a nd the recovery of nitrogen and phosphate have all been proved possible. The feasib ility of removal and transformation of

PAGE 5

v low level (e.g. ng/L) trace organic compounds (TOrC s) in wastewater was studied in microbial fuel cells (Chapter 6), because it was hy pothesized that MFC can be effective in this task by providing both reductive (anode) an d oxidative (cathode) environments, and the biodegradation can be enhanced with electro chemical transformation. Results show that MFC was effective in removing most of the 34 TOrCs tested with different efficiencies, and electricity was produced from all reactors. The form and content of this abstract are approved. I recommend its publication. Approved: Z. Jason Ren

PAGE 6

vi DEDICATION I dedicate this doctoral dissertation to my husband Tianhua Guo, and the Wang and Guo families for their continuous love and supp ort.

PAGE 7

vii ACKNOWLEDGMENTS I would like to express my sincerest gratitude to m y advisor, Dr. Zhiyong “Jason” Ren, who offered me the opportunity to the Ph.D program and walked me through all the difficulties in my study and research. I am deeply grateful of his support, guidance, and encouragement, which not only helped me finish the doctoral study, but also will benefit my professional development. I would also like to t hank Dr. Jae-Do Park from Electrical Engineering, for his great guidance and tremendous help in the energy harvesting project. I am grateful to Dr. Zhuangchun Wu, Dr. Peter Jenki ns, Dr. Atousa Plaseied, and Dr. Pei Xu for their insightful and constructive discussion s. I thank Dr. JoAnn Silverstein and Dr. Angela Bielefeldt for serving in the committee and providing great advices for my research and course work. My research was supported by the Ofce of Naval Research (ONR) and US National Science Foundation (NSF). I am grateful to all my committee members for their valuable participation and helpful comments. My thanks would also go to all th e members in the lab for their direct and indirect assistance. Finally, I also owe a special debt of gratitude to my family who are halfway across the globe. I thank my parents for their unco nditional support and love. I also owe my deepest gratitude to my beloved husband who is a lways there waiting for me all these years.

PAGE 8

viii TABLE OF CONTENTS CHAPTER 1. The X Factor: a Review of Bioelectrochemical Sys tem as a Platform Technology ....... 1 1.1 Abstract ...................................... ................................................... ................................ 1 1.2 Introduction .................................. ................................................... .............................. 2 1.3 The Shared Principle in the BES Anode Chamber ................................................... ... 3 1.4 The X Factor .................................. ................................................... .......................... 10 1.5 MFC-based Systems for Electricity Generation .. ................................................... .... 12 1.5.1 Wastewater microbial fuel cell (wastewater MF C) ................................................ 12 1.5.2 Benthic microbial fuel cell (benthic MFC) ... ................................................... ........ 14 1.5.3 Microbial remediation cell (MRC) ............ ................................................... ........... 15 1.5.4 Microbial solar cell (MSC) .................. ................................................... ................. 17 1.6 MEC-based Systems for Chemical Production...... ................................................... .. 19 1.7 MES-based Systems for Chemical Production ..... ................................................... ... 21 1.8 MDC-based Systems for Water Desalination and Be neficial Reuse .......................... 22 1.9 Outlook ....................................... ................................................... ............................. 24 1.10 Acknowledgement .............................. ................................................... ................... 26 2. Recycled Tire Crumb Rubber Anodes for Sustainabl e Power Production in Microbial Fuel Cells ........................................ ................................................... ............................... 27 2.1 Abstract ...................................... ................................................... .............................. 27 2.2 Introduction .................................. ................................................... ............................ 28 2.3 Materials and Methods ......................... ................................................... .................... 29 2.3.1 MFC construction and operation .............. ................................................... ............ 29 2.3.2 Statistical and electrochemical analyses .... ................................................... ........... 30 2.4 Results and Discussion ........................ ................................................... .................... 32

PAGE 9

ix 2.4.1 Resistance characterization of coated crumb r ubber electrode ................................ 32 2.4.2 Surface characterization of coated crumb rubb er electrode ..................................... 33 2.4.3 Power production from tire rubber MFCs and gr aphite granule MFCs................... 35 2.4.4 Cost-benefit outlook......................... ................................................... ..................... 36 2.5 Conclusions ................................... ................................................... ........................... 37 2.6 Acknowledgements .............................. ................................................... .................... 37 3. Carbon Nanotube Modified Air-cathodes for Electr icity Production in Microbial Fuel Cells ............................................. ................................................... .................................. 38 3.1 Abstract ...................................... ................................................... .............................. 38 3.2 Introduction .................................. ................................................... ............................ 39 3.3 Materials and Methods ......................... ................................................... .................... 41 3.3.1 Cathode construction ........................ ................................................... .................... 41 3.3.2 MFC construction and operation .............. ................................................... ............ 42 3.3.3 Electrochemical and Microscopy Analysis ..... ................................................... ...... 43 3.4 Results and Discussion ........................ ................................................... .................... 44 3.4.1 Electrochemical performance ................. ................................................... .............. 44 3.4.2 Performance of MFCs with nano-modified air-ca thodes ......................................... 46 3.4.3 FIB / SEM analysis .......................... ................................................... ..................... 50 3.5 Conclusions ................................... ................................................... ........................... 51 3.6 Acknowledgement ............................... ................................................... .................... 52 4. Active Energy Harvesting from Microbial Fuel Cel ls at the Maximum Power Point without Using External Resistors................... ................................................... ................ 53 4.1 Abstract ...................................... ................................................... .............................. 53 4.2 Introduction .................................. ................................................... ............................ 54 4.3 Materials and Methods ......................... ................................................... .................... 56 4.3.1 MFC construction and operation .............. ................................................... ............ 56

PAGE 10

x 4.3.2 Maximum power point circuit (MPPC) design and operation ................................. 57 4.3.3 Analyses .................................... ................................................... ............................ 60 4.4 Results and Discussion ........................ ................................................... .................... 61 4.4.1 MPPC can operate the MFC at the maximum power harvesting range ................... 61 4.4.2 MPPC harvests energy more actively and effici ently ............................................. 63 4.4.3 The numbers of capacitors for energy storage ................................................... ...... 65 4.4.4 Conversion efficiency of the MPPC ........... ................................................... .......... 67 4.5 Outlook ....................................... ................................................... ............................. 70 4.6 Acknowledgement ............................... ................................................... .................... 71 5. Power Electronic Converters for Microbial Fuel C ell Energy Extraction: Effects of Inductance, Duty Ratio, and Switching Frequency ... ................................................... .... 72 5.1 Abstract ...................................... ................................................... .............................. 72 5.2 Introduction .................................. ................................................... ............................ 74 5.3 Materials and Methods ......................... ................................................... .................... 76 5.3.1 MFC construction and operation .............. ................................................... ............ 76 5.3.2 Energy extraction circuit design ............ ................................................... ............... 78 5.3.3 Tests ....................................... ................................................... ............................... 79 5.4 Results and Discussion ........................ ................................................... .................... 81 5.4.1 Effects on MFC voltage and current .......... ................................................... ........... 81 5.4.2 Effects on MFC Energy and Efficiency ........ ................................................... ........ 85 5.4.3 Discussion .................................. ................................................... ........................... 88 5.5 Acknowledgement ............................... ................................................... .................... 89 6. Removal Mechanisms of Trace Organic Compounds in Microbial Fuel Cells .......... 90 6.1 Abstract ...................................... ................................................... .............................. 90 6.2 Introduction .................................. ................................................... ............................ 91 6.3 Materials and methods ......................... ................................................... .................... 91

PAGE 11

xi 6.3.1 MFC construction and operation .............. ................................................... ............ 92 6.3.2 Experimental Procedures ..................... ................................................... ................. 93 6.3.3 Analysis..................................... ................................................... ............................ 97 6.4 Results and Discussion ........................ ................................................... .................... 97 6.4.1 Performance of Single-chamber and Two-chamber MFCs ..................................... 97 6.4.2 TOrCs removal in Single-chamber and Two-chamb er MFCs ............................... 100 References .................................................. ................................................... .................. 105

PAGE 12

xii LIST OF TABLES Table 1.1 Summary of All Types of BES/MXC. ............. ................................................... ......... 4 3.1 List of Cathode Materials and Modifications Us ed in This Study and Their Specifications. ................................... ................................................... ............................. 41 4.1 Analysis of Energy Extraction Efficiency by MP PC. ............................................... 69 6.1 34 TOrCs Detected by LC-MS/MS in ESI(+) and ES I(-) Methods and Selected Physicochemical Properties. ....................... ................................................... ................... 94

PAGE 13

xiii LIST OF FIGURES Figure 1.1 Number of Published Journal Articles on MXCs C ontaining the Phrases “Microbial Fuel Cell”, “Microbial Electrolysis Cell”, “Microbi al Desalination Cell” or “Microbial Electrosynthesis” (Source: Scopus on 3/8/2013; Docu ment Type: Journal; Language: English; Remove Duplicates from Searching Results)……………………………………………… ................................................. ..... 3 1.2 Basic Principles in Four Typical BESs (Left Ch amber: Anode; Right Chamber: Cathode). (A) Electricity Generation in Air-cathode Microbial Fuel Cell (MFC); (B) Hydrogen Generation with External Power Supply in M icrobial Electrolysis Cell (MEC); (C) Chemical Production by Microbial Electrosynthes is (MES); (D) Middle Chamber Desalination by Electric Drive in Microbial Desalin ation Cell (MDC)……………………………………………… ................................................. ..... 11 1.3 MFC-based Systems for Electricity Generation: (A) Wastewater Microbial Fuel Cell,5 (B) Benthic Microbial Fuel Cell,119 (C) Microbial Remediation Cell,126 and (D) Microbial Solar Cell.111……………………………………………… ............................ 13 1.4 Some Advanced MXC Systems: (A) Microbial Rever se-electrodialysis Electrolysis Cell (MREC) for H2 Production,71 (B) Microbial Electrosynthesis (MES) for Organic Synthesis,76 and (C) Microbial Capacitive Desalination Cell (MC DC) for Desalination.86……………………………………………… .......................................... 20 2.1 Box Plot of Resistance Measurement and Statist ics on Tire Crumb Particle Surface with Different Coating Layers……………………………………………… ................. 32 2.2 System Resistance of Single Chamber Bottle Rea ctor Filled with Graphite Granules and Tire Particles with Different Coating Layers……………………………………………… ................................................. ...... 33 2.3 Pore Size Distribution of (A) Rubber Particle with 4-layer Coating, and (B) Graphite Granule as the MFC Anode……………………………………………… ..................... 34 2.4 Voltage and Power Density as a Function of Cur rent Density for Coated Tire Anode MFCs and Graphite Granule Anode MFCs …………………………………… ………… ................................................... ................................................... ..................................... 35 3.1 LSV Results (Current Density vs Potential) of Newly Modified Cathodes Before Installing in MFCs. Current Density Range Was Marke d Based On the Values Shown in Figure 3.3……………………………………………… ................................................. 45 3.2 Voltage Generation as a Function of Time for t he Different Cathodes ……………………………………………… ................................................... ................ 46

PAGE 14

xiv 3.3 Power Density as a Function of Current Density (A) and Polarization Curves (B) for MFCs Operated Using Different Air-cathodes Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... ................ 48 3.4 Comparison of LSV Electrochemical Test Results between New and Used Cathodes of CC-Pt and SWNTn-PtÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ......................... 49 3.5 SEM/FIB Images of New and Used Cathodes: (A) N ew CC-Pt, (B) Used CC-Pt After MFC Operation, (C) New CNTM, (D) Used CNTM-Pt After MFC Operation, (E) New SWNTn-Pt, and (F) Used SWNTn-Pt After MFC OperationÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... 51 4.1 Components in MPPCÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ...................... 57 4.2 Block Diagram of the Maximum Power Point Circu it (MPPC): (A) Harvesting Converter Controller. (B) Whole electric Circuit Di agram; (C) CHARGE Phase, MOSFET is On While Diode is Off, Extracted Energy i s Stored in the Inductor; (D) DISCHARGE Phase, MOSFET is Off While Diode is On, E xtracted Energy is Stored in the Capacitors..................................... ................................................... ............................ 59 4.3 MFC Polarization Curve and Power Density Curve Obtained by Linear Sweep Voltammetry (LSV). The Scan Rate of the Polarizatio n was 0.1 mV/s. : Operating Point of the Charge Pump. : Operating Range of the MPP C. Recirculating-flow MFC Open Circuit Potential was 688 mVÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… .................. 61 4.4 Snapshot of On/Off Cycle of the MPPC During Ac tive Energy Harvesting from MFCs and the Voltage and Current Profiles. One divi sion of X-Axis Represents 100 sec. The Figure Shows the Waveforms of 1 msec Dura tion in Terms of Current, Voltage, and On/Off Switch ChangesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… Â… ...... 62 4.5 Comparison of MFC Voltage, Cathode Potential, and Anode Potential between the MPPC Active Energy Harvesting Condition and 23 Ohm External Resistor Condition. The Optimum External Resistance was Calculated to b e 23 ohm Based on Polarization Curve that Could Yield the Maximum Power DensityÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................. .... 64 4.6 (A) Comparison of Energy Harvesting by the MPP C and the Charge Pump and Energy Stored in Capacitors. (B) Comparison of COD Removals in the MPPC and Charge Pump Conditions. In the MPPC Test, 12 Capaci tors were Connected in Parallel for Energy Storage. In the Charge Pump Test, one Ca pacitor was Enough to Store All the Harvested Energy from MFCÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… .................... 65 4.7 (A) Voltage Profile and (B) Energy Storage Dif ferences by Using 3, 6, 9, and 12 Capacitors in Parallel During MPPC Active Energy Ha rvesting Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... ................ 66 4.8 Efficiencies through 18-hour Test. ........... ................................................... .............. 68

PAGE 15

xv 4.9 Energy Conversion Efficiency and Distribution of Internal Energy Loss in the MPPC. The Distribution was Quantified Based on an 1 8-hour, 12-capacitor OperationÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... 69 5.1 Schematic Diagram of the Experimental Setup Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... ................ 77 5.2 Block Diagram of Energy Extraction Circuit: (a ) Energy Harvesting Converter, MOSFET is Controlled ON/OFF in Different Duty Ratio s and Frequencies by Function Generator; (b) CHARGE Mode, MOSFET is On and Switch ing Diode is Off, Energy Extracted from MFC is Stored in the Inductor Tempor ally; (c) DISCHARGE Mode, MOSFET is Off and Switching Diode is On, Energy Sto red in the Inductor is Transferred to the Capcitor. .................................. ................................................... ............................. 79 5.3 MFC Voltages (a)-(c) and MFC Current (d)-(f) D uring Energy Extraction under Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130mHÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… Â…Â…Â…Â… ................................................... ................................................... ..................................... 81 5.4 Waveforms of MOSFET Gating Signal (Top), MFC V otlage (Middle), and MFC Current (Bottom). MOSFET is Turned on When the Gati ng Signal is in High State. (a) Inductance 0.45mH, Switching Frequency 100Hz, Duty Ratio 50%; (b) Inductance 14mH, Switching Frequency 1000 Hz, Duty Ratio 50%.Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................. ........ 83 5.5 Anode Potentials (a)-(c) and Cathode Potential s (d)-(f) During Energy Extraction under Different Inductances, Duty Ratios and Freque ncies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130mH.Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... ... 85 5.6 MFC Generated Energy (a)-(c), Harvested Energy (d)-(f), and Efficiencies (g)-(i) During Energy Extraction Under Different Inductance s, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130mH.Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................... ... 86 5.7 Waveforms of MFC Current (Top), MFC Votlage (M iddle), and MOSFET Gating Signal (Bottom). MOSFET is Turned on When the Gatin g Signal is in High State. (a) Inductance 0.45 mH, Switching Frequency 1000 Hz, Du ty Ratio 25%; (b) Inductance 130 mH, sSwitching Frequency 5000 Hz, Duty Ratio 25%; ( c) Inductance 0.45mH, Switching Frequency 100Hz, Duty Ratio 50%; (d) Inductance 14m H, Switching Frequency 500 Hz, Duty Ratio 50%; (e) Inductance 0.45 mH, Switching F requency 2000 Hz, Duty Ratio 75%; (f) Inductance 14 mH, Switching Frequency 5000 Hz, Duty Ratio 75%.Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… ................................................. ........ 87 6.1 Polarization Curve and Power Density Curve Obt ained by Linear Sweep Voltammetry (LSV) in Single-chamber (A) and Two-cha mber Reactors (B) Filled by Sodium Acetate Spiked with TOrCs (Dark Blue) and wi thout TOrCs (Orange). The Scan Rate of the Polarization Was 0.1 mV/s. ............ ................................................... ............. 98

PAGE 16

xvi 6.2 Total Organic Carbon (TOC) Removal in Single-c hamber and Two-chamber Reactors Filled by Sodium Acetate Spiked with TOrCs under 167 ohm and Open Circuit. ................................................... ................................................... ................................... 100 6.3 TOrCs Removal in Single-chamber Reactors and C athode-chamber of Two-chamber Reactors (A) and Anode-chamber of Two-chamber React ors (B) Filled by Sodium Acetate Spiked with TOrCs under 167 ohm and Open Ci rcuit. TOrCs are Divided into Three Categories Based on Charge. Biodegradibility Probability is Indicated in Parentheses after the Name of Each Compound. ...... ................................................... ... 103

PAGE 17

1 1. The X Factor: a Review of Bioelectrochemical Sys tem as a Platform Technology1 1.1 Abstract Bioelectrochemical system (BES) is an emerging tech nology that uses microorganisms to covert the chemical energy stored in biodegradable materials to direct electric current and chemicals. Compared to traditi onal treatment-focused, energyintensive environmental technologies, BES offers a new and transformative solution for integrated waste treatment and energy and resource recovery, because it offers a flexible platform for both oxidation and reduction reaction oriented processes. All BESs share one common principle in the anode, in which biodegr adable substrates, such as waste materials, are oxidized and generate electrical cur rent. In contrast, a great variety of applications have been developed by utilizing this in situ current, such as direct power generation (microbial fuel cell, MFC) or chemical p roduction (microbial electrolysis cell, MEC). This study provides a comprehensive review of the different functions developed based on the BES platform to date and summarized ne arly 50 corresponding systems as MXCs – with the X standing for the different functi ons and systems. It also discusses the “X factor” the future development of this promisi ng yet early-stage technology. 1 The work presented in this chapter is co-authored b y Heming Wang and Zhiyong Ren and in review by Energy Environ. Sci

PAGE 18

2 1.2 Introduction Bioelectrochemical system (BES), microbial electroc hemical system (MES), or MXC are different collective names for an emerging environmental technology called microbial electrochemical technology (MET).1-4 While this platform technology has only been intensively studied and developed in the past decade, it opens up a new interdisciplinary field for research and developmen t which integrates microbiology, electrochemistry, materials science, engineering, a nd many related areas together. BES not only provides a unique environment to understan d the largely unexplored microbial electrochemistry, it also offers a flexible platfor m for many different engineering functions to be developed. While many existing envi ronmental technologies have only one or two functions, the BES platform is so flexib le that dozens of functions have been discovered. Almost all BESs share one common princi ple in the anode, in which biodegradable substrates, such as waste materials, are oxidized by microorganisms and generate electrical current. The current can be cap tured directly for electricity generation (microbial fuel cells, MFCs),5-7 or used to produce hydrogen, methane, and other va lueadded chemicals (microbial electrolysis cells, MECs ).8-10 The electrons can also be used in the cathode chamber to synthesize organic compou nds (microbial electrosynthesis, MES) or remediate contaminants (microbial remediati on cells, MRCs).11-15 The potential across the electrodes can also drive desalination ( microbial desalination cells, MDCs).1620 The production of current associated with microbia l catabolism was first reported a century ago by M. C. Potter,21 but research interests in this concept have only b lossomed in the past decade, resulting in an exponential gro wth in the number of journal articles (Figure 1.1). There are several excellent reviews t hat provided information on the history

PAGE 19

3 and development of BESs22-25 and the substrates, materials, and microbial commu nities in BESs,26-30 but there has been no comprehensive review that di rectly addresses the factor of “X”, where all the known functions were originat ed from and future functions will be based upon. As shown in Table 1.1, this article aim s to provide the first comprehensive review to summarize all the X functions that have b een developed using the BES platform and shed lights on future system developme nt for energy and environmental science and engineering. Figure 1.1 Number of Published Journal Articles on MXCs Containing the Phrases “Microbial Fuel Cell”, “Microbial Electrolysis Cell ”, “Microbial Desalination Cell” or “Microbial Electrosynthesis” (Source: Scopus on 3/8/2013; Document Type: Journal; Language: English; Remove Duplicates from Searching Results). 1.3 The Shared Principle in the BES Anode Chamber Compared to traditional chemical fuel cells, the BE S platform uses low-cost and self-sustaining microorganisms to oxidize organic a nd inorganic substrates, mainly waste materials, and transfer electrons to the anode elec trode. This microbial oxidation reaction is a shared principle for almost all BES or MXC rea ctors, as shown in Table 1.1. However, how to use these electrons on the cathode side gives the most beauty of this 0 50 100 150 200 250 300 350 400 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Number of journal articlesYear MES MDC MEC MFC

PAGE 20

4 platform technology, because any reduction-based re action can be realized in the cathode chamber which creates numerous possibilities. Based on the different functions, the BES platform has been specified into many different nam es that researchers name them MXCs, where X stands for the different applications.2, 4 Table 1.1 summarizes all the Xs to date and demonstrates the shared principle on the anode and the versatile functions on the cathode. Table 1.1 Summary of All Types of BES/MXC. Types of BES/MXC Electron donor for anode oxidization Electron acceptor for cathode reduction Main Products Ref. MFC-based systems for electricity generation Microbial Fuel Cell (MFC) – in general Any biodegradable material Oxygen, Potassium ferricyanide, or other oxidants Electricity 31 32 Tubular Microbial Fuel Cells (Tubular MFC) Acetate, glucose, domestic wastewater, hospital wastewater, digester effluent from a potato processing plant Potassium ferricyanide Electricity 33 Upflow Microbial Fuel Cell (UMFC) Sucrose Potassium ferricyanide, oxygen Electricity 34 35 Bafed Air-cathode Microbial Fuel Cell (BAFMFC) Glucose, liquid from corn stover steam explosion process Oxygen Electricity 36 Stacked Microbial Fuel Cell (Stacked MFC) Sodium acetate Potassium ferricyanide Electricity 37 Submersible Microbial Fuel Cell (SBMFC) Domestic wastewater Oxygen Electricity 38 Benthic Microbial Fuel Cell (BMFC) Sediment Oxygen Electricity 39 41

PAGE 21

5 Types of BES/MXC Electron donor for anode oxidization Electron acceptor for cathode reduction Main Products Ref. Sediment Microbial Fuel Cell (AKA Benthic Unattended Generator or BUG) Acetate and other fermentation products in the sediment Oxygen Electricity 30 Self-stacked Submersible Microbial Fuel Cell (SSMFC) Sediment, acetate Oxygen Electricity 42 Microbial Remediation Cell (MRC) Diesel, ethanol, 1,2-dichloroethane, pyridine, phenol Chlorinated solvents, perchlorate, chromium, and uranium Reduced/ non-toxic chemicals 11 13 43-48 Photo-Microbial Fuel cell (p-MFC) Water Potassium ferricyanide Electricity 49 Microbial Photoelectrochemical Solar Cell Marine sediment Oxygen Electricity, glucose, oxygen 50 Solar-powered Microbial Fuel Cell Succinate, propionate Oxygen Electricity, hydrogen 51 52 Photobioelectrochemical Fuel Cell Organic acids, alcohols Potassium ferricyanide Electricity, hydrogen 53 Photosynthetic Microbial Fuel Cells (PMFC) Water Oxygen Electricity 54 Photosynthetic Electrochemical Cell Water, glucose Potassium ferricyanide Electricity 55 Solar-driven Microbial Photoelectrochemical Cell (Solar MPC) Trypticase soy broth (TSB) Proton Electricity 56 Plant Microbial Fuel Cell (PMFC) Plant-derived organics (root exudates) Oxygen, potassium ferricyanide Electricity 57 Phototrophic Microbial Fuel Cells (Phototrophic MFC) Sediment Oxygen Electricity 58 Table 1.1 (conÂ’t)

PAGE 22

6 Types of BES/MXC Electron donor for anode oxidization Electron acceptor for cathode reduction Main Products Ref. Photosynthetic Algal Microbial Fuel Cell (PAMFC) Algae Potassium ferricyanide Electricity 59 Microbial Electrochemical Snorkel (MES, AKA short-circuited microbial fuel cell) Wastewater Oxygen Treated wastewater, no electricity 60 Acid-mine drainage fuel cell (AMD-FC) Ferrous ion Oxygen Electricity, removing iron 61 Integrated photobioelectrochemical system (IPB) Wastewater Oxygen Electricity, algal biomass 62 Osmotic Microbial Fuel Cell (OsMFC) Sodium acetate Oxygen Diluted draw solution, electricity 63 Microbial Reverse Electrodialysis Cell (MRC) Sodium acetate Oxygen Electricity 64 65 MEC-based systems for chemical production Microbial Electrolysis Cell (MEC) – in general Any biodegradable material Proton Hydrogen, hydrogen Peroxide, methane, sodium hydroxide 8 66 68 Bioelectrochemically assisted microbial reactor (BEAMR) Wastewater Proton Hydrogen 69 Solar-powered Microbial Electrolysis Fuel (Solar MEC) Acetate Proton Hydrogen 70 Table 1.1 (con’t)

PAGE 23

7 Types of BES/MXC Electron donor for anode oxidization Electron acceptor for cathode reduction Main Products Ref. Microbial Reverse-electrodialysis Electrolysis Cells (MREC) Acetate Proton Hydrogen 71 Microbial Electrolysis Struvite-precipitation Cell (MESC) Sodium acetate Proton Hydrogen, struvite 72 Submersible Microbial Electrolysis Cell (SMEC) Acetate Proton Hydrogen 73 MES-based systems for chemical production Microbial Electrosynthesis (MES) – in general organic, poised anode, hydrogen sulfide Acetic acid or other organics, carbon dioxide Ethanol, Acetate, 2-oxobutyrate, formate 14 74-78 Microbial Carbon Capture Cell (MCC) Glucose Carbon dioxide Algal biomass, electricity 79 MDC-based systems for water desalination and benefi cial reuse Microbial Desalination Cell (MDC) – in general Any biodegradable material Oxygen, potassium ferricyanide, organics, or other oxidants Desalinated water 16 Microbial Saline-wastewater Electrolysis Cell (MSC) Sodium acetate Hydrogen Treated saline wastewater, electricity 80 Osmotic MDC (OsMDC, MODC) Sodium acetate, xylose, wastewater Oxygen, potassium ferricyanide, proton Desalinated water, electricity 81 82 Microbial Desalination Cell with capacitive adsorption capability (cMDC) Sodium acetate Potassium ferricyanide Desalinated water 83 Table 1.1 (con’t)

PAGE 24

8 Types of BES/MXC Electron donor for anode oxidization Electron acceptor for cathode reduction Main Products Ref. Microbial Desalination Cell packed with ion-exchange resin (R-MDC) Sodium acetate Oxygen Desalinated water, electricity 84 Microbial Electrolysis Desalination Cell (MEDC) Sodium acetate Proton Hydrogen, desalinated water 19 Microbial Electrolysis Desalination and Chemicalproduction Cell (MEDCC) Sodium acetate Oxygen Desalinated water, sodium hydroxide, hydrochloric acid 85 Microbial Capacitive Desalination Cell (MCDC) Sodium acetate Oxygen Desalinated water 86 Capacitive Deionization coupled with Microbial Fuel Cells (CDI-MFC) Sodium acetate Potassium ferricyanide Desalinated water 87 Upow Microbial Desalination Cell (UMDC) Sodium acetate Oxygen Desalinated water, electricity 18 Stacked Microbial Desalination Cells (SMDC) Sodium acetate Oxygen Desalinated water, electricity 88 Recirculation Microbial Desalination Cell (rMDC) Xylose Oxygen Desalinated water, electricity 89 Ideal anodic reactions in BESs or MXCs generally in clude dynamic and effective microbial activity and community, higher substrate conversion rate and electron transfer efficiency, and lower material and system cost. The conversion of chemical energy to electrical energy in the BES anode chamber requires the respiration of the insoluble Table 1.1 (conÂ’t)

PAGE 25

9 anode, where a unique group of microbes called elec trochemically active bacteria (EAB), exoelectrogen, electricigen, or anode respiring bac teria have been used.29, 30, 90, 91 Such microorganisms are able to transfer electrons out o f cell membranes to the electrode either directly through immobilized structures or u sing mobile electron shuttles. For example, recent studies showed that Geobacter sulfurreducens requires conductive pili as nanowires for cell-to-cell electron conduction and c -type cytochrome OmcZ to promote electron transfer onto the electrode.92, 93 In contrast, Shewanella species were reported to make both direct electrode contact through conducti ve filaments and indirect electron transfer via mediators, such as riboflavin or flavi n ademine mononucleotide (FMN).94-96 Many other bacteria can produce and use soluble red ox mediators or electron shuttles, which transport the electrons from the cell to the electrode. For example, Pseudomonas species can produce phenazines as extracellular ele ctron shuttles, and other bacteria can use externally provided mediators, such as neutral red, anthraquinone-2,6-disulfonate (AQDS), thionine, methyl viologen, methyl blue, and some humics.13, 97-101 Using microorganisms as biocatalysts, BESs or MXCs can theoretically be used to convert any biodegradable substrate into energy and chemicals. Besides simple sugars and derivatives used in most lab scale studies, man y complex waste materials have also been utilized, such as different wastewaters from m unicipal and industrial sources, biomass waste, and inorganic substrates such as amm onia, sulfide, and acid mine drainage.27, 61, 102-104 The utilization of complex waste materials require s the cooperation of polymer-degrading bacteria and electrochemically active bacteria, with the first group breaks down the complex organic matters into monome rs or solvents, and the second group oxidizes the fermentation products with the a node serving as the electron

PAGE 26

10 acceptor.6, 105-107 In terms of waste treatment in the anode chamber, BES represents a new generation of technology, because it carries the po tential to transform traditional energyintensive, treatment-focused processes into integra ted systems that recover energy, nutrient, water, and other value-added products. 1.4 The X Factor As shown in Table 1.1, there have been nearly 50 sy stems with different functions developed using the BES platform, and people used M XCs to represent the different functions and systems. Though no specific rules hav e been established to name the Xs, this article attempts for the first time to summari ze and categorize all the Xs that have been reported so far and provides some insights on the unknown “X factor” regarding to technology development. In general, an X simply presents the main function and benefit of a specific cell. For example, microbial fuel cell (MFC) is the very original type BES or MXC, whose main function is direct electricity generation (Fig ure 1.2A).108 When an external power source is added in an MFC reactor to reduce cathode potential, the system becomes a microbial electrolysis cell (MEC), where hydrogen g as, methane gas, and other products can be generated through electrolysis (Figure 1.2B) .8, 9, 67-69 If the main function of the system is to use the cathode to reduce oxidized con taminants, such as uranium, perchlorate or chlorinated solvents, the cell can b e named microbial remediation cell (MRC),11-13 and if the main goal of the system is to synthesiz e value-added bio-chemicals through cathodic reduction, the system can be named microbial electrosynthesis (MES) (Figure 1.2C).14, 15 Another system called microbial desalination cell (MDC) (Figure

PAGE 27

11 1.2D),16, 71 includes additional chambers between the anode and cathode and uses the internal potential to drive water desalination. Figure 1.2 Basic Principles in Four Typical BESs ( Left Chamber: Anode; Right Chamber: Cathode). (A) Electricity Generation in Ai r-cathode Microbial Fuel Cell (MFC); (B) Hydrogen Generation with External Power Supply in Microbial Electrolysis Cell (MEC); (C) Chemical Production by Microbial Electrosynthesis (MES); (D) Middle Chamber Desalination by Electric Drive in Microbial Desalination Cell (MDC). There are also many different sub-systems within ea ch main category. Take MFCs as an example, based on different substrates u sed in MFC reactors, there are wastewater MFCs, sediment or benthic MFCs, etc.109, 110 By utilizing different photosynthetic organisms for solar energy capturing people have developed plant-MFC, phototrophic-MFC, and algae-MFC.57, 58, 111 By integrating other technologies with the BES platform, new systems with superior performance can be developed. For instance, Anode Cathode Resistor Membrane (optional) e-eOrganic CO2H2O O2 PS Anode Cathode Resistor Membrane (optional) e-eOrganic CO2H2H+ PS Anode Cathode Membrane e-eH2O O2Organics CO2 Anode Cathode Resistor AEM e-eOrganic CO2H2O O2 Na+ClCEM Anode Bacteria O2 H2 Organics CO2 Cathode Bacteria A B C D

PAGE 28

12 by incorporating reverse-electrodialysis (RED) with MEC, the microbial reverseelectrodialysis electrolysis cell (MREC) can produc e H2 without any external power supply.71 By integrating capacitive deionization (CDI) with MDC, the microbial capacitive desalination cell (MCDC) could improve d esalination efficiency by 7-25 times.86 Other names may come from the combination of multi ple functions in one system, and they are generally straightforward, suc h as microbial electrolysis desalination cell (MEDC),19 microbial electrolysis desalination and chemical-p roduction cell (MEDCC),85 osmotic microbial fuel cell (OsMFC),63 and microbial electrolysis struviteprecipitation cell (MESC),72 etc. 1.5 MFC-based Systems for Electricity Generation 1.5.1 Wastewater microbial fuel cell (wastewater MF C) MFC refers to the reactor systems that focus on ele ctricity production from biodegradable materials. Table 1.1 provides a compl ete list of different MFCs to date. Early lab scale MFC studies mostly used acetate, gl ucose, or other simple substrates to characterize the performance of materials, reactor configurations, or microbial activities.112, 113 The first MFC study that used real wastewater as s ubstrate was reported in 2004,109 and since then hundreds of studies have been publi shed to report power production from different substrates, including bot h organic and inorganic waste streams using various electrode or separator materials and reactor configurations. Several review articles have provided comprehensive information on the substrates,27 electrode materials,28 separator materials,114 and reactor configurations108 used in different MFC studies.

PAGE 29

Classic MFC designs include the single developed by Liu et al, which in its cubic designs for the first time eliminated the membrane and therefore significantly reduced system internal resistance and cost ( 1.3A).5, 112 Tubular d esigns (Tubular MFC) with different flow patterns s implified construction process and reduced system resistance. designed to increas e organic loading rate, direct voltage or current output while also enhance substrate systems used in wastewater applications may convert the information of substrate concentrat ion, toxicity, or dissolved oxygen concentration into electronic signals as MFC sensor s. Figure 1.3 MFCbased Microbial Fuel Cell,5 (B) B Cell,126 and (D) M icrobial C Classic MFC designs include the single -chamber aircathode MFCs (SCMFCs) developed by Liu et al, which in its cubic designs for the first time eliminated the membrane and therefore significantly reduced system internal resistance and cost ( esigns (Tubular MFC) with different flow patterns s implified construction process and optimized systems with increased electrode surface area and reduced system resistance. 33, 34 Baffled aircathode microbial fuel cell (BAFMFC) e organic loading rate, 36 and stacked MFCs were able to incr direct voltage or current output while also enhance substrate oxidation.37 systems used in wastewater applications include submersible MFCs (SBMFC) may convert the information of substrate concentrat ion, toxicity, or dissolved oxygen concentration into electronic signals as MFC sensor s. based Systems for Electricity Generation: (A) W astewater (B) B enthic Microbial Fuel Cell,119 (C) M icrobial icrobial Solar Cell.111 A D B 13 cathode MFCs (SCMFCs) developed by Liu et al, which in its cubic designs for the first time eliminated the membrane and therefore significantly reduced system internal resistance and cost ( Figure esigns (Tubular MFC) with different flow patterns s implified optimized systems with increased electrode surface area and cathode microbial fuel cell (BAFMFC) was and stacked MFCs were able to incr ease Other MFC include submersible MFCs (SBMFC) ,38 which may convert the information of substrate concentrat ion, toxicity, or dissolved oxygen astewater icrobial Remediation

PAGE 30

14 The main advantages of using MFCs in wastewater tre atment come from the savings of aeration energy and sludge disposal.62, 115, 116 For traditional activated sludge system, aeration can amount to 45-75% of plant ener gy costs, so the conversion of aeration tank to MFC units is very beneficial becau se it not only eliminates aeration energy consumption, studies also showed that the MF C can produce 10-20% more energy that can be used for other processes.27, 117 The reported maximum power density from lab scale air-cathode MFCs has reached to 2.87 kW/m3, making it possible for commercialization development.118 Another main benefit of MFC system is the low biomass production. Because MFC is a biofilm based system, the cell yield of electrochemically active bacteria (0.07-0.16 gVSS/g COD) was much less than the activated sludge (0.35-0.45gVSS/gCOD), so it can re duce sludge production by 5070%,117, 118 which in turn may reduce 20-30% of the plant opera tion cost. Other benefits may include nutrient removal and the production of value-added products, such as caustic solutions for disinfection, or H2 or biogas for energy, which will be discussed more extensively in the following sections. 1.5.2 Benthic microbial fuel cell (benthic MFC) Benthic MFC (BMFC), also known as sediment MFC (SMF C) is a system that utilizes the naturally occurred potential differenc e between the anoxic sediment and oxic seawater to produce electricity.30 Microorganisms oxidize the substrates in the sedim ent and transfer electrons to the anode either embedded in or rested on top of the sediment, and then the electrons are transferred to the catho de suspended in the overlying seawater, where dissolved oxygen is reduced to water (Figure 1.3B).119 The abundant availability of substrates in the sediment makes BMFC a very promis ing power source for autonomous

PAGE 31

15 marine sensors and underwater vehicles, because the y can provide consistent and maintenance-free power supply for a long period of time without using batteries. This is a huge advantage compared to batteries, because batte ries are limited in service life for about 2-4 years, and the replacement can be very ex pensive, especially in deep water. It was estimated that the initial organic matters in 1 L marine sediment could generate an average current of 0.3 mA continuously for 22 years .50 While the concept of BMFC was only introduced in 2001 by Reimers et al.,110 it is a type of BES device that is closest toward commercialization. The first demonstration o f BMFC as a viable power source was reported by Tender et al. in 2008, where an 18 mW meteorological buoy was powered for nearly 7 months.40 Another study showed a chambered BMFC was used to power an acoustic modem interfaced with an oceanogr aphic sensor for over 50 days with an average power density of 44 mW/m2.41 Different configurations of BMFCs have been developed and deployed. Initial designs include sim ple graphite plates buried in the sediment with suspended cathode in water, but such designs are fragile and the power output is very low.120 Nielsen et al. developed a chamber-based BMFC that incorporates a suspended and semi-enclosed anode, which reduced system footprint and increased power output to a range of 380 mW/m2 (3.8 W/m3).39 A Self-stacked submersible microbial fuel cell (SSMFC) showed an open circuit voltage (OCV) of 1.12 V and a maximum power density of 294 mW/m2.42 1.5.3 Microbial remediation cell (MRC) Another emerging application of the BES/MXC platfor m is using the electrodes to serve as inexhaustible electron acceptors (anode ) or donors (cathode) for underground contaminant remediation.121-123 Like sediment MFCs, the MRCs used in groundwater o r

PAGE 32

16 soil remediation can be a single or an array of ele ctrodes without using enclosed containers. Such bioelectrochemically enhanced appr oach can stimulate microbes to concurrently degrade underground pollutants and pro duce additional electricity. Such process is considered sustainable because it elimin ates the injection of expensive chemicals and reduces operational energy cost as co mpared to other technologies. Microbial electrochemical remediation of petroleum contaminants was demonstrated by using electrode as a channel linkin g underground hydrocarbon oxidation and upground O2 reduction. One study showed that the active MRC in creased the degradation of diesel range organics (DRO) by 164% as compared to open circuit potential,43 and another study using U-tube MFC showed crude oi l degradation can be increased by 120% at the location near the electrod e.124 Similar remediation studies on other reductive pollutants including diesel, ethano l, 1,2-dichloroethane, pyridine, and other contaminants were also reported.45-47 Conversely, oxidative contaminants, such as chlorinated solvents, perchlorate, chromium, and ur anium, can be reduced using the electrode as the electron donor.11-13, 125 For instance, studies showed that a negatively polarized electrode could act as an electron donor for the reductive dechlorination of trichloroethene (TCE) to ethene by a mixed culture of microorganisms.13 Similar approach was also used in both lab and field tests for U(VI) reduction, where the horizontally distributed anodes and cathodes enable d direct correlation between acetate injection and uranium reduction, and current produc tion may be an effective proxy for monitoring in situ microbial activity and remediation performance (Fi gure 1.3C).126

PAGE 33

17 1.5.4 Microbial solar cell (MSC) Microbial solar cell is a collective name for diffe rent BES/MXC systems that integrate the photosynthetic reaction with microbia l electricity (or chemical) production using synergistic relationships between photosynthe tic organisms and EAB.111 While the EABs are generally the same bacterial groups in oth er MXC systems, the organisms that are responsible to convert solar energy to organic matters may include higher plants, photoautotrophic bacteria, and algae. A very wide v ariety of names and systems related to MSCs have appeared in literature, such as photomicrobial fuel cell (p-MFC),49 microbial photoelectrochemical solar cell,50 solar-powered microbial fuel cell,52 photobioelectrochemical fuel cell,53 photosynthetic microbial fuel cells (PMFC),54 photosynthetic electrochemical cell,55 and solar-driven microbial photoelectrochemical cell (solar MPC).56 Despite the variations in system designs, the basi c principle of MSCs usually include 4 steps, as described by Strik et a l., and illustrated in Figure 1.3D, (i) photosynthesis of organic matter; (ii) transport of organic matter to the anode compartment; (iii) anodic oxidation of organic matt er by EAB; and (iv) cathodic reduction of oxygen or other electron acceptors.111 Here we categorize and discuss the MSCs into 3 groups based on the organisms responsib le for photosynthesis – plant MSC, phototrophic MSC, and algae MSC. More detailed info rmation can be found in other reviews.57, 58, 111 The most popular MSCs are plant MSCs, which use the organic rhizodeposits excreted from living higher plants to feed EAB for electricity production. Reed mannagrass and rice plants were used first to demon strate the syntrophic relations, with a maximal power output of 67 mW/m2 and 26 mW/m2, respectively.127, 128 Other plants

PAGE 34

18 such Spartina anglica Arundinella anomala and Arundo donax were also investigated for concurrent electricity and biomass production. The A. donax failed,129 but S. anglica was able to generate current for up to 119 days.130 Despite the low power output at current stage, an European research consortium estimated th at the power production from plantMSCs could reach 1000 GJ/ha/year (3.2 W/m2).111 Unlike plant MSCs, the phototrophic MSCs do not require the cooperation between the two groups of microbes, because studies showed that strains of photosynthetic bacte ria such as Rhodobacter sphaeroides can generate electricity through the metabolic acti vity of in situ oxidation of photobiological hydrogen,53 and the power density can be comparable with nonphotosynthetic MFCs.131 A self-assembling self-repairing marine sediment s ystem with photosynthetic microbes was reported to genera te electricity from sunlight without providing constant flux of glucose and oxygen.50 The algae MSC is an emerging system, because the functions of algae and EAB are compleme ntary. The consortium not only can convert solar energy to electric energy, it can als o remove nutrients and produce valueadded chemicals, such as protein and biodiesel. Bot h microalgae ( Chlorella vulgaris ) and macroalgae ( Ulva lactuca ) have been used in algae MSCs to provide substrate s for EAB.102 In addition to traditional batch reactors, Strik e t al. developed a flow through photosynthetic algal microbial fuel cell (PAMFC) to automatically feed algae to MFCs.59 Another study integrated photobioreactor, anaerobic digester, and MFC reactors together to recover both biogas and electricity.132 Other systems include recycling anode off gas (CO2) into an algae grown cathode for additional carbon capture,79 and an integrated photobioelectrochemical system with an MFC enclosed inside an algal bioreactor.62 Utilizing the algae, cyanobacteria and protozoa, St rik et al. reported an MSC with a

PAGE 35

19 reversible bioelectrode that the electrode can func tion as a biocathode during illumination for photosynthesis reaction and also can switch as the anode in the dark for organic degradation.52 MSCs are the only BESs that do not rely on externa l electron donors but convert inexhaustible solar energy into electrical energy and chemicals, so they carry great potentials if current challenges such as low power output are addressed. 1.6 MEC-based Systems for Chemical Production The concept of microbial electrolysis cell was orig inated in 2005, with the key feature of using an external voltage on top of the MFC potential to enable hydrogen gas evolution at the cathode through the reduction of p rotons.66, 133 Early studies used external power supplies ranged from 0.6-1.0 V to ca talyze H2 evolution, which was much lower than the 1.8-2.0 V used in traditional water electrolysis.9, 66 Another advantage was that the substrates can be from renewable and waste materials rather than fossil fuels, and the H2 production rate can be more than 1 m3/d/m3 reactor with a yield upto 11 mol H2/mol glucose, more than 3 times higher than dark fe rmentation.9, 10 Several excellent reviews summarized the material and system developm ent of the MECs for H2 production.9, 10, 134 The elimination of membranes or separators converte d dual chamber MECs to single chamber reactors and significantly increased H2 generation rate, but the produced H2 was more likely consumed by methanogenesis to gene rate CH4.9, 10 Researchers have tried different inhibition approaches such as addin g expensive methanogen inhibitors, periodically expose solution in aerobic environment and control the pH and redox potentials, but the CH4 contamination of H2 in single chamber MECs still remains a major obstacle.9, 10, 135 The small external voltage can be supplied by MFC stacks or other

PAGE 36

20 renewable power sources such as solar and wind.70, 136 Recently, reverse electrodialysis (RED) was added into MEC generating a new system ca lled microbial reverseelectrodialysis electrolysis cells (MREC) with spon taneous H2 production by combing together the driving forces from anode organic oxid ation and salinity gradient energy (Figure 1.4A), and salt solutions could be continuo usly regenerated with waste heat (40 C).65, 71 Figure 1.4 Some Advanced MXC Systems: (A) Microbia l Reverse-electrodialysis Electrolysis Cell (MREC) for H2 Production,71 (B) Microbial Electrosynthesis (MES) for Organic Synthesis,76 and (C) Microbial Capacitive Desalination Cell (MC DC) for Desalination.86 A B C

PAGE 37

21 By using similar strategies in MECs, other inorgani c chemicals have been produced in the cathode chamber. Cusick and Logan d iscovered that phosphate can be recovered as struvite (MgNH4PO4 6H2O) in a modified microbial electrolysis struviteprecipitation cell (MESC).72 Rozendal et al. reported that hydrogen peroxide ca n be produced by reducing oxygen through the two electro n reduction, and the proof-ofconcept study showed at an applied voltage of 0.5 V H2O2 can be generated at a rate of 1.9 0.2 kg H2O2/m3/day1 at a concentration of 0.13 0.01 wt% with an over all efficiency of 83.1 4.8%.67 The same group later used a similar approach to pr oduce alkaline solutions, as they found that by using ace tate as the electron donor in the anode, the MEC generated up to 1.05A in current at 1.77 V applied voltage, which allowed for the production of caustic to 3.4 wt%.68 Such chemicals can be produced during wastewater treatment process and then used as low-c ost disinfectants for many industries. 1.7 MES-based Systems for Chemical Production Microbial electrosynthesis is an emerging area in B ES research and development, and it uses the electrons derived from the cathode to reduce carbon dioxide and other chemicals into a variety of organic compounds, espe cially those with multiple carbons that are precursors for desirable value-added chemi cals or liquid transportation fuels.14, 15, 74 The potential of MES not only comes from the doubl e benefits of carbon sequestration and organic production, it may also address the har vesting, storage, and distribution problems associated with energy crops, solar and wi nd farms, and natural gas exploration, because the electrons can be from any renewable sou rce, and microbes may harvest solar energy in a 100-fold higher efficiency than biomass -based chemical production.

PAGE 38

22 The concept of microbial electrosynthesis was only introduced in 2009-2010, with the initial findings associated with methane genera tion from a reactor with an abiotic anode and a biocathode acclimated with Methanobacterium palustre .8 Another early study demonstrated that biofilms of Sporomusa ovata could use the electrons supplied by the cathode to reduce carbon dioxide into acetate a nd small amounts of 2-oxobutyrate. Electrons appearing in these products accounted for over 85% of the electrons consumed (Figure 1.4B).76 In general, acetogenic bacteria use hydrogen as th e electron donor for carbon dioxide reduction, but it was found that man y acetogenic bacteria, such as Clostridium ljungdahlii, Clostridium aceticum, Spor omusa sphaeroides and Moorella thermoacetica were all able to consume electrical current and pr oduce organic acids.77 Studies also show that ethanol can be produced by r educing acetate at the cathode, but some processes required the addition of mediators, such as methyl viologen (MV).75 Mixed culture originated from brewery wastewater we re reported to generate methane, acetate, and hydrogen gas from a biocathode poised at -590 mV (vs SHE) with CO2 as the only carbon source,137 and research on genetically modified microorganism s may significantly facilitate electron uptake and organi c synthesis. As discussed in several conceptual review articles, the microbial electrosy nthesis carries great potentials, but there are also many technological and economic chal lenges to be solved before it can be implemented in large scale.14, 15, 74 1.8 MDC-based Systems for Water Desalination and Be neficial Reuse Water desalination using the MDC process was first introduced in 2009 by Cao, et al, and the proof-of-concept study was selected as the top technology paper by Environmental Science & Technology .16 The basic principle of MDC is to utilize the

PAGE 39

23 electric potential generated across the anode and c athode to drive desalination in situ Compare to other MXCs, MDC has a third chamber for desalination by inserting an anion exchange membrane (AEM) and a cation exchange membr ane (CEM) in between the anode and cathode chambers. When bacteria in the an ode chamber oxidize biodegradable substrates and produce current and protons, the ani ons (e.g., Cl-) in the middle chamber migrate to the anode and the cations (e.g., Na+) are drawn to the cathode for charge balance, thus the middle chamber solution is desali nated.16, 20 Recently, other approaches were developed to achieve desalination as well. For example, by switching the CEM to the anode side and AEM to the cathode side, microbi al saline-wastewater electrolysis cell (MSC) desalinates anolyte and catholyte by driving salts into the middle chamber.80 Osmotic microbial fuel cell (OsMFC) or osmotic MDC (OsMDC, MODC) uses a forward osmosis membrane to replace the AEM and wit hdraw pure water from wastewater to the draw solution, and then water can be recovered during draw solution regeneration.63, 82 Capacitive microbial desalination cell (cMDC) inco rporates capacitive deionization into MDC to improve desalination effic iency.83, 86, 87 In addition to desalination, acid (HCl) and base (NaOH) solutions can be produced if a bipolar membrane is placed into the MDC next to the anode c hamber, creating a four-chamber system called microbial electrolysis desalination a nd chemical-production cell (MEDCC).85 The MDC can be used either stand-alone for simultan eous organic and salt removal with energy production or serve as a pretre atment for conventional desalination processes such as reverse osmosis (RO) to reduce th e feed solution salt concentration, and minimize energy consumption and membrane foulin g. Compared with current

PAGE 40

24 technologies that use 6-68 kWh to desalinate 1 m3 of seawater, MDC studies showed that 180-231% more energy can be recovered as H2 than the reactor energy input when desalinating 5-20 g/L1 NaCl solutions,17, 19 and it was estimated that MDC may produce upto 58% of the electrical energy required by downs tream RO systems.18 Higher desalination efficiency and current output can be a chieved through membrane stacks,88, 138 and electrolyte recirculation was shown effective in stabilizing electrolyte pH.89, 139 Traditional MDC designs accomplish desalination by transporting ions from the middle chamber to the anode and cathode chambers, which in creases the conductivity of the anolyte and catholyte. This change has been shown b eneficial to electricity generation due to improved mass transfer, but the increased sa linity may also affect effluent water quality and prevent subsequent beneficial use of tr eated wastewater.20 One solution for complete salt removal from all solutions may involv e the physical and electrical adsorption of ions onto high surface area membrane electrode assemblies, such as microbial capacitive desalination cell (MCDC), whic h showed upto 25 times of increase in salt removal and complete salt recovery (Figure 1.4C).86 Similar as many membrane based technologies, one challenge for MDC may come from membrane fouling due to biofilm growth and scaling due to the deposition of hardness-causing cations, but studies on understanding and addressing such problems are j ust getting started, and solutions remain to be found.139, 140 1.9 Outlook In one decade of research and development, the func tionality of BES/MXC has expanded dramatically and the performance has impro ved exponentially. Taking MFC as an example, the power density has increased by orde rs of magnitude, from less than 1

PAGE 41

25 mW/m3 to 2.87 kW/m3 (or 10.9 kA/m3),118 and the projected wastewater treatment capacity of MFC can reach to 7.1 kg chemical oxygen demand (COD)/m3 reactor volume/day, which is even higher than conventional activated sludge systems (~0.5–2 kg COD/m3 reactor volume/day).3 However, there are still remaining challenges that need to be addressed before the technology can be applied i n commercial scale. Despite the elimination of expensive metal catalysts and membra nes, the overall cost of MXCs is still considered expensive for wastewater treatment, unle ss an estimated threshold of internal resistance <40 m m2 in combination of a current density around 25 A/m2 can be reached.25 Most studies are still limited in lab scale, and s everal pilot scale plants with capacities between 20-1000 liters have yet shown st able and high enough performance due to the problems of water leaking, low power out put, influent fluctuation, and unfavorable products.141-143 To achieve practical implementation, BESs will nee d to be scaled-up to at least cubic meter scale, the reacto r configurations have to be easily integrated with current infrastructure, and effecti vely harvesting systems instead of resistors have to be developed to deliver usable po wer.144, 145 Multiple reviews have summarized the progresses of MFC system development and provided insights in further directions.3, 28, 92, 142 Compared to electricity generation in MFCs, chemica l production and desalination from BESs have been considered technic ally and economically more feasible due to the higher price of chemical and relatively simple collection process. But such processes are relatively new and mainly in lab scal e, and there has been few reports in scale-ups.141, 143 Among the many different functions developed using this BES/MXC platform technology, as discussed across this artic le, it is not clear where the BES can

PAGE 42

26 contribute the most to the current environmental in frastructure and chemical industry. There have been very limited evaluations of BESs/MX Cs regarding to their life cycle in terms of function selections or comparisons with es tablished technologies which they can complement.146, 147 It has been assumed that the most environmental be nefits from BESs come from the displacement of fossil fuel dependent resources (i.e. grid electricity, or chemical manufacture) through co-product production (i.e. electricity, chemicals) from renewable sources, but the energy and environmental footprints of BESs have to be clearly quantified before implementing large scale applications. Despite the remaining challenges, BES/MXC has been widely considered the next generation of platform technologies that will provide the multidisciplinar y “X factor” for energy and environmental sustainability. 1.10 Acknowledgement This work was supported by the US National Science Foundation under Award CBET-1235848.

PAGE 43

27 2. Recycled Tire Crumb Rubber Anodes for Sustainabl e Power Production in Microbial Fuel Cells2 2.1 Abstract One of the greatest challenges facing microbial fue l cells (MFCs) in large scale applications is the high cost of electrode material We demonstrate here that recycled tire crumb rubber coated with graphite paint can be used instead of fine carbon materials as the MFC anode. The tire particles showed satisfacto ry conductivity after 2 to 4 layers of coating. The specific surface area of the coated ru bber was over an order of magnitude greater than similar sized graphite granules. Power production in single chamber tireanode air-cathode MFCs reached a maximum power dens ity of 421 mW/m2, with a coulombic efficiency (CE) of 25.1%. The control gra phite granule MFC achieved higher power density (528 mW/m2) but lower CE (15.6%). The light weight of tire pa rticle could reduce clogging and maintenance cost but posts chal lenges in conductive connection. The use of recycled material as the MFC anodes brings a new perspective to MFC design and application and carries significant economic and en vironmental benefit potentials. Keywords: tire crumb rubber, microbial fuel cell, a node, electricity 2 The work presented in this chapter has been publis hed by Heming Wang, Matthew Davidson, Yi Zuo, and Zhiyong Ren in J. Power Sources 2011, 196, 5863-5866.

PAGE 44

28 2.2 Introduction Microbial fuel cells (MFCs) are an emerging bioelec trochemical technology that produces electrical energy from organic matter cata lyzed by exoelectrogenic bacteria on the anode.148-150 In less than a decade, researchers have increased power densities by several orders of magnitude, from mW/m3 to kW/m3.143, 150 These increases have come primarily from addressing the physical and chemical constraints on MFC performance by exploring new materials and optimizing reactor arch itectures. A remaining challenge for MFCs as they become more technically feasible for f ull-scale applications such as simultaneous wastewater treatment and bioenergy rec overy is the high cost of electrode material currently used in lab scale studies. Many anode materials have been tested to improve biofilm attachment and conductivity. The po pular materials include graphite granules,33 carbon paper,44 carbon cloth,151 carbon mesh,148 and activated carbon.35 The recent development of graphite brush anodes with hi gh specific surface area and an open structure to prevent fouling problems provides a so lution for scaling up.152 However, the cost of most electrode materials, from ~$50/m2 to over $1,000/m2, is prohibitive to use in large scale.143, 148 We investigated the performance of a recycled mater ial crumb rubber (granular particles produced by grinding waste tires) as a potential inexpensive and abundant alternative material for MFCs. The Rubber Manufactu res Association (RMA) estimated that 303.2 million scrapped tires were produced in the U.S. in 2007; approximately one discarded tire per person per year.153 In addition, more than 300 million tires are curre ntly stockpiled throughout the country due to the lack o f end-use markets. These stockpiles pose great environmental, safety, and health concer ns. The materials are fire hazards,

PAGE 45

29 non-biodegradable, and occupy significant landfill space. Current disposal solutions include incineration for tire derived fuels, reuse of crumb rubber as surfaces for playgrounds and sports fields, and reuse of tire ru bber in asphaltic concrete mixtures.154 Recently Tang et al. developed a new crumb rubber f iltration system to treat wastewater and ship ballast water.155, 156 They found that crumb rubber filters significantly reduced clogging compared to sand filters without compromis ing pollutant removal. Tire derived rubber particles also showed better organic adsorpt ion capacity than sand particles, as well as exhibited good performance as high surface area, non-toxic media for biofilm attachment in bioreactors.157 In this study, we tested for the first time the fea sibility of using crumb rubber with a conductive graphite coating as the anode material for electricity production in MFCs, and compared its performance with graphite granule anodes. Statistical and electrochemical analyses were conducted to evaluate the performance of the coated material in terms of conductivity, resistances, and specific surface area. The potential benefits of using crumb rubber as MFC electrode was also discussed. 2.3 Materials and Methods 2.3.1 MFC construction and operation Single-chamber MFCs were constructed from Wheaton g raduated media bottles (250 mL, Wheaton, NJ) by adding one glass extension tube on the side.152, 158 Rubber top caps were used to provide an air-tight condition. A ir cathodes (projected area of 4.5 cm2, one side) were made by applying Pt/C (0.5 mg/cm2) and four PTFE diffusion layers on 30% wet-proofed carbon cloth (Fuel Cell Earth, MA, USA) as previously described.159 Recycled tire crumb rubber was donated by AcuGreen Inc. (CO, USA). The crumb

PAGE 46

30 rubber was pre-shredded from recycled tires and sie ved to collect particles with 4-8 mm diameter. The rubber particles were washed with dei onized water and air dried before the application of the conductive coating (E-34, Superi or Graphite Co. OH, USA) to the particle surface.160 To determine the optimal coating condition, additi onal coatings were applied to some of the particles after the previous coating was completely dried in air. Coated crumb rubber was packed into the MFC anode c hamber to a volume of 140 mL (71.5 g). A twisted titanium wire was inserted into the anode pack as a current collector and connected to the external circuit. The same vol ume (140 mL, 133.6 g) of graphite granules (D=2-6 mm, Graphite Sales Inc. OH) were us ed as the control anode material in a separate MFC.MFCs were inoculated with anaerobic sludge obtained from the EnglewoodLittleton Wastewater Treatment Plant (Englewood, CO ). The reactors were fed with 190 mL medium containing: 1.25 g/L of sodium acetate, 0 .31g/L of NH4Cl, 0.13 g/L of KCl, 3.321 g/L of NaH2PO42H2O, 10.317 g/L of Na2HPO412H2O, 12.5 mL/L of mineral solution and 5 mL/L of vitamin solution.158 All MFCs were operated in fed-batch mode at room temperature. Growth media was replaced with fr esh media when the voltage dropped below 50 mV (1000 resistance) 2.3.2 Statistical and electrochemical analyses The optimal number of coatings on the crumb rubber was determined by resistance measurement and statistical analysis. Af ter each coat was finished and completely dried in air, 35 coated rubber samples w ere randomly selected and the ohmic resistance across a 4 mm distance was measured repe atedly by a programmable multimeter. The t distribution was used to calculat e confidence intervals (CIs) for the

PAGE 47

31 mean changes between the resistances on two adjacen t coatings. The 95% CIs were calculated by 2.032 XSn , where 1 n i i XXn == was sample mean, 2 1() 1 n i iXX nS== was sample standard variation, and n was the sample size. MFC cell voltage was continuously monitored using a data acquisition system (Keithley Instruments, OH). The circuits were opera ted under a fixed load of 1000 During the stable power production stage of each ba tch experiment polarization measurements were made using a variable resistor bo x (50 to 50 k). Current (I = V/R), power (P = IV), and coulombic efficiency (CE, based on COD) were calculated as previously described.6 Electrochemical impedance spectroscopy (EIS) tests were conducted using a Potentiostat (PC 4/300, Gamry Ins truments, NJ, USA) to measure the internal resistances with the anode as the working electrode, and the cathode as the counter electrode and reference electrode. The scan range was from 105 Hz to 0.005 Hz with a small sinusoidal perturbation of 10 mV. Specific surface area and pore size distribution of the particles were estimated by Brunauer-Emmett-Teller (BET) method using a five-po int N2 gas adsorption technique (ASAP 2020; Micromeritics, Norcross, GA).161 The average pore size and pore size distribution were determined from desorption of N2 according to the method developed by Barrett, Joyner, and Halenda (BJH).162

PAGE 48

32 2.4 Results and Discussion 2.4.1 Resistance characterization of coated crumb r ubber electrode Maximum ( /mm) 131 6 6.2 4.8 4.9 Minimum ( /mm) 6.5 2.0 1.7 1.4 1.3 Median ( /mm) 16 3.5 2.8 2.1 2.2 95%CI ( /mm) 22.08.1 3.50.3 3.00.4 2.50.3 2.30.2 n=35, t0.025,34=2.032 Figure 2.1 Box Plot of Resistance Measurement and Statistics on Tire Crumb Particle Surface with Different Coating Layers. To convert the almost non-conductive crumb rubber i nto electrically conductive electrode, multiple layers of graphite paint were a pplied to the rubber particle surface. Such approach has been successfully applied on ultr afiltration membrane MFC cathodes, as described previously.160 Figure 2.1 is the box plot showing the statistics of ohmic resistance variations of the 35 randomly selected p articles coated with different layers of graphite paint. It appears that additional coatings reduced the particle surface ohmic resistance and heterogeneity. After the first coati ng, a particle has an average ohmic resistance of 22.0 /mm, but the numbers across the 35 samples varied s ignificantly, from 6.5 to 131 /mm, resulting in a huge standard deviation. The se cond layer of coating reduced the ohmic resistance by a factor of 6, to an average of 3.5 /mm.The 0 20 40 60 80 100 120 140 1-layer2-layer3-layer4-layer5-layerOhmic resistance ( /mm) 0 2 4 6 82-layer3-layer4-layer5-layer

PAGE 49

33 variation was also considerably reduced. Additional coatings showed only minor improvements: the difference between 4 coatings and 5 coatings was not statistically significant at the 5% level (Figure 2.1). In order to conserve coating material and reduce cost, the rubber particles with 4 layers of coating was used in further MFC and electrochemical characterizations. The average ohmi c resistance of the particles was 2.5 /mm, ~10 times greater than graphite granules. EIS s hows a similar trend in reactor ohmic resistance in coated crumb rubber or graphite granules MFC reactors (Figure 2.2). The system resistance decreased along with addition al coatings. The average system resistance of the MFC with 4-layer coated rubber wa s 574 and the resistance using same volume of graphite granule was 210 Figure 2.2 System Resistance of Single Chamber Bot tle Reactor Filled with Graphite Granules and Tire Particles with Different Coating Layers. 2.4.2 Surface characterization of coated crumb rubb er electrode High specific surface area is a crucial parameter o f the MFC anode. It allows higher biofilm density and thus making higher curre nt output possible. Surface characterization shows that the crumb rubber partic le with 4-layer coating has an average -1200-1000 -800-600-400-200 0 020040060080010001200 Z img (ohm) Zreal(ohm) 1-layer 2-layer 3-layer 4-layer Granule

PAGE 50

34 BET surface area of 4.5 m2/g (32,143 m2/m3), more than one order of magnitude greater than the graphite granule (0.3 m2/g, or 2,143 m2/m3). The BJH desorption cumulative pore volume of the coated rubber particle was 0.013 cm3/g and BJH desorption average pore diameter was 88 , also significantly higher t han the graphite granule in terms of the same parameters. The granule has a BJH desorption c umulative pore volume of 0.0006 cm3/g and BJH desorption average pore diameter of 54 Figure 2.3 compares the pore size distribution of the 4-layer coated tire partic le and graphite granule. The incremental pore area and cumulative pore area of the tire part icle are each one order of magnitude greater than those of the graphite granule. Specifi cally, the desorption cumulative pore area of tire particle and graphite granule were 5.7 8 cm2/g and 0.45 cm2/g, respectively. Figure 2.3 Pore Size Distribution of (A) Rubber Pa rticle with 4-layer Coating, and (B) Graphite Granule as the MFC Anode. 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.00 0.10 0.20 0.30 0.40 0.50 050100150200Incremental pore area (m2/g) Cumulative pore area (m2/g)Pore size () Cumulative Incremental Graphite granule 0.00 0.20 0.40 0.60 0.80 1.00 1.20 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 050100150200Incremental pore area (m2/g) Cumulative pore area (m2/g)Pore size () Cumulative Incremental 4-layer tire particleB A

PAGE 51

35 2.4.3 Power production from tire rubber MFCs and gr aphite granule MFCs Repeatable cycles of power production were obtained from both tire rubber and graphite granule MFCs after 3-4 feeding cycles with fresh media. The stable voltages over a 1000 external resistor in tire and graphite MFCs were a round 390 mV and 430 mV, respectively. The tire reactor generally showed longer batch durations than the graphite MFC. A regular batch cycle for graphite MF Cs took around 20 days before the media change, while a batch cycle for tire MFCs too k about 30 days. Polarization and power density curves obtained by varying the extern al circuit resistances from 50 to 50,000 showed that the crumb rubber MFC produced less pow er than graphite granule MFC. Figure 2.4 shows that the maximum power densit y of the 4-layer coating tire MFC was 421 mW/m2 (cathode projected area), ~20% less than the power density of the graphite MFC (528 mW/m2). The COD removal of the tire reactor (85.0 %) wit hin a batch was less than that from the graphite MFC (92. 8%), but the columbic efficiency obtained from the tire reactor (25.1%) was nearly o ne and half times higher than that calculated from the graphite granule MFC (15.6%). Figure 2.4 Voltage and Power Density as a Function of Current Density for Coated Tire Anode MFCs and Graphite Granule Anode MFCs. 0 150 300 450 600 750 0 150 300 450 600 750 900 01234Voltage (mV) Power Density (mW/m2)Current Density (A/m2) Granule-Voltage Tire-Voltage Granule-Power Tire-Power

PAGE 52

36 The difference in power production from the two typ es of reactors is a result of several factors. It was noted that the specific are a of the coated rubber particle was much higher than the graphite granule, which could resul t in higher attachment and current output, but the high ohmic resistance of the coated tire particle outweighed the benefit of surface area and caused lower power generation. Th e conductive coating only allows the electrons transfer across the surface of the tire p article rather than through the diameter of the granule, which increased the length of the tran sfer route. Additionally, the density of the crumb rubber is about 1.1 g/cm3, only a little greater than water but much less th an the density of the graphite (2.2 g/cm3). The low density of crumb rubber media has the benefits with reduced maintenance cost and clogging potential,155 but it results in a loose packing that hinders conductive connection. The in tegration of metal current collectors into the anode pack could alleviate the problem in larger scale systems by compacting tire particles and generating a highly conductive networ k for more efficient electron transfers.163 2.4.4 Cost-benefit outlook The use of recycled tire crumb rubber instead of ex pensive carbon products as MFC electrode material is believed to carry economi cal and environmental benefits. Compared to the high cost of refined carbon electro de ($50 to over $1,000/m2), the crumb rubber is free except for the minimal cost of the c oating material. Our preliminary cost calculation shows the cost of coated tire electrode is $0.71/m2 and $1.42/m2 for 2-layer coating and 4-layer coating, respectively, which is comparable to graphite granules ($1.29/m2). Moreover, many countries currently have tire dis posal tax ($1 $3 per tire) and reuse subsidy programs ($0.1-$0.5 per tire) to encourage tire recycle and reuse.

PAGE 53

37 These policies make the use of crumb rubber more ec onomically competitive and can reduce the cost of rubber electrode by another $0.2 0-0.40/m2.164 In addition, government regulations and higher public expectation on waste recycle and renewable energy production make the technology more attractive to i ndustries, as the use of recycled tire rubber in MFCs will reduce the cost of tire disposa l and bring more environmental benefits by increasing tire reuse, treating wastewa ter, and generate alternative energy. 2.5 Conclusions Recycled tire crumb rubber was tested for the first time as an alternative electrode material in microbial fuel cells. The tire particle s showed good conductivity after 2-4 layers of graphite coating. The specific surface ar ea of the coated tire particle was more than 10 times greater than similarly sized graphite granules and provides improved attachment surface for microbes. The single chamber air-cathode tire MFC produced the same level of power, COD removal, and coulombic eff iciency as the graphite granule MFCs did. The concept of using recycled material as MFC electrodes opens up a whole new approach toward MFC design and application that carries significant economic and environmental benefits. 2.6 Acknowledgements This work was supported by the Office of Naval Rese arch (ONR) under Awards N000140910944. We thank Dr. Peter Jenkins and Dr. P ei Xu for valuable discussion and Superior Graphite Co. for donating graphite coating materials.

PAGE 54

38 3. Carbon Nanotube Modified Air-cathodes for Electr icity Production in Microbial Fuel Cells3 3.1 Abstract The use of air-cathodes in microbial fuel cells (MF Cs) has been considered sustainable for large scale applications, but the p erformance of most current designs is limited by the low efficiency of the three-phase ox ygen reduction on the cathode surface. In this study we developed carbon nanotube (CNT) mo dified air-cathodes to create a 3-D electrode network for increasing surface area, supp orting more efficient catalytic reaction, and reducing the kinetic resistance. Compared with traditional carbon cloth cathodes, all nanotube modified cathodes showed higher performanc e in electrochemical response and power generation in MFCs. Reactors using carbon nan otube mat cathodes showed the maximum power density of 329 mW/m2; more than twice that of the peak power obtained with carbon cloth cathodes (151 mW/m2). The addition of Pt catalysts significantly increased the current densities of all cathodes, wi th the maximum power density obtained using the Pt/carbon nanotube mat cathode at 1118 mW /m2. The stable maximum power density obtained from other nanotube coated cathode s varied from 174 mW/m2 to 522 mW/m2. Scanning electron micrographs showed the presence of conductive carbon nanotube networks on the CNT modified cathodes that provide more efficient oxygen reduction. Keywords: microbial fuel cell, carbon nanotube, cat hode, electricity 3 The work presented in this chapter has been publis hed by Heming Wang, Zhuangchun Wu, Atousa Plaseied, Peter Jenkins, Lin Simpson, Ch aiwat Engtrakul, and Zhiyong Ren in J. Power Sources 2011, 196, 7465-7469.

PAGE 55

39 3.2 Introduction Microbial fuel cells (MFCs) are renewable energy sy stems that employ bacteria to convert chemical energy stored in biodegradable mat erials to electrical energy. The MFC technology carries great potential as it concurrent ly removes pollutants from the environment and produces energy. An MFC reactor gen erally consists of an anode, a cathode, and sometimes a separator between the two electrodes.149 In order to provide good access for bacteria and improve power generati on, the electrodes in MFCs need to have high surface area, high conductivity, and be r esistant to physical and chemical corrosion. Many anode materials have been tested, i ncluding woven graphite mat,33 carbon paper,158 carbon cloth,151 and activated carbon.35 A recent development of graphite brush anodes provides a solution for scali ng up because it has very high specific surface area (7170 18200 m2/m3) and an open structure to prevent fouling problems .152 Moreover, this makes the anode no longer the main l imitation on power production, but instead brings up the challenge of effective cathod e development.165, 166 Even though ferricyanide or permanganate can provid e a higher cathode open circuit potential, oxygen is considered as the elec tron acceptor for eventual MFC applications due to its availability and high redo x potential (E0=1.23 V).166, 167 However, the tri-phase reaction among oxygen gas, solid cata lyst, and liquid electrolyte does not make satisfactory reaction kinetics. Moreover, the specific surface area (~100 m2/m3) of current popular air-cathode designs are orders of m agnitude lower compared with the brush anode, making the cathode the primary limitin g factor for improved power production in MFCs. Previous results showed that MF C power output improved along with the increase of cathode specific surface area. For example, Deng et al., found the

PAGE 56

40 power density increased from 67 mW/m2 to 315 mW/m2 by replacing carbon paper cathodes with high specific surface area activated carbon felt cathodes.168 Cheng and Logan recently demonstrated that the volumetric pow er density has a linear relationship with the cathode specific surface area.169 Carbon nanotubes (CNTs) have exhibited great potent ial as electrode materials in fuel cell applications due to their high surface-to -volume ratio and unique electrical and mechanical properties. Several studies investigated the performance of CNT-modified anodes in MFCs and found that the anode biofilm act ivity was not affected by the carbon nanotubes but the power density was improved.170-172 Though the CNT-modified anodes do not provide direct large enough macro-scale poro sity for more microbial colonization, the highly conductive nanotube network serves as na nowires to facilitate the electron transfer between the microbes to the electrodes. Th e reported power densities obtained from CNT-modified anodes varied from 22 mW/m2 (multi-walled CNT)173 to 1098 mW/m2 (3-D CNT-texile).171 However, to date little effort has been reported o n CNTmodified cathodes, even though the benefits can be more significant because the CNTmodified cathodes provide higher conductivity for i mproved electron transfer efficiency and allow more nano-sized catalyst particles to be deposited in a 3D texture to facilitate the tri-phase reaction kinetics. Therefore, in this study we used carbon nanotubes to modify the air-cathode using several different meth ods and characterized the MFC performance under each condition.

PAGE 57

41 3.3 Materials and Methods 3.3.1 Cathode construction Different coating methods were used to modify MFC a ir-cathodes using carbon nanotubes. In addition, carbon nanotube mat and car bon cloth modified with Pt nanoparticles were tested for comparison. Table 3.1 shows the list of 8 different electrodes that were characterized and the specific ations of their modification. Table 3.1 List of Cathode Materials and Modificati ons Used in This Study and Their Specifications. Cathode Name Material Type Pt Coating Method CC Carbon cloth — CC-Pt Carbon cloth with Pt 10%Pt/carbon black mixture (brush) CNTM Carbon nanotube mat — CNTM-Pt Carbon nanotube mat with Pt 10%Pt/carbon black mixture (brush) 1 SWNT n SWNT coated electrode — 1 SWNT n -Pt SWNT coated electrode with Pt H2PtCl6 (Microwave) 2 SWNT c -Pt SWNT coated electrode with Pt 2 MWNT c -Pt MWNT coated electrode with Pt 1 Nanotube synthesized in National Renewable Energy L aboratory 2 Nanotube purchased from Cheap Tubes Inc. Carbon cloth cathodes (CC) were made by applying on e layer of carbon black nanoparticles and four PTFE diffusion layers on the air side of the carbon cloth according to Cheng et al.159 In some cases, a catalyst layer containing 0.5mg/c m2 platinum nanoparticles was applied on the water side of the carbon cloth to improve reaction kinetics (CC-Pt). Carbon nanotube mat (CNTM) that c ontains more than 90% carbon nanotubes was donated by Nanocomp Technologies Inc. (OH, USA) and was used directly as the MFC cathode. The same amount of Pt catalyst was applied in some of the CNTM cathodes for comparison (CNTM-Pt).

PAGE 58

42 Single-walled carbon nanotubes (SWNTs) were synthes ized by Laser Vaporization methods174, 175 at the National Renewable Energy Laboratory (NREL) (CO, USA). The as-prepared nanotubes were purified using a method introduced by Dillon, which consisted of treatment in 3 molar nitric acid flux for 16 hours, followed by an acetone wash and air burn at 525C.174 Commercially available SWNTs and multi-walled carbon nanotubes (MWNTs) were purchased from CheapT ubes Inc., (NJ, USA) and purified by the same procedure. The purified carbon nanotubes were suspended in 1% sodium dodecyl sulfate (SDS) surfactant water solut ion to prevent agglomeration. The suspended carbon nanotubes were then deposited onto a PTFE membrane (Advantec MFS, Inc, CA, USA) by vacuum filtration to form a c ompact cathode. Sufficient DI water washing was applied after to remove the extra surfactant. In some cases, Pt nanoparticles (0.5mg/cm2) were uniformly coated on the surface of carbon na notubes by using a controlled temperature microwave method (25 0 W, 140 ) for 90 seconds.176 The cathodes containing nanotubes synthesized at NR EL with and without Pt coatings were labeled as SWNTn-Pt and SWNTn, respectively. Similarly, the cathodes using commercial SWNTs and MWNTs from CheapTubes Inc. wit h Pt coating were labeled as SWNTc-Pt MWNTc-Pt, respectively (Table 3.1). 3.3.2 MFC construction and operation Single chamber cubic-shaped reactors were construct ed as previously described.5, 152 The total volume of each reactor was 28 mL. Graphi te fiber brushes were used as the anodes in all of the experiments. Prior to the test s, the brushes were treated by soaking in acetone overnight and then heated at 450C for 30 m inutes.148 All reactors were initially equipped with CC-Pt cathodes and inoculated using e ffluent from an air-cathode MFC

PAGE 59

43 operated for nearly one year. When exoelectrogenic bacteria were acclimated and all reactors showed repeatable and comparable voltages (51010 mV), nanotube modified air-cathodes were transferred to MFC reactors for p erformance characterization. Medium solution was prepared containing 1.25 g/L CH3COONa, 0.31g/L NH4Cl, 0.13g/L KCl, 3.321g/L NaH2PO42H2O, 10.317g/L Na2HPO412H2O, 12.5 mL/L mineral solution, and 5 mL/L vitamin.19 Reactors were operated in fed-batch mode at room t emperature and refilled with new medium solution when voltages red uced below 30 mV forming one cycle of operation. 3.3.3 Electrochemical and Microscopy Analysis Voltage across an external resistor (1000 ) was recorded at 10-minute intervals using a data acquisition system (Keithley Instrumen t, OH, USA) connected to a computer. Polarization power density curves were obtained by altering external resistances from 50000 to 50 during the stable power production stage of each b atch. The calculations of power density was performed according to Ren et al.177 Linear Sweep Voltammetry (LSV) was applied using a potentiostat (PC4/300, Gamry Instruments, NJ, USA) to characterize the abi otic electrochemical oxygen reduction behavior on different cathodes. LSV tests were conducted in the same reactor filled with same media solution but without carbon source and bacterial inoculums.178 The working electrode was the cathode (4.9 cm2 projection area), the counter electrode was a titanium wire (diameter: 0.081 cm, length: 40 cm), and the reference electrode was an Ag/AgCl electrode (RE-5B, BASi, IN, USA). The po tential was scanned from open circuit potential to -0.2 V at a rate of 1.0 mV/s. Reactor internal resistances were measured by Electrochemical Impedance Spectroscopy (EIS) with the anode as the

PAGE 60

44 working electrode, and the cathode as the counter e lectrode and reference electrode. The scan range was from 105 Hz to 0.005 Hz with a small sinusoidal perturbatio n of 10 mV.158 All tests were repeated at least three times to ve rify consistency. Selected cathode samples were examined using a dual beam focused ion beam scanning electron microscope (FIB/SEM, NOVA 600i, F EI Company). Samples were fixed overnight at 4C by KarnovskyÂ’s Fixative (Ele ctron Microscopy Sciences, CA, USA), washed three times in phosphate buffer (0.2 M pH 7.2), and then dehydrated stepwise in a series of water/ethanol solutions wit h increasing ethanol concentration (50, 70, 80, 90, 100 %). Samples were then kept in a des iccator prior to Pd/Pt sputtering and SEM observation. 3.4 Results and Discussion 3.4.1 Electrochemical performance The electrochemical performance of the cathodes wit h respect to current density was evaluated using LSV tests in the absence of bac teria (Figure 3.1). Among the cathodes without Pt catalyst coating, those electro des consisting of or modified with carbon nanotubes (e.g. CNTM and SWNTn) showed much larger current responses than the carbon cloth cathode (CC). SWNTn cathodes showed the most positive onset potential, followed by the CNTM cathode, while the CC showed m inimal current response across the potential scan range. The larger current respon se from the nanotube cathodes indicates a higher limiting current density and bet ter electrochemical performance, presumably due to the higher specific surface area generated by nanotube modification.179 In comparison, the flat current response of CC cath odes over the range of voltages examined indicates that the carbon cloth itself did not catalyze oxygen reduction.

PAGE 61

45 Figure 3.1 LSV Results (Current Density vs Potenti al) of Newly Modified Cathodes Before Installing in MFCs. Current Density Range Wa s Marked Based On the Values Shown in Figure 3.3. The addition of Pt catalyst on the cathodes signifi cantly increased the current densities of all cathodes (Figure 3.1). The current densities of CNTM-Pt, SWNTn-Pt, and CC-Pt cathodes all increased substantially compared with their non-Pt counterparts, and the SWNTn-Pt showed the best performance across the potentia l range. Considering the same amount of catalyst was applied on the electrod es, the LSV results suggest that the 3D structure created by the carbon nanotubes on SWNTn-Pt allowed more Pt nanoparticles to be deposited inside the electrode space rather t han on the surface, leading to increased reaction kinetics.180 The electrochemical performance of cathodes modifi ed by commercial single-walled nanotubes (SWNTc-Pt) and multi-walled nanotubes (MWNTcPt) with Pt coatings was also evaluated using the s ame LSV tests. In general, both cathodes showed lower current densities compared to the other three electrodes described previously. The reason could be attributed to the d ifferent nanotube synthesis approaches. The laser vaporization method used for SWNTn-Pt has been known to produce higher

PAGE 62

46 quality nanotubes than the chemical vaporization me thod used for SWNTc-Pt. The higher crystallization and electric conductivity of SWNTn resulted in higher electron transfer efficiency thus higher performance. The SWNTc-Pt showed larger current response than the MWNTc-Pt across the scanned potentials. This is presumab ly due to the high resistance thus more electrochemical loss of MWNTs compared with SWNTs. The resistance of the reactor with the MWNTc-Pt cathode was 94 which is nearly 3 times higher than with the SWNTc-Pt cathode (35 ). 3.4.2 Performance of MFCs with nano-modified air-ca thodes Figure 3.2 Voltage Generation as a Function of Tim e for the Different Cathodes. The nano-modified air-cathodes were transferred to pre-acclimated single chamber MFCs after LSV tests. Rapid voltage generat ion was observed in all reactors. Voltage / mVTime / h SWNTc-Pt MWNTc-Pt Voltage / mVTime / h CNTM-Pt CNTM Voltage / mVTime / h SWNTn Pt SWNTn Voltage / mVTime / h CC-Pt CC (mV) (mV) (h) (h) (h) (h) (mV) (mV)

PAGE 63

47 Figure 3.3 shows the power density and polarization curves obtained from each MFC reactor during steady power production stages. Figu re 3.2 shows the voltage profiles as a function of time. For MFC cathodes without Pt catal yst coating, CNTM reactor showed the maximum power density of 329 mW/m2, more than twice that of the peak power obtained from CC reactor (151 mW/m2). The SWNTn reactor showed a lower power density (117 mW/m2) than the other two reactors without Pt. Similar t o the results observed during LSV tests, the reactor performance was greatly improved by the application of Pt catalyst in the cathodes. During the first several batches, the power densities of SWNTn-Pt and CNTM-Pt reactors, calculated from voltages at 1000 were 735 mW/m2 and 723 mW/m2, respectively, which were higher than that of CC-P t reactor (672 mW/m2). The voltages of CNTM-Pt and CC-Pt MFCs kept stab le during multiple batches, but the voltage of the SWNTn-Pt reactor decreased gradually from the first batc h and stabilized after the fourth batch, resulting in a voltage drop from more than 600 mV to around 300 mV. As shown in Figure 3.3, the power density curves obtained at the steady state operation of each reactor show the max imum power densities from CNTM-Pt and CC-Pt MFCs of 1118 mW/m2 and 1071 mW/m2, respectively, while the maximum power density of SWNTn-Pt after stabilizing was 302 mW/m2. A similar voltage decline was observed in SWNTc-Pt and MWNTc-Pt MFCs in the first batches. The maximum power density produced by SWNTc-Pt was 522 mW/m2, which was about two times higher than that from MWNTc-Pt (174 mW/m2), confirming that SWNT material performs better as MFC cathodes due to its lower oh mic resistance compared with MWNT material.

PAGE 64

48 Figure 3.3 Power Density as a Function of Current Density (A) and Polarization Curves (B) for MFCs Operated Using Different Air-ca thodes. The electrochemical performance of CC-Pt and SWNTn-Pt cathodes were analyzed again by LSV tests after MFC operation in order to understand the chemical and microbial effects on the cathode performance. Figur e 3.4 shows the electrochemical performance of the used CC-Pt cathode decreased sli ghtly compared with the new cathode, while the current response of the used SWN Tn-Pt cathode declined significantly compared with the new SWNTn-Pt. This drop may explain why the steady state pow er density of the SWNTn-Pt reactor was lower than the CNTM-Pt and CC-Pt re actors despite a better LSV performance with new SWNTn-Pt cathodes. Such finding could also 0 200 400 600 800 1000 1200 02468Power Density / mW m-2 CC CC-Pt CNTM CNTM-Pt SWNTn SWNTn-Pt MWNTc-Pt SWNTc-Pt 0 120 240 360 480 600 720 840 02468Voltage / mVCurrent Density / A m-2 CC CC-Pt CNTM CNTM-Pt SWNTn SWNTn-Pt MWNTc-Pt SWNTc-Pt A B ( A/m 2 ) (mW/m 2 ) (mV)

PAGE 65

49 be confirmed by the changes of open circuit potenti al (OCP). The OCP of the SWNTn-Pt cathode dropped by 35%, from 357 mV in batch 1 to 2 33 mV in batch 4, while the OCP of CC-Pt only dropped by 9%, from 345 mV to 310 mV during the same period. Figure 3.4 Comparison of LSV Electrochemical Test Results between New and Used Cathodes of CC-Pt and SWNTn-Pt. Similar findings in air-cathode performance decline were reported by several other studies, and the reasons were believed to be due to the adherence of biofilm and chemical deposits on the cathode surfaces that redu ced charge transfer and catalyst activity.181, 182 For the SWNTn-Pt cathode, the large amount of pores, cavities, a nd curving paths among carbon nanotubes that increase the specific surface area also enables adsorption of impurities that may cover the surface s of the cathode and reduce charge transfer. Adding a separator on the nanotube modifi ed electrode to block adsorption or switching the nanotube layer to the air-face side c ould be possible solutions.183, 184 It may also be possible that the SWNTn-Pt cathode performa nce decreased due to the loss of the Pt catalyst overtime since the microwave deposition method has not been optimized for -14 -12 -10 -8 -6 -4 -2 0 2 -0.3-0.2-0.100.10.20.30.4Current Density / A m-2Potential / V,vs Ag/AgCl (V, vs Ag/AgCl) (A/m 2 ) SWNTn-Pt-used CC-Pt-used CC-Pt-new SWNT n Pt new

PAGE 66

50 MFC operation. However, this was not confirmed usin g energy dispersive spectroscopy element percentage test during FIB/SEM characteriza tion. 3.4.3 FIB / SEM analysis Scanning electron micrographs were taken for the CC -Pt, CNTM-Pt, and SWNTnPt cathode materials before and after MFC operation Figure 3.5 shows a distinct difference in morphology between the new CC-Pt cath ode and new CNTM-Pt and SWNTn-Pt cathodes. The CC-Pt cathode surface was covered by aggregated particles or short fibers (Figure 3.5A), while the new CNTMPt and SWNTn-Pt electrode surfaces clearly showed the carbon nanotube network (Figure 3.5C, 5E). The deposited Pt catalyst shows up in the images as bright nanoparticles depo sited across the nanotube network. Denser microbial biofilms were formed on the surfac e of the CC-Pt cathode (Figure 3.5B) compared with the CNTMPt and SWNTn-Pt cathodes (Figure 3.5D, 5F), but more chemical deposit covered a large portion o f the surface of the two CNTmodified electrodes. Studies showed that the excess accumulation of biofilm and chemical scales could adversely affect the system p erformance due to the decrease in active cathode specific surface area and increase i n diffusion resistance in oxygen.181, 182 Such observations may also explain the reduced elec trochemical activities in certain MFCs after a period of operation, where the presenc e of cavities and the adsorption of impurities played a major role as compared to catho de biofilms.

PAGE 67

51 Figure 3.5 SEM/FIB Images of New and Used Cathodes : (A) New CC-Pt, (B) Used CC-Pt After MFC Operation, (C) New CNTM, (D) Used C NTM-Pt After MFC Operation, (E) New SWNTn-Pt, and (F) Used SWNTn-Pt After MFC Operation. 3.5 Conclusions Microbial fuel cells provide direct and efficient e lectricity generation from renewable sources, but the current 2-D air-cathode configuration limits system A B C D E F

PAGE 68

52 performance due to the low kinetics of oxygen reduc tion. Carbon nanutubes were used in this study to modify air-cathodes in single chamber MFCs to create a 3-D structure for improved surface area and reaction kinetics. Compar ed with traditional carbon cloth cathodes, all nanotube modified cathodes showed gre ater electrochemical performance as well as higher power density in MFCs. The maximum p ower density obtained from carbon nanotube mat cathodes (329 mW/m2) was more than double the power output from traditional carbon cloth cathodes (151 mW/m2). The addition of Pt catalyst on the cathodes increased the current densities of all cat hodes, with the maximum power density achieved by a CNTM-Pt of 1118 mW/m2. Electrodes made from commercial single wall carbon nanotubes (SWNTc-Pt) have much lower ohmic resistance than those ma de from multiwall nanotubes (MWNTc-Pt) and showed larger current response and higher power. The customized SWNTn electrode showed great electrochemical responses, but its performance declined gradually due to the depositio n of chemical and microbial impurities which blocked reaction surfaces. Scannin g electron micrographs demonstrated different electrode surface morphologies, with the CNT modified cathodes showing carbon nanotube networks and carbon cloth cathodes showing aggregated particles on the electrode surface. 3.6 Acknowledgement This research was supported by the Office of Naval Research (ONR) Grant N000140910944. The authors thank Nanocomp Technolog ies Inc. for donating carbon nanotube mat, and Dr. Paul Rice at University of Co lorado Nanomaterials Characterization Facility (NCF) for helping with FI B/SEM operation.

PAGE 69

53 4. Active Energy Harvesting from Microbial Fuel Cel ls at the Maximum Power Point without Using External Resistors4 4.1 Abstract Microbial fuel cell (MFC) technology offers a susta inable approach to harvest electricity from biodegradable materials. Energy pr oduction from MFCs have been demonstrated using external resistors or charge pum ps, but such methods can only dissipate energy through heat or receive electrons passively from the MFC without any controllability. This study developed a new approac h and system that can actively extract energy from MFC reactors at any operating point wit hout using any resistors, especially at the peak power point to maximize energy producti on. Results show that power harvesting from a recirculating-flow MFC can be wel l maintained by the maximum power point circuit (MPPC) at its peak power point, while a charge pump was not able to change operating point due to current limitation. W ithin 18-hour test, the energy gained from the MPPC was 76.8 J, 76 times higher than the charge pump (1.0 J) that was commonly used in MFC studies. Both conditions resul ted in similar organic removal, but the Coulombic efficiency obtained from the MPPC was 21 times higher than that of the charge pump. Different numbers of capacitors could be used in the MPPC for various energy storage requirements and power supply, and t he energy conversion efficiency of the MPPC was further characterized to identify key factors for system improvement. This active energy harvesting approach provides a new pe rspective for energy harvesting that can maximize MFC energy generation and system contr ollability. 4 The work presented in this chapter has been publis hed by Heming Wang, Jae-Do Park, and Zhiyong Ren in Environ. Sci. Technol., 2012, 46, 5247–5252.

PAGE 70

54 4.2 Introduction A microbial fuel cell (MFC) is a bioelectrochemical system (BES) that employs exoelectrogenic bacteria to oxidize organic matter and produce direct electrical current. Because MFC offers a sustainable solution for remot e sensing and simultaneous pollution control and energy production, it has been intensiv ely researched in recent years, and the improvements in reactor configurations, materials, and operations have led to orders of magnitude increase in power density, from less than 1 mW/m2 to the level of 6.9 W/m2.143, 165 However, most studies operate the MFC with a stati c external resistance or applied potential and report the power density usin g a polarization curve, which assumed that the maximum power density is achieved when the applied external resistance is equal to the MFC internal resistance.177, 185, 186 Such characterizations do represent the theoretical potential of MFC power output, but no u sable energy could be captured, because the electricity generated in such systems i s actually dissipated into heat instead of being utilized by electronics. Moreover, the fix ed external resistance cannot always match the system internal resistance and recover th e maximum power output during MFC operation, because the internal resistance of an MF C varies constantly with changes in microbial activities and operational parameters, su ch as substrate concentration, pH, and temperature.112, 187-189 Studies showed that MFCs may lose more than 50% of produced power across the internal resistance if the operati ng voltage is not at the maximum power point voltage.190 To effectively and efficiently harvest MFC energy, unnecessary resistors need to be eliminated, and technologies need to be develope d to track and harvest energy at the peak level with sufficient controllability. Progres ses have been made in maximum power

PAGE 71

55 point tracking (MPPT) and harvesting systems, such as using perturbation and observation or gradient method to track and optimiz e external resistance.186, 190 For example, Pinto et al. find that MFC power output ca n be significantly improved when real-time resistance optimization was implemented d uring long term operation.186 However, traditional MPPT techniques still use exte rnal resistances and cannot capture and utilize the energy directly. Another harvesting approach is using capacitor-based circuits such as super capacitors and charge pumps, which capture MFC energy passively and transfer it to a boost converter.191, 192 For example, a recently study by Liang et al., showed that current production from a BES reactor c an be increased by 22-32% if an alternative charging and discharging method is used In such operation, a capacitor is firstly charged by the reactor but then discharges the electrons back to the reactor. This is different from traditional intermittent charging, w here a capacitor discharged the collected electrons to a resistor.191 Another study by Kim et al., showed that parallel charging of multiple capacitors can avoid potential voltage reversal while series discharging could increase MFC output voltage.193 The problem of directly using capacitors or charge pumps is that such devices can only passively receive MFC energy at a fixed operating point without any control on t he MFC reactor, and the operating points cannot be adjusted to capture energy at the maximum power density point. In this study, we developed a new energy harvesting approach and system that not only can capture the maximum power from the MFC, bu t also harvests energy actively without using any resistance. Instead of passively receiving electrons from the MFC reactor, this controller can actively extract energ y from the MFC at any operating point, especially at the peak power point to maximize ener gy production. The energy harvesting

PAGE 72

56 efficiency, organic removal, and Coulombic efficien cy of the MFC operated by this maximum power point circuit (MPPC) was characterize d and compared with a common charge pump operation. The energy storage capacity using different numbers of capacitors and system energy conversion efficiency was also investigated for system optimization. 4.3 Materials and Methods 4.3.1 MFC construction and operation Each MFC reactor consisted of two polycarbonate cub e-shaped chambers that were separated by a cation exchange membrane (38 cm2, CMI-7000, Membranes International, NJ).20 The empty volume of either anode or cathode chambe r was 150 mL. Heat treated graphite brushes were used as the anod es, and carbon cloth (projected surface area 38 cm2) was selected as the cathode material.148, 194 MFCs were inoculated with anaerobic sludge obtained from Longmont Wastew ater Treatment Plant (Longmont, CO). The anode chamber was fed with growth medium c ontaining (per liter) 1.25g CH3COONa, 0.31g NH4Cl, 0.13g KCl, 3.32g NaH2PO4 2H2O, 10.32g Na2HPO4 12H2O, 12.5mL mineral solution, and 5mL vitamin solution.177 Phosphate buffered potassium ferricyanide solution (50 mM) was used as the catho lyte to minimize the cathode effects on system performance.20 MFCs were operated in fed-batch mode at the acclim ation stage until repeatable voltage profiles were obtain ed. Reactors were then operated by recirculating anolyte with a 1000 mL reservoir at a flow rate of 45 mL/min and recirculating catholyte with another reservoir at a flow rate of 114 mL/min, respectively. Such operation was aimed to maintain stable substra te and pH conditions so energy harvesting characterization could be focused.19, 195

PAGE 73

57 4.3.2 Maximum power point circuit (MPPC) design and operation NO. Component Function Manufacture Capacitor Stores energy extracted from MFC Taiyo Yu den Inductor Intermediate energy storage before it is transferred to capacitor Triad Magnetics Diode Transfers energy from inductor to capacitor and blocks reverse power flow Micro Commercial Components MOSFET Main switch of energy harvesting converter Vishay Comparator Generates hysteresis voltage band National Semiconductor Potentiometer Adjusts MFCÂ’s working voltage General Potentiometer Adjusts width of hysteresis voltage band General n Connector Monitors MFC voltage TE Connectivity Connector Reference voltage for comparator TE Conne ctivity Figure 4.1 Components in MPPC. The MPPC consisted of a metal-oxide-semiconductor f ield-effect transistor (MOSFET), a comparator, an inductor, a diode, capac itors, potentiometers, and connectors. The detailed information of each MPPC c omponent is listed in Figure 4.1, and the circuit design details is described by Park and Ren.196 Figure 4.2 shows the principles of the energy harvesting MPPC. The MPPC is able to operate the MFC reactor in the vicinity of the maximum power operating poin t, which is regulated by a hysteresis 1 2 3 4 5 6 7 8 9

PAGE 74

58 controller (Figure 4.2A). The hysteresis controller confines the MFC voltage in a predefined range to avoid voltage collapse and ensure enough recovery time of the MFC reactor, and the upper (VthH) and lower (VthL) voltage thresholds can be defined by equation (1).196 () 3 1 2 2// R R R R V Vcc thH+ = () 3 2 2 3 2// // R R R R R V Vcc thL+ = (1) Where Vcc is the external voltage for MPPC circuit, R1, R2, and R3 are internal resistors to set the harvesting hysteresis voltage band, and the double slash means parallel connections of resistors (Figure 4.2). For comparis on with traditional passive energy harvesting approaches, a charge pump (S-882Z24, Sei ko Instruments) was used in a control experiment with the same reactor configurat ion and operation. The operation of the MPPC consists of two modes, CH ARGE and DISCHARGE, according to the energy flow on the inductor connec ted with the MFC (Figure 4.2B). During CHARGE mode, the MOSFET switch is on and dio de is off, and the energy is extracted from the MFC and charged to the inductor (Figure 4.2C). Due to energy extraction, the voltage of the MFC decreases in thi s mode. During DISCHARGE mode, the MOSFET switch is off and diode is on, and the e nergy stored in the inductor is discharged to the capacitor (Figure 4.2D). MFC volt age increases in this mode as it recovers from energy extraction. The controller tur ns off the MOSFET automatically when the MFC voltage reaches lower threshold in CHA RGE mode and turns it back on when the MFC voltage gets to the upper threshold in DISCHARGE mode. The duty ratio and switching frequency can vary depending on the g enerating capacity and recovery time of the operating MFC. The comparator generated hysteresis voltage band according

PAGE 75

59 to the MFC voltage, and the voltage band can be eas ily tracked and adjusted by potentiometers.145 Figure 4.2 Block Diagram of the Maximum Power Poin t Circuit (MPPC): (A) Harvesting Converter Controller. (B) Whole electric Circuit Diagram; (C) CHARGE Phase, MOSFET is On While Diode is Off, Extr acted Energy is Stored in the Inductor; (D) DISCHARGE Phase, MOSFET is Off Wh ile Diode is On, Extracted Energy is Stored in the Capacitors. + i i + MFC -+ -Inductor Capacitor Diode i Inductor -+ Inductor MFC MFC -MOSFET + -Capacitor i A B C D

PAGE 76

60 4.3.3 Analyses The MFC voltage, capacitor voltage, and the output voltage across a current probe (K110, AEMC Instruments) were recorded at 66-sec in tervals using a data acquisition system (Keithley Instrument, OH). The anode potenti al and cathode potential were measured against a Ag/AgCl reference electrode (RE5B, Bioanalysis) inserted in the anode chamber. An oscilloscope (TPS2014B, Tektronic s) was used to continuously monitor MFC voltage, output current and the main sw itch on/off signal. Chemical oxygen demand (COD) was measured using a standard colorime tric method (Hach Company, CO). Polarization curves were obtained by linear sw eep voltammetry (LSV) using a potentiostat (PC4/300, Gamry Instruments, NJ). The scan rate of LSV was 0.1 mV/s with the anode as working electrode and the cathode as c ounter and reference electrode.178, 194 The output power ( P ) of MFC was calculated by PUI = where U is the voltage across the MFC anode and cathode, and I is the MFC output current monitored by the current meter. Power density and current density we re normalized by the projected area of the cathode (38 cm2). Energy ( Wc) consumed by an external resistor ( R) was calculated by 2 c WURdt = and the energy (Wp) supplied by the MFC during harvesting by the MPPC or charge pump was expressed as p WPdt = where dt is 66 sec. The Energy (E) stored in the capacitors was calculated by 2 0.5 ECV =, where C is the capacitance, and V is the capacitor voltage. Energy conversion efficie ncy (ECE) was calculated by 100% pECEEW = .The energy (J) consumption in each MPPC component was calculated based on E = VIt during the harvesting period. Coulombic efficiency ( CE ) was presented as 8000 CEIdtFVCOD =D where F is Faraday constant, V is total volume,

PAGE 77

61 and COD D is COD concentration change. Duty ratio (D) was defined as the ratio of turn-on time to the total switching time, () ononoff Dttt =+, where on t and off t is the on and off time of the MOSFET, respectively. 4.4 Results and Discussion 4.4.1 MPPC can operate the MFC at the maximum power harvesting range Figure 4.3 MFC Polarization Curve and Power Densit y Curve Obtained by Linear Sweep Voltammetry (LSV). The Scan Rate of the Polar ization was 0.1 mV/s. : Operating Point of the Charge Pump. : Operating Range of the MPPC. Recirculating-flow MFC Open Circuit Potential was 6 88 mV. Figure 4.3 shows the polarization and power density curves obtained in the steady-state recirculating flow MFC reactor. The ma ximum power density produced by the MFC was around 1370 mW/m2 when the reactor voltage was between 372 mV and 316 mV, with an average of 344 mV. The correspondin g external resistor at the peak power density was 23 In order to harvest the maximum power identified by the MFC 0 200 400 600 800 1000 1200 1400 1600 0 100 200 300 400 500 600 700 800 02468Power Density (mW/m2) Voltage (V)Current Density (A/m2) Charge pump operating point MPPC operating range 372 mV 316 mV 633 mV

PAGE 78

62 power density curve, the upper voltage threshold at 372 mV and lower voltage threshold at 316 mV were determined to form an energy extract ion band for the hysteresis controller in MPPC. In contrast, the charge pump wa s only able to harvest the MFC energy at the 633 mV (317 mW/m2) due to the current limitation of the charge pump (Figure 4.3). The difference in the operating point s on the power density curve demonstrates that the MPPC could be modulated to ha rvest energy at the range of the peak point while the power harvesting by the charge pump was limited at lower points due to the lack of controllability. Figure 4.4 Snapshot of On/Off Cycle of the MPPC Du ring Active Energy Harvesting from MFCs and the Voltage and Current Pr ofiles. One division of XAxis Represents 100 sec. The Figure Shows the Wave forms of 1 msec Duration in Terms of Current, Voltage, and On/Off Switch Change s. Typical operation cycles of the MPPC harvesting are shown in Figure 4.4. When the MFC voltage reaches to 372 mV, the MPPC activel y extracts energy from the MFC and charge the inductor (CHARGE mode). The extracti on stops when the voltage drops to 316 mV. While waiting for the MFC voltage to rec over, the controller discharges the energy from the inductor to the capacitor (DISCHARG E mode). Once the MFC voltage Charge Discharge On/off time Voltage Current

PAGE 79

63 recovers back to 372 mV, the controller charges the inductor again. The switching frequency between CHARGE and DISCHARGE phase was ve ry fast and in the order of kHz. The duration of CHARGE and DISCHARGE phases de pend on the MFC condition, and the DISCHARGE phase was also affected by the ta rget of the capacitor voltage. The range of operation can be tracked and controlled by the Hysteresis controller.145 Anolyte and catholyte recirculation operation was used in t his study because such system could maintain a relatively stable substrate concentratio n, pH, and other operating conditions as compared to fed-batch operation and thus reduces th e effects of environmental factors. 4.4.2 MPPC harvests energy more actively and effici ently MFC power density curves demonstrate that when the applied external resistance is equal to the MFC internal resistance, the maximu m power can be achieved. Figure 4.3 shows that the peak power of the MFC used in this s tudy could be obtained at 23 and Figure 4.5 shows the MPPC-controlled MFC was operat ed nearly as the same condition as the reactor operated under a 23 resistor. The operating curves of the anode potent ial, cathode potential, and reactor voltage in both cond itions basically overlapped each other, indicating very similar operating conditions, where the maximum power could be generated. However, instead of dissipating the ener gy into heat as resistors do, the MPPC captured the energy and stored energy into capacito rs. Energy harvesting results show that the MPPC-contro lled MFC was able to charge multiple capacitors (Taiyo Yuden, PAS1016LR2 R3205). After 18 hours of operation, the voltage of the 12 capacitors connect ed to the MPPC controller increased from 0 V to 2.5 V, and the MPPC extracted 214.1 J o f energy from the MFC, in which 76.8 J were stored in the capacitors (Figure 4.6A). In comparison, the charge pump was

PAGE 80

64 able to charge 1 capacitor to 1.0V during the same period, and the total extracted and stored energy was 23.2 J and 1.0 J, respectively (F igure 4.6A). The results show that by actively extracting energy at the maximum power poi nt, the MPPC harvested 76 times more energy than the charge pump. Figure 4.5 Comparison of MFC Voltage, Cathode Pote ntial, and Anode Potential between the MPPC Active Energy Harvesting Condition and 23 Ohm External Resistor Condition. The Optimum External Resistance was Calculated to be 23 Ohm Based on Polarization Curve that Could Yield th e Maximum Power Density. Comparable substrate degradation was observed in bo th energy harvesting operations, as the COD removals were 49.8% and 47.1 % for the MPPC and charge pump, respectively (Figure 4.6B). However, the Columbic e fficiency of the MPPC operation was 15.6%, 21 times higher than that of the charge pump (0.7%). This finding is consistent with previous studies that higher Columb ic efficiency can be achieved by operating MFCs at an optimal external resistance.186 Moreover, compared to other studies that selected the optimal resistance to demonstrate the power generation potential, the MPPC actually captured the energy at the maximum po wer point that is available for -500 -300 -100 100 300 500 700 048121620Voltage/Potential (mV)Time (h) MFC_MPPC Cathode_MPPC Anode_MPPC MFC_23ohm Cathode_23ohm Anode_23ohm

PAGE 81

65 electronic utilization. Ferricyanide solution was u sed as the catholyte in this 2-chamber MFC study, so no cathode biofilm was observed to co nsume substrate and affect Columbic efficiency. Figure 4.6 (A) Comparison of Energy Harvesting by the MPPC and the Charge Pump and Energy Stored in Capacitors. (B) Compariso n of COD Removals in the MPPC and Charge Pump Conditions. In the MPPC Test, 12 Capacitors were Connected in Parallel for Energy Storage. In the Ch arge Pump Test, one Capacitor was Enough to Store All the Harvested Energy from M FC. 4.4.3 The numbers of capacitors for energy storage Different numbers of capacitors were tested in the study for energy storage and to provide stable electricity for electronic devices. While resistors donÂ’t capture any energy, and the charge pump was only able to charge 1 capac itor, the MPPC harvested so much energy that multiple capacitors had to be used for energy storage. The charging behavior of using 3, 6, 9, and 12 capacitors during 18-hour harvesting was characterized. The capacitors were connected in parallel in order to m aintain charging efficiency. 0 100 200 300 400 500 600 700 800 900 1000 0 50 100 150 200 250 Energy extracted from MFC Energy stored in capacitors Start_CODEnd_CODEnergy (J) COD (mg/L) Maximum power point circuit Charge pumpB A

PAGE 82

66 Figure 4.7 (A) Voltage Profile and (B) Energy Stor age Differences by Using 3, 6, 9, and 12 Capacitors in Parallel During MPPC Active En ergy Harvesting. Figure 4.7 shows that with the increase of charging time, the voltage of the 3capactior condition increased faster than other con ditions and reached the saturated level of 2.9 V in 4 hours. The stored energy in the 3 cap acitors was 25.2 J within this period. After 4 hours, the 3-capacitors could not store mor e energy due to saturation, but the MPPC kept harvesting energy from the MFC, as eviden ced by the 6, 9, and 12-capacitor conditions. With increasing numbers of capacitors, the voltage increase rate declined, but 0 500 1000 1500 2000 2500 3000 3500 0 4 8 12 16 20 Capacitor voltage (mV) 0 10 20 30 40 50 60 70 80 90 048121620Energy (J)Time (h)3C 6C 9C 12C 3C 6C 9C 12C A B

PAGE 83

67 the total amount of energy stored in capacitors inc reased. The energy storage in the 6, 9, and 12-capacitor condition was 47.0 J, 65.6 J, and 76.8 J, respectively, and the corresponded voltage after 18 hours was 2.8 V, 2.7 V, and 2.5 V, respectively. Because the larger the capacitance, the longer the charging time is required, which resulted in differences in capacitor voltages and energy storag e. More energy can be stored in higher capacitance conditions if the harvesting continues. These results indicate that different numbers of capacitors or capacitors with different capacitance can be used as energy storage for operating electronic devices. The requi red number of capacitors, charging time and energy storage capacity are determined by the characteristics of end users, and the MPPC-controlled MFC should not be a limiting fa ctor under stable operating condition. 4.4.4 Conversion efficiency of the MPPC The MPPC can actively harvest energy at the maximum power point thus significantly increased energy generation from MFCs However, like any electronic device, the MPPC consists multiple electronic compo nents that consume energy, which reduce MPPCÂ’s energy conversion efficiency. When th e MFC was operated under 23 the optimal external resistance that lead to the ma ximum power density, the MFC could provide 215.7 J (prorated by duty ratio) of energy in 18-hour, even though all such energy will be dissipated through heat by the resis tor. In comparison, the MPPC harvested 214.1 J of energy from the MFC without ex ternal resistors and transferred 76.8 J to the capacitors. This further confirmed that th e MPPC was able to harvest 99.2% of the energy from the MFC, but it also shows that onl y 35.9% of the harvested energy was transferred to the capacitors. Even though the MPPC -transferred energy is still 76 times

PAGE 84

68 higher than that from the charge pump (1.0 J), it i s important to identify the limiting factors within the MPPC circuit and improve the con version efficiency. Figure 4.8 shows the MPPC efficiency through an 18-hour test, and it can be seen that the efficiency increased sharply at the beginning and then stabili zed till a decline was observed due to the saturation of capacitance. The highest energy e xtraction efficiencies occured between 8.3 to 11.5 hour, with an efficiency of 42.1%. Figure 4.8 Efficiencies through 18-hour Test. The power consumption of each component in the MPPC circuit was calculated in Table 4.1. Figure 4.9 illustrates the percentage of energy loss in each component during MFC energy extraction by the MPPC during an 18-hour period. While most MPPC components consumed minimum amount of energy, the d iode contributed to 58.8% of the energy loss within the MPPC, indicating that it is the single element that needs to be replaced or improved. The diode is used in the MPPC to transfer the energy from the 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5 10 15 20 25 30 35 40 45 0.05.010.015.020.0Voltage (V) Efficiency (%)Time (h) Efficiency MFC voltage Capacitor voltage

PAGE 85

69 inductor to capacitors and blocks reverse flow. Adv anced converters are currently being developed to replace the diode and improve MPPC con version efficiency. Table 4.1 Analysis of Energy Extraction Efficiency by MPPC. Components Information from datasheet Calculation Energy stored in 12 capacitors after 18 h Capacitance: 2F n r Inductor DC Resistance Max: 0.58ohm n r n Capacitor ESR (Equivalent Series Resistance): 50.0 mohm for each n r Diode Forward voltage drop: 0.72V at 12mA n MOSFET Drain-Source On-State Resistance: 0.033 ohm at V GS =1.8V r Total extracted energy n ! n Total energy provided by MFC Error n Energy extraction efficiency # Average current in the circuit is 12 mA; Duty ratio is 138s:40s. Figure 4.9 Energy Conversion Efficiency and Distri bution of Internal Energy Loss in the MPPC. The Distribution was Quantified Based on an 18-hour, 12-capacitor Operation. Diode 58.8% Inductor 2.5% MOSFET 0.1% Others 2.6% Net Gained Energy 36%

PAGE 86

70 4.5 Outlook MFC technology has been considered as a sustainable method to directly produce energy from biodegradable substrates, but the impro vement of power density has been stagnant for several years after significant advanc ements in reactor configuration and material development. Compared with traditional app roaches that use external resistors and charge pumps, this study demonstrates a new act ive approach to harvest energy from MFCs. Instead of passively receiving electrons from the MFC, the MPPC actively extracts energy from the MFC at the peak power poin t. The remarkable increase in energy generation by the MPPC compared to the commo n charge pump shows this approach is much more efficient and effective to ca pture MFC energy. There are very few charge pumps available for MFC systems, and the charge pump used in this study is representative, because it has been used by many ot her studies in different conditions.119, 197, 198 The active energy harvesting approach is new to MFC operation, and there are many questions remain to be answered. For example, one unique feature of MFCs is the variable biocatalyst density on the electrodes. Exo electrogenic bacteria transfer electrons to the anode electrode and gain energy during anaer obic respiration. Within the capability of bacterial extracellular electron transfer, the m ore electrons get extracted from the external circuit, the less electrons and energy bec ome available for microbial growth. Therefore, it is important to understand how the ac tive harvesting affects microbial activity, community, and metabolisms, so a balanced and sustainable reactor performance can be maintained. We did not find active harvestin g negatively affect MFC performance in terms of power density and Coulombic efficiency in recirculation operation. In

PAGE 87

71 addition, further optimization of the MPPC circuit needs to be conducted to reduce internal energy loss. Systems with precise tracking capability will allow the circuit to adjust and maintain the maximum energy extraction b ased on real-time changes of MFC condition due to the variations of environmental co nditions such as pH, temperature, and substrate concentration. 4.6 Acknowledgement This work was supported by the Office of Naval Rese arch (ONR) under Award N000140910944. We thank Drs. Bruce Logan and Peter Jenkins for constructive discussions.

PAGE 88

72 5. Power Electronic Converters for Microbial Fuel C ell Energy Extraction: Effects of Inductance, Duty Ratio, and Switching Frequency5 5.1 Abstract Power converter based microbial fuel cell (MFC) ene rgy harvesting has been recently researched to replace the external resisto rs that have been utilized to show MFC output in many studies. The electronic circuit can operate as an equivalent external resistor, but the energy generated from MFC can be harvested in storage instead of being dissipated. However, there is limited information i n the literature about the effects of operating configuration of power electronic circuit s on MFC energy harvesting. In this study, a boost-converter based energy harvester cir cuit was examined in terms of inductance, duty ratio, and switching frequency. Th e results showed that all of these factors play important roles for the performance of MFC and energy harvesting, and their effects can be cross linked. Current and voltage is generally proportional and inversely proportional to the inductance, respectively. The t otal harvested energy and efficiency vary significantly by combinations of duty ratio an d switching frequency. For the MFC reactor tested in the study, the highest energy har vested was 3.48 J which was under the combination of 14 mH inductance, 75% duty ratio and 5000 Hz frequency, comparing to the highest efficiency of 67.7% happened at 130 mH inductance, 25% duty ratio and 4000 Hz frequency. When using the smallest inductan ce of 0.45 mH the highest energy and efficiency were only 1.38 J (50% duty ratio and 5000Hz frequency) and 19.9% (25% duty ratio and 5000Hz frequency), respectively. Reg ardless of the voltages and currents 5 The work presented in this chapter has been publis hed by Heming Wang, Zhiyong Ren, and Jae-Do Park in J. Power Sources 2012, 220, 89–94

PAGE 89

73 produced in various operating configurations, anode potentials were stable, suggesting that there were enough electrons available to be ut ilized for current generation. An optimal operating configuration that provides ideal system performance can be found for different reactors and applications. Keywords: microbial Fuel Cell, energy extraction, D C/DC converter, inductance, duty ratio, switching frequency

PAGE 90

74 5.2 Introduction Microbial fuel cells (MFCs) are considered new gree n energy sources due to their wastewater treatment and simultaneous energy produc ing capability. In most studies so far, an external resistor was inserted between the anode and cathode of an MFC, so the voltage across the resistor can be monitored to dem onstrate the power production. However, this method does not harvest any energy be cause the energy is dissipated on the resistor as heat. Furthermore, the relatively l ow voltage (<1.0 V) and low power (~W/m3) output from an MFC cannot directly support majori ty of commercial electrical devices, which is one of the biggest obstacles for practical application of MFC. Recently power electronics based harvester circuits for MFCs has been researched,145, 193, 199-201 aiming significantly improved MFC energy harvest an d output voltage boost, which can be a crucial step to make MFC technology commercial ly viable. Different from conventional operations using external resistors, e nergy generated from MFCs will be collected and stored, which in turn will be utilize d to power electrical devices, for example, wireless sensors to monitor environment.119, 198 To design boost-converter based circuits for more v ersatile and efficient MFC energy harvesting, there are three fundamental fact ors to consider: the inductance of an inductor, extraction duty ratio (also known as duty cycle), and extraction frequency (i.e. converter switching frequency). The inductor is the intermediate energy storage for an MFC and determines the rate of current change and l evel of energy extraction. Large inductance makes the current changing slowly while small inductance makes it faster, and this contributes to determine the MFC terminal voltage. The duty ratio governs the relative duration of energy extration in a certain switching period. In other words, it

PAGE 91

75 determines the time period of energy extraction and MFC recovery. In MFC recovery period, the energy harnessed in inductor is transfe red to the capacitor storage. Higher duty ratios lead to longer energy extraction and sh orter time for MFC recovery, while low duty ratios allow more time for MFC to recover and energy transfer. The switching frequency determines the number of energy extractio n and transfer within a given time period. Higher switching frequency means the energy extraction happens more frequently, but each extraction is shorter for a given duty rat io. The total energy extraction time is determined by duty ratio regardless of switching fr equency, which can be given as extprd TDT =. For example, the energy extraction with 50% duty ratio at 1000Hz switching frequency has the same energy extraction time of 500 msec per 1 second period as the 50% duty ratio at 2000Hz frequency, but the number of energy extraction cycle for the first case is 500 times, only half of the latte r case. However, it has been revealed that different combinations of duty ratio and switching frequency affect MFC energy harvesting results, even if the energy is harnessed for the same amount of time. There are very few studies that investigated how to capture MFC energy more effectively through the design and optimization of electronic harvesting circuits, especially by using high-speed switching converters Dewan et al.202 concluded that intermittent energy harvesting (IEH) by alternative ly collecting energy in the capacitor and dispensing it through a resistor was more effec tive than continuous energy harvesting (CEH) with constant energy extraction. The capacito r was charged for hours but discharged for only less than a minute, which indic ated that electroactive species around the electrode was replenished while the capacitor w as being discharged. Gardel et al.203 obtained similar results with duty cycling based en ergy harvesting from a multi-anode

PAGE 92

76 MFC, which suggests that it was necessary to replen ish depleted electron donors within the biofilm and surrounding diffusion layer to maxi mum charge transfer. Grondin et al.204 also investigated the power output as a function of duty cycle, but the effect of extraction frequency was not studied. We recently developed a boost converter based energ y harvesting circuit for MFCs,144 and our results showed that the new active harvest ing approach was much more efficient than passive charge pump method, as the e nergy output increased by 76 times.144 In the active harvesting circuit, energy extraction was controlled within a voltage band at the MFCÂ’s maximum power points. The selection of in ductance was based on a fixed condition of the MFC, but the duty ratio and switch ing frequency were flexibly controlled based on MFC condition144 In this study, we investigated the energy extracti on with different inductances, duty ratios, and switching f requencies to characterize how these parameters affect MFC energy output performance. Th e energy harvesting frequency or switching frequency of the power converter ranges f rom 100 to 5000 Hz, which means that our switching periods (10 msec 200 sec) wer e orders of magnitude shorter than previous studies, which were in the range of hours,202 minutes191 and seconds.203, 204 5.3 Materials and Methods 5.3.1 MFC construction and operation As shown in Figure 5.2, a two-chamber MFC reactor w ith anode and cathode chamber separated by cation exchange membrane (38 c m2, CMI-7000, Membranes International) was used in this study.144 The reactor was originally inoculated by anaerobic sludge from Longmont Wastewater treatment Plant (Longmont, CO) and has been operated stably for nearly one year. The empty volume of anode or cathode chamber

PAGE 93

77 each was 150 mL. Anode was a treated graphite fiber brush (Gordon Brush) and cathode was a 38 cm2 plain carbon cloth (Fuel Cell Earth). To maintain stable conditions of both anode and cathode during tests, anolyte and catholy te was separately recirculated from a 1000 mL reservoir. The flow rates of recirculation were 45mL/min and 114 mL/min for anolyte and catholyte, respectively. The anolyte wa s sodium acetate dissolved in 50 mM phosphate buffer containing 1.25g of CH3COONa, 0.31g of NH4Cl, 0.13g of KCl, 3.32g of NaH2PO42H2O, 10.32g of Na2HPO412H2O, 12.5 mL of mineral solution, and 5 mL of vitamin solution.177 The catholyte was potassium ferricyanide dissolved in 50 mM phosphate buffer contains 16.5g of C6FeK3N6, 3.32g of NaH2PO42H2O and 10.32g of Na2HPO412H2O. All of the tests were conducted at room temperat ure. Figure 5.1 Schematic Diagram of the Experimental S etup. PC monitor Datalogger Waveform generator Catholyte reservoir Anolyte reservoir Oscilloscope Electric circuit Current probe MFC

PAGE 94

78 5.3.2 Energy extraction circuit design Energy extraction circuit based on boost converter was composed of a metaloxide-semiconductor eld-e$ect transistor (MOSFET, Si3460BDV, VISHAY), an inductor (Triad Magnetics), a Schottky diode (1N571 1, Micro Commercial Components) and a 2 F supercapacitor (PAS1016LR2R3205, Taiyo Yu den). The MOSFET is the main switch of the circuit and controlled by a 15 MHz fu nction/arbitrary waveform generator (33120A, Agilent Techonologies). The function/arbit rary waveform generator can generate square waves in various duty ratios and fr equencies that turn MOSFET on and off to extract energy from MFC in different conditi ons. The inductor is a temporary energy storage while the MOSFET is on, and the stor ed energy is transfered to capacitor when the MOSFET is off. The Schottky diode blocks r everse power flow from capacitor to inductor and automatically turns on when the MOS FET is off due to the induced voltage across the inductor. The capacitor is the t erminal energy storage in this study. Figure 5.2(a) shows block diagram of the energy ext raction circuit controlled by the function generator. The circuit was operated un der two modes: CHARGE (Figure 5.2(b)) and DISCHARGE (Figure 5.2(c)). Under CHARGE mode, MOSFET is on and switching diode is off, energy harvested from MFC i s stored temporarily in the inductor; Under DISCHARGE mode, MOSFET is off and switching d iode is on, energy stored in the inductor is transferred to the capacitor. After alternative operation under CHARGE and DISCHARGE modes, energy can be cumulated in the capacitor. Detailed operation of the energy harvester can be found in authors ¢ previous work.144, 201

PAGE 95

79 Figure 5.2 Block Diagram of Energy Extraction Circ uit: (a) Energy Harvesting Converter, MOSFET is Controlled ON/OFF in Different Duty Ratios and Frequencies by Function Generator; (b) CHARGE Mode, MOSFET is On and Switching Diode is Off, Energy Extracted from MFC i s Stored in the Inductor Temporally; (c) DISCHARGE Mode, MOSFET is Off and S witching Diode is On, Energy Stored in the Inductor is Transferred to the Capcitor. 5.3.3 Tests In this study, three different inductors (RC-7 (0.4 5mH), CST206-1A (14mH), and CST206-3A (130mH), Triad Magnetics), three duty rat ios (25%, 50% and 75%) and seven switching frequencies (100Hz, 500Hz, 1000Hz, 2000Hz, 3000Hz, 4000Hz and 5000Hz) were examined. In each set of the tests, sw itching frequency was changed with one fixed inductor and duty ratio. So there were to tal nine sets and each set includes i i -+ Inductor MFC i + MFC -+ -Inductor Capacitor i -+ Diode Inductor MFC -MOSFET + Capacitor Function Generator (a) (b) (c)

PAGE 96

80 seven tests. Each test lasts for 30 min to examine one specific configuration of inductor, duty ratio and switching frequency. Before each tes t batch, MFC reactor was fed with new anolyte and catholyte and operated under a 23 resistor to facilitate the recovery of electrochemical active bacteria. A 23 resistor was used because it is correlated to the highest power density based on the systemÂ’s polariz ation curve. If MFC reactor output voltage was stable at 3555mV with 23, meaning the MFC was maintained at the maximum power point, the reactor was assumed to be ready for the test. For each test in a different condition, the reactor was initially kept at open circuit condition until it reached an open circuit voltage of 7055mV. Then the charac terization was conducted from this open circuit condition. The MFC voltage, anode potential, cathode potential capacitor voltage, and voltage across the current probe were recorded ever y 66 seconds by data acquisition system (Model 2700, Keithley Instrument). Anode pot ential and cathode potential were measured against an Ag/AgCl reference electrode (RE -5B, Bioanalysis) inserted in anode chamber and cathode chamber, respectively. A digita l storage oscilloscope (Tektronix TPS2014) was used to continuously monitor MFC volta ge, output current and duty ratio. The energy stored in the storage capacitor (E) was calculated by 20.5 ECVdt = where C is the capacitance, and V is storage capacitor voltage. Energy supplied by t he MFC during harvesting ( W ) was expressed as MFCMFC WVIdt = where VMFC is the voltage across the MFC anode and cathode, IMFC is the MFC output current and the sampling time dt is 66 seconds. Energy harvesting efficiency ( EHE ) was calculated by /100% EHEEW = Duty ratio ( D ) was defined as /() ononoff Dttt =+, where on t and

PAGE 97

81 off t is the on and off time of the MOSFET, respectively 144. Arithmetic mean values of MFC voltage, MFC current, anode potential, and cath ode potential are computed for each 30-minute test. 5.4 Results and Discussion 5.4.1 Effects on MFC voltage and current Figure 5.3 MFC Voltages (a)-(c) and MFC Current (d )-(f) During Energy Extraction under Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130m H. The MFC voltages showed the same trend with differe nt inductances, duty ratios and frequencies (Figure 5.3(a) – 3(c)). MFC voltage s decreased with increasing duty ratios and decreasing frequencies. Higher duty rati os and low frequencies mean that more time was used for energy extraction from the MFC, s o MFC voltage decreases from open circuit voltage in response to the energy extractio n. Comparing different inductances, small inductances led to lower MFC voltages than la rger inductances, because the smaller

PAGE 98

82 inductance will introduce larger current. The two l arger inductances (14 mH and 130 mH) showed similar MFC voltages especially with high sw itching frequencies because of the similar current amplitudes. The current does not sh ow much difference between 14 mH and 130 mH when its amplitude is very low. The relationship between MFC output current and the inductor can be given as (1/)MFCMFC ILVdt = where L is the inductance, VMFC is the MFC voltage, and dt is sampling time. As shown in Figure 5.2(b), the volta ge across the inductor VL is identical with MFC output voltage when the MOSFET is on in CH ARGE mode. Hence, average MFC output current is inversely proportional to the inductance for a given MFC voltage output, which can be seen in Figure 5.3(a) 3(f). Although smaller inductance extracts more current, the corresponding voltage is low ther efore resulting in low power output. The rate of current change also depends on the indu ctance. As can be seen in the voltage and current relationship in the differentia l form, (/) MFCMFC VLdIdt = the current is increasing fast if the inductance is small for a given voltage. This causes a fast decrease of MFC voltage as well in DISCHARGE period, which g enerates fluctuations in MFC output voltage, current, and power in the given swi tching frequency. On the contrary, larger inductance will make them less fluctuating a nd closer to a constant value due to smaller dI/dt. Typical instantaneous MFC current and voltage wav eforms of small and large inductance are shown in Figure 5.4(a) and (b) It can be seen that the voltage and current are smoothed and the current level is low w ith the high inductance.

PAGE 99

83 Figure 5.4 Waveforms of MOSFET Gating Signal (Top) MFC Votlage (Middle), and MFC Current (Bottom). MOSFET is Turned on When the Gating Signal is in High State. (a) Inductance 0.45mH, Switching Freque ncy 100Hz, Duty Ratio 50%; (b) Inductance 14mH, Switching Frequency 1000 Hz, D uty Ratio 50%. The MFC voltage and current are also a function of duty ratio. As the duty ratio represents the conduction time of MOSFET, which is the time duration for MFC output terminals connected to the inductor, the average cu rrent increases and voltage decreases with increasing duty ratio. Under a low duty ratio, ON time is short and fewer electrons are extracted, so the produced current is low altho ugh there is a long OFF time to replenish electron donors; if ON time is long and m ore electrons are extracted with a high duty ratio, so the produced current becomes high bu t the short OFF time may reduce time for electron donor replenishment. Hence, there shou ld be an optimum duty ratio to balance ON and OFF time for a given MFC condition. The experiment with lower inductance case shows more linear relationship betw een voltage and current in terms of duty ratio. The generated energy surface in low ind uctance experiment shows clear peak point similar to the polarization curve as can be s een in Figure 5.6(a). For higher inductances, the current and voltage do not change as much (Figure 5.3(a) 3(f)).

PAGE 100

84 The switching frequency also affects the MFC voltag e and current. Similar as the duty ratio, it has higher effects when inductance i s smaller. For a given duty ratio, MFC generates more current with lower switching frequen cy because more electrons were extracted at low switching frequency and the reacto r can have more time for recovery. This should be very important factor for switching converter-based energy harvesting system designs, because MFC voltage may collapse wi thout proper amount of recovery time. However, due to the high switching frequencie s (100Hz 5000Hz) used in this study, the cycle times were very short ranging from 200 sec to 10 msec. Therefore, when energy was extracted within this short period, there were enough electrons available around the electrode for next extraction because only a small portion of the electrons was pulled out. At each duty ratio, curre nt decreased as switching frequency increased because fewer electrons were extracted du e to short energy extraction time. The MFC current and voltage showed a tendency to be sta bilized after a certain switching frequency (Figure 5.3(a) 3(f)). In the experiment s in this paper, recirculation of anolyte helped improve the mass diffusion and replenish ele ctrons at the electrode. As can be seen in Figure 5.3(a) – (f), cathode pote ntials showed a significant differences for a given duty ratio and switching fr equency between low (0.5mH) and high inductances (14mH and 130mH) even with enough catho lyte provided during the tests, compared to the anode potentials that were relative ly stable. This result suggested that the different duty ratios and frequencies in energy har vesting would not affect the activity of mature anode biofilm (developed on the anode for ne arly one year). This confirms that the duty cycling itself had little or no effect on gross community composition on the anode.203 The stable anode potential also suggested that eno ugh electrons which could be

PAGE 101

85 stored in the haem groups of exocytoplasmic cytochr omes205 were available to be utilized for current generation. If there are not enough ele ctrons produced from bacteria, the problem of overshoot would happen which will reduce the voltage.206 Figure 5.5 Anode Potentials (a)-(c) and Cathode Po tentials (d)-(f) During Energy Extraction under Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130m H. 5.4.2 Effects on MFC Energy and Efficiency The energy that can be stored in the inductor, EL, can be given as 2 (0.5) LL ELI =, where IL is the average inductor current. Neglecting small r esistance in inductor and assuming long enough time for discharge, this energ y can represent the generated energy from MFC. For a given inductance, the generated pow er and energy are dominated by duty ratio changes (Figure 5.6(a) – 4(c)), because the duty ratio mainly determines the MFC current and voltage. In the experiment with low inductance (0.5mH), the generated energy surface shows maximum energy generation poin ts for each switching frequency

PAGE 102

86 set because the low inductance allows the MFC gener ates high enough current to show the polarized characteristics with the given duty r atios (Figure 5.6(a)). With other high inductances, the reactor was not able to generated maximum power available even with a high duty ratio. Therefore it can be suggested that the inductance should be carefully selected if the maximum power point operation is re quired. Figure 5.6 MFC Generated Energy (a)-(c), Harvested Energy (d)-(f), and Efficiencies (g)-(i) During Energy Extraction Under Different Inductances, Duty Ratios and Frequencies. Left Column: 0.45mH, Middle Column: 14mH, and Right Column: 130mH.

PAGE 103

87 (a) (b) (c) ( d) (e) (f ) Figure 5.7 Waveforms of MFC Current (Top), MFC Vot lage (Middle), and MOSFET Gating Signal (Bottom). MOSFET is Turned on When the Gating Signal is in High State. (a) Inductance 0.45 mH, Switching Frequency 1000 Hz, Duty Ratio 25%; (b) Inductance 130 mH, sSwitching Frequency 50 00 Hz, Duty Ratio 25%; (c) Inductance 0.45mH, Switching Frequency 100Hz, Duty Ratio 50% ; (d) Inductance 14mH, Switching Frequency 500 Hz, Duty Ratio 50%; ( e) Inductance 0.45 mH, Switching Frequency 2000 Hz, Duty Ratio 75%; (f) In ductance 14 mH, Switching Frequency 5000 Hz, Duty Ratio 75%.

PAGE 104

88 The energy harvesting efficiencies can reach up to 60% (Figure 5.6(h), 6(i)) with large inductances. These were much higher than thos e with small inductance which was under 20% (Figure 5.6(g)). Although MFC reactor can generate more energy under small inductance than large inductances (Figure 5.6(a) – (c)), only a small portion of the energy was harnessed because of the low MFC voltage. This is due to the relatively high voltage drop in the harvesting circuit compared to the low input voltage level. Similarly, efficiency is high when duty ratio is low and switc hing frequency is high (Figure 5.6(h), (i)), but energy output is low (Figure 5.6(e), (f)) The two larger inductances had the similar results on MFC generated energy because the y had the comparable current and voltage. The same trend was achieved by all of the three ind uctances: the highest efficiency had been reached at the lowest duty rati o and the highest frequency. However, this operating point may not be desirable because t he power level is quite low due to the low current. As can be seen in Figure 5.6, the harv ested energy increases as duty ratio increases, but the efficiency decreases slightly. T herefore an optimal combination of the duty ratio and switching frequency for a given indu ctance can be found for certain operating conditions, such as maximum power or maxi mum current operation. 5.4.3 Discussion This energy harvesting circuit can be seen as a var iable resistance box as a function of inductance, duty ratio and frequency wh ich can operate MFC at different conditions. Although energy harvesting circuit work s as a resistor, it is utterly different for its controllability and energy harvesting capab ility. Resistors only show the energy output potential but cannot harness any usable ener gy. Energy harvesting circuit can

PAGE 105

89 control a specific bioelectrochemical system (BES) as needed and meanwhile energy can be harvested and stored for different uses. For exa mple, for microbial electrolysis cell (MEC) or microbial desalination cell (MDC) reactors higher current can increase hydrogen production rate9 or desalination rate,19 so low inductance, large duty ratio and low frequency should be considered. If running a mi crobial fuel cell (MFC) reactor chasing the maximum power output, medium inductance medium duty ratio and medium frequency would be a better choice. However, high e nergy extraction efficiencies were obtained under the large inductance. Our previous w ork 144 when using maximum power points circuit (MPPC) to extract energy from MFC, t he three parameters were happened to be 14 mH inductance, 77.5% duty ratio and 5500 H z frequency. At this combination, the harvested energy was 214 J and the efficiency w as only 36% within 18 h, which was much lower than the highest efficiency (67.7%) achi eved in this study. The future work could adjust the three interdependent parameters to extract energy at the maximum power points to obtain higher energy and extraction effic iency. It would be necessary to balance among inductance, duty ration and frequency to find out the most suitable operation parameters for a specific BES. 5.5 Acknowledgement This research was supported by the Office of Naval Research (ONR) under Award N000140910944.

PAGE 106

90 6. Removal Mechanisms of Trace Organic Compounds in Microbial Fuel Cells6 6.1 Abstract Microbial fuel cell (MFC) is an environmental platf orm technology that integrates microbial degradation and electrochemical conversio n to remove organic and inorganic contaminants from wastewater. The objective of the study is to understand and characterize the removal mechanisms of different tr ace organics (TOrCs) in a unique MFC environment, where the anode chamber provides a n anaerobic and reductive condition and the cathode chamber provides an oxida tive and aerobic environment. In the study, 34 TOrCs with various physico-chemical prope rties were spiked into artificial wastewater containing acetate. Single-chamber air-c athode microbial fuel cells (SMFCs) and two-chamber air-cathode microbial fuel cells (T MFCs) were constructed to provide combined or separated oxidation/reduction environme nts for TOrCs removal. Results showed that the spiked TOrCs had no negative impact s on bacteria on the electrodes, and all the 34 TOrCs showed significant removal in MFCs with extra power production. 6 Manuscript in preparation for Environ. Sci. Technol.

PAGE 107

91 6.2 Introduction Microbial fuel cell (MFC) is a bioelectrochemical s ystem (BES) that any biodegradable substrate in wastewater.33, 207-210 theoretically can be oxidized by donating electrons to the anode, while O2 or other terminal electron acceptors can be reduce d by accepting electrons on the cathode. Electricity is produced during the electron transfer from the anode to the cathode via the external circ uit. Theoretically, all reductive pollutants can be oxidized using the electrode as e lectron acceptor in the anode chamber, while oxidative pollutants can be reduced using the electrode as electron donor in the cathode chamber. For example, it was reported that diesel, ethanol, 1,2-dichloroethane, pyridine, and other contaminants were successfully oxidized at the anode, 45-47 while chlorinated solvents, perchlorate, chromium, and ur anium, can be reduced at the cathode.11-13, 125 Compared with other treatment systems, MFC provide s such a flexible platform that it allows both oxidation and reductio n reactions occurring at the same time in one reactor. Traditional wastewater treatment processes such as suspended or attached systems are effective in removing major organic compounds a s measured as COD or BOD, but studies showed that they have challenges to effecti vely remove trace organic compounds (TOrCs), because such contaminants generally presen t in a very low level (eg, ng/L) and difficult for biodegradation. TOrCs are common poll utants exist in domestic wastewater resulting from human activities including the use o f personal care product, pharmaceuticals, and pesticides. TOrCs in wastewate r are in a diverse range of physical and chemical properties, such as molecular weight, hydrophobicity and charge state. In traditional wastewater treatment plant, part of the TOrCs were removed by primary

PAGE 108

92 settling, chemical precipitation, volatilization, a nd activated sludge sorption, while most of the TOrCs were degraded by microorganisms.211 MFCs as new environmental technology could be an ef fective approach for TOrCs removal from wastewater. On the one hand, MFC s can provide both oxidation and reduction reactions in one system with electrochemi cal reactions; on the other hand, anaerobic anode and aerobic cathode grow different bacteria communities with different removal functions. The electrode may also adsorb tr ace organics for microbial degradation. The unique environments provided by MF C systems may enhance the removal of different kinds of TOrCs in wastewater, and therefore in this study, singlechamber MFCs and double-chamber MFCs were used to c reate combined and separated oxidation-reduction conditions. Thirty four differe nt types of TOrCs were used as target compounds that were spiked into reactors to charact erize the feasibility and removal mechanisms of trace organic removal in MFCs. 6.3 Materials and methods 6.3.1 MFC construction and operation Single-chamber MFC (SMFC) reactors were built by on e polycarbonate cubeshaped chamber with the empty volume of 110 mL. Dou ble-chamber MFC (DMFC) reactors were constructed by two polycarbonate cube -shaped chambers separated by a cation exchange membrane (CEM, ASTOM) with the empt y volume of either anode or cathode chamber of 120 mL. For all the reactors, pl ain carbon cloth (projected surface area 38 cm2, Fuel Cell Earth) without water-proof were used as anode, and water-proof carbon cloth (projected surface area 38 cm2, Fuel Cell Earth) with 4 PTFE diffusion layers, one carbon base and one catalyst layer (0.5 mg Pt/cm2) were used as air-

PAGE 109

93 cathode.159 The MFCs were inoculated with anaerobic sludge obt ained from Littleton Englewood Wastewater Treatment Plant (Englewood, CO ). During the acclimatization time, all the chambers were fed with medium solutio n containing (per liter) 1.24g CH3COONa, 0.031g NH4Cl, 0.013g KCl, 0.2452g NaH2PO4H2O, 0.4576g Na2HPO4, 1.25mL mineral solution, and 0.5mL vitamin solution A 167 ohm external resistor was connected to MFCs to show the potential of power pr oduction. When the voltages dropped to lower than 40 mV, fresh medium was refil led. MFCs were put in a shaker to accelerate mass transfer inside the reactors. All t he tests were conducted in room temperature. 6.3.2 Experimental Procedures When repeated stable performances were generated, t he reactors were ready for TOrCs removal tests. During the tests, there were f our control experiments. The 34 target TOrCs in the study are listed in Table 6.1. Firstly medium solution were filled in MFCs for one complete cycle to conduct baseline controls ; secondly, medium solution spiked with 34 TOrCs was used to feed the reactors for one complete cycle to allow TOrCs adsorption inside MFCs, which were adsorption contr ols; thirdly, TOrCs removal in closed circuit under 167 ohm external resistor were carried out; finally, TOrCs removal were conducted without external resistor to run ope n circuit controls. In the last two steps, 25 mL initial influent samples and 25 mL fin al effluent samples were collected for TOrCs analysis. To minimize the contamination of sa mples, all the referred vessels are glassware cleaned by methanol and Milli-Q water thr ee times. MFCs and glassware were wrapped by aluminum foil to avoid light. Samples we re collected in amber glass bottles

PAGE 110

94 and preserved by adding 1g/L NaN3 to prevent microbial degradation. Sample bottles were stored at 4C before analysis. Table 6.2 34 TOrCs Detected by LC-MS/MS in ESI(+) and ESI(-) Methods and Selected Physicochemical Properties. Name Class Structure Molar Mass (g/mol) Biodegradability probabilitya logDa Charge (pH=7) DL (ng/L) ESI (+) method Acetaminophen analgesic 151.2 LM = 1.0015 AL = -0.1124 0.2685 neutral 10 Amitriptyline antidepressant 277.4 LM = 0.5196 AL = -1.4265 4.9487 positive 25 Atenolol beta-blocker 266.3 LM = 1.33 AL = -0.1861 -0.0259 positive 10 Atrazine herbicide 215.7 LM =0.0045 AL = -0.5787 2.8175 neutral 5 Benzophenone preservative 182.2 LM = 0.9238 AL = -0.0737 3.1471 neutral 250 Caffeine stimulant 194.2 LM = 0.6551 AL = 0.5019 0.1564 neutral 10 Carbamazepine antiepileptic 236.3 LM = 0.6351 AL = -0.0744 2.2484 neutral 25 Cimetidine histamine 252.3 LM = 0.6821 AL = 0.334 0.574 neutral 50 DEET insecticide 191.3 LM = 0.9213 AL = -0.5924 2.2579 neutral 25 Diazepam benzodiazepine tranquilizer 284.7 LM = 0.7678 AL = -0.8789 2.6997 neutral 5

PAGE 111

95 Name Class Structure Molar Mass (g/mol) Biodegradability probabilitya logDa Charge (pH=7) DL (ng/L) Dilantin anticonvulsant 252.3 LM = 0.699 AL = -0.0161 2.1613 neutral 25 Diphenhydramine antihistamine 255.4 LM = 0.3295 AL = -1.088 3.1063 positive 25 Fluoxetine antidepressant 309.3 LM = 0.4937 AL = 0.5562 4.6483 positive 5 Hydrocodone analgesic 299.4 LM = 0.5412 AL = -1.4424 2.1644 positive 25 Meprobamate anti-anxiety agent 218.3 LM = 0.6188 AL = 0.4467 0.9826 positive 10 Norfluoxetine fluoxetine metabolite 295.3 NA NA neutral 25 Oxybenzone ingredient in sunscreens 228.2 LM =1.0215 AL = 0.0781 3.5248 neutral 100 Primidone anticonvulsant 218.3 LM = 1.0081 AL = -1.0665 0.726 neutral 25 Sulfamethoxazole anti-infective 253.3 LM = 0.4479 AL = -0.2907 0.484 negative 5 Trimethoprim anti-infective 290.3 LM = 0.5922 AL = 0.1677 0.7289 Positive/n eutral 10 TCEP flame retardant 250.2 NA NA neutral 10 Table 6.1 (conÂ’t)

PAGE 112

96 Name Class Structure Molar Mass (g/mol) Biodegradability probabilitya logDa Charge (pH=7) DL (ng/L) TCPP ame retardant 327.6 LM = 0.5714 AL = 0.6608 2.8865 neutral 25 TDCPP ame retardant 430.9 LM = 0.188 AL = 0.9335 3.6485 neutral 50 ESI (-) method Bisphenol A plasticizer 228.3 LM = 0.6866 AL = -0.2593 3.643 neutral 50 Diclofenac antiinammatory agent 296.1 LM = 0.1353 AL = -0.8493 4.0157 negative 10 Gemfibrozil antilipidemic 250.3 LM = 0.7584 AL = -0.2444 4.7672 negative 10 Ibuprofen antiinammatory agent 206.3 LM = 0.8314 AL = 0.0334 3.7971 negative 100 Ketoprofen antiinammatory agent 254.3 LM = 0.8888 AL = -0.1707 3.0001 negative 50 Methylparaben personal care product 152.2 LM = 0.9651 AL = 0.6274 1.9967 neutral 10 Naproxen antiinammatory agent 230.3 LM = 0.8972 AL = 0.3878 3.1029 negative 10 Nonylphenol surfactant 220.4 LM = 0.9215 AL = 0.1979 5.9889 neutral 50 Propylparaben preservative 180.2 LM = 0.9517 AL = 0.6793 2.9789 neutral 5 Triclocarban antimicrobial 315.6 LM = 0.05 AL = -1.0387 4.9018 neutral 10 Table 6.1 (conÂ’t)

PAGE 113

97 Name Class Structure Molar Mass (g/mol) Biodegradability probabilitya logDa Charge (pH=7) DL (ng/L) Triclosan antimicrobial 289.5 LM = 0.3102 AL = -0.6846 4.663 neutral 10 a EPI SuiteTM v4.11; LM=Linear biodegradability prob ability; AL=Anaerobic linear biodegradibility proba bility; A probability 0.5 indicates biodegrades fast; A probability < 0. 5 indicates does not biodegrade fast. NA=not availa ble. 6.3.3 Analysis Voltages of MFCs were recorded every 66 sec by a da ta acquisition system (Keithley Instrument, OH). Dissolved oxygen (DO) an d pH were measured by a DO meter and pH meter (HACH Co., CO). TOrCs were analy zied by liquid chromatography/tandem mass spectrometry (LC-MS/MS) assembled with a highperformance liquid chromatography (HPLC, 1200 Serie s, Agilent Technologies Inc., CA) and tandem mass spectrometer (3200 Q Trap, Applied Biosystems, CA) using electrospray ionization (ESI) in both positive and negative modes. To alleviate the issue of matrix suppression, isotope dilution for each ta rget compound was used to correct matrix effects.212 Biodegradibility probability of each compound was estimated using EPI SuiteTM v4.11. Linear biodegradability probability model was used to calculate the possibilities of TOrCs biodegradation in SMFC and c athode-chamber of DMFC, wihle anaerobic linear biodegradibility probability model was applied for anode-chamber of DMFC. 6.4 Results and Discussion 6.4.1 Performance of Single-chamber and Two-chamber MFCs Voltages of the MFC reactors began to increase slow ly after inoculation. When repeated voltages were generated filling with only sodium acetate, SMFC and DMFC can reach voltages as high as 23410 mV and 10015 mV a t 167 ohm, respectively. There Table 6.1 (conÂ’t)

PAGE 114

98 were not many differences on MFCs performance when filling the reactors with sodium acetate spiked with TOrCs. The concentration of eac h spiked TOrC was around 500 ng/L which may not have harm on bacteria in a short time running. As shown in Figure 6.1, power production from SMFC filled with sodium aceta te spiked with (147 mW/m2) and without TOrCs (144 mW/m2) were comparable, the similar results were also ob tained in DMFC when spiked with (47.2 mW/m2) and without TOrCs (51.4 mW/m2). Figure 6.1 Polarization Curve and Power Density Cu rve Obtained by Linear Sweep Voltammetry (LSV) in Single-chamber (A) and Two-cha mber Reactors (B) Filled by Sodium Acetate Spiked with TOrCs (Dark Blue) and wi thout TOrCs (Orange). The Scan Rate of the Polarization Was 0.1 mV/s. 0 10 20 30 40 50 60 0 100 200 300 400 500 600 700 800 00.050.10.150.20.25Power density (mW/m2) Voltage (mV)Current density (A/m2) 0 20 40 60 80 100 120 140 160 0 100 200 300 400 500 600 700 800 00.30.60.91.2Power density (mW/m2) Voltage (mV) Current density (A/m2)A B

PAGE 115

99 The cycle times of running SMFC and DMFC at 167 ohm were 3 days and 4 days, respectively. At the end of each cycle, the voltage of SMFC reduced to lower than 10 mV and 97.4% of TOC was removed (Figure 6.2), while th e voltage of DMFC was merely lower than 40 mV and TOC removal efficiencies in an ode-chamber and cathode-chamber were 37.9% and 97.9%. Because of the water loss thr ough air cathode, the final volumes of SMFC and cathode-chamber of DMFC were about 90 m L and 95 mL, whereas the volume of anode-chamber was the same as initial vol ume of 120 mL. DO concentration in SMFC decreased from the beginning (5.11.1 mg/L) to 0.340.12 mg/L and then increased to 4.20.28 mg/L, which was the similar p erformance in cathode-chamber of DMFC that DO went down from 5.50.5 mg/L to 0.370. 01 mg/L and finalized at 4.80.1 mg/L. However, DO in anode-chamber of DMFC reduced from 4.80.5 mg/L to 0.60.4 mg/L and increased a little to 1.20.1 mg/L when the cycles were ended. It was inevitable to introduce oxygen to the reactor when refilling new medium, the lost water from the reactor would further make room for intrud ed air which led to DO increase, although aerobic bacteria existed in the reactor ca n consume certain amount of oxygen. Thus, there could be many aerobic bacteria grown in SMFC and cathode-chamber of DMFC because of the high DO, while in anode-chamber of DMFC there should be mainly anaerobic bacteria or electrochemically acti ve bacteria (EAB). Since aerobic bacteria grow faster than anaerobic bacteria, this can also explain why TOC removal efficiencies were much higher in SMFC and cathode-c hamber of DMFC than that in anode-chamber of DMFC. Under open circuit condition TOC removal efficiencies in SMFC and cathode-chamber of DMFC were still compara ble with that at 167 ohm, but less TOC was removed in anode-chamber of DMFC (Figu re 6.2). Because of the

PAGE 116

100 different types of bacteria, there was no much infl uence on aerobic bacteria under open circuit condition or not, but EAB had no drive to c onduct electrons under open circuit condition, so less organic carbon was consumed. pH (7.550.28) in SMFC was relatively stable during the cycle even low buffer was used, h owever, pH reduced below 6 in anode-chamber of DMFC and increased to 8 in cathode -chamber of DMFC since different reduction-oxidation reactions happened in the two chambers. Figure 6.2 Total Organic Carbon (TOC) Removal in S ingle-chamber and Twochamber Reactors Filled by Sodium Acetate Spiked wi th TOrCs under 167 ohm and Open Circuit. 6.4.2 TOrCs removal in Single-chamber and Two-chamb er MFCs Figure 6.3 presents the results of TOrCs removal in both SMFC and DMFC. It is clearly shows that most of the TOrCs can be removed in different levels in MFC reactors. Both nonylphenol and benzophenone got poor results in each sample, so there were no data available for these two compounds. As discusse d in the previous section, there were similar pH and DO conditions in SMFC and cathode-ch amber of DMFC. Thus, the performance of TOrCs removal in SMFC and cathode-ch amber of DMFC were similar 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% singletwo_anodetwo_cathode Removal Efficiency 167ohm open

PAGE 117

101 except the situations of TCEP and Diclofenac (Figur e 6.3A). Comparatively, the removal of TOrCs in anode-chamber of DMFC which had a condi tion of lower DO and pH was much worse (Figure 6.3B). Because of oxygen intrusi on through air-cathode, there was a layer of aerobic biofilm grown on the liquid-facing side of the cathode in both SMFC and cathode-chamber of DMFC. Some suspended aerobic bac teria inevitably existed in the catholyte. In anode-chamber of DMFC, electrochemica lly active bacteria (EAB) were the main microorganisms which were grown on the brush a node with some minor anaerobic bacteria suspended in the anolyte. Totally differen t bacteria communities were a crucial reason to explain the different performances on TOr Cs removal in SMFC and DMFC. There were not many differences of TOrCs removal in SMFC and cathode-chamber of DMFC at 167 ohm or open circuit, but the overall tr end of TOrCs removal in anodechamber of DMFC at open circuit was lower than that at 167 ohm. This result was also related to the different bacteria communities that aerobic bacteria still consumed comparable amount of TOrCs even at open circuit, bu t EAB would degrade less compounds since no pressure on electron transfer at open circuit. TOrCs were divided into three classifications based on charge. Usually, the removals of positively charged TOrCs were higher th an that of negatively charged TOrCs because positively charged compounds can be easily attracted and adhered on the negatively charged bacteria for degradation due to electrostatic attraction,213 but negatively charged TOrCs would be repulsed by the b acteria which was not beneficial for degradation, for example, the removal efficiencies of positively charged TOrCs were 7898% in SMFC and cathode-chamber of DMFC and 44-86% in anode-chamber of DMFC, whereas only 5-29% of negatively charged TOrCs were removed in SMFC and cathode-

PAGE 118

102 chamber of DMFC and 4-37% in anode-chamber of DMFC. Two exceptions were positively charged meprobamate and negatively charg ed sulfamethoxazole with possible reasons discussed later. The removal of uncharged T OrCs had a wide range of 17-98% in SMFC and cathode-chamber of DMFC, which was compara ble in anode-chamber of DMFC (3-97%), but SMFC and cathode-chamber of DMFC removed more type of compounds. The removal performances of all TOrCs were not cons istent with biodegradability probabilities estimated by EPI SuiteTM v4.11 BioWin software. The possible reasons were as follows: first, the limitation and accuracy of the models used in the analysis may bring any error; second, DO and pH were always chan ged and not stable during the process in MFC reactors. DO would affect the activi ties of bacteria and pH would change the species percentage between charged species and uncharged species; third, octanolwater partition coefficient (D=kow) for each compound would also affect removal behavior. logD greater than 2 is considered hydroph obic. Hydrophobic TOrCs were likely to be adsorbed on the reactors wall or catio n exchange membrane in DMFC, for example, the removal efficiencies of diclofenac (lo gD=4.0157) and TCPP (logD=2.8865) were -20% and -13% in open circuit, respectively. I t is possible that when the adsorption became over saturated, desorption would happen and increased the concentration in the effluent, fourth, sodium acetate was used as the ma in substrate to feed the reactors which led to a simple bacteria community. The limited bac teria may not effectively degrade each compound especially in a mixture with many oth er interference factors, such as DO, pH, charge and kow and so on.

PAGE 119

103 Figure 6.3 TOrCs Removal in Single-chamber Reactor s and Cathode-chamber of Two-chamber Reactors (A) and Anode-chamber of Two-c hamber Reactors (B) Filled by Sodium Acetate Spiked with TOrCs under 16 7 ohm and Open Circuit. TOrCs are Divided into Three Categories Based on Ch arge. Biodegradibility Probability is Indicated in Parentheses after the N ame of Each Compound. -40% -20% 0% 20% 40% 60% 80% 100%Atenolol(1.3) Meprobamate(0.6) Hydrocodone(0.5) Amitriptyline(0.5) Fluoxetine(0.5) Diphenhydramine(0.3) Naproxen(0.9) Ketoprofen(0.9) Ibuprofen(0.8) Gemfibrozil(0.8) Sulfamethoxazole(0.4) Diclofenac(0.1) Oxybenzone(1.0) Primidone(1.0) Acetaminophen(1.0) Methylparaben(1.0) Propylparaben(1.0) DEET(0.9) Diazepam(0.8) Dilantin(0.7) Bisphenol A(0.7) Cimetidine(0.7) Caffeine(0.7) Carbamazepine(0.6) Trimethoprim(0.6) TCPP(0.6) Triclosan(0.3) TDCPP(0.2) Triclocarban(0) Atrazine(0) TCEP Norfluoxetine Removal Efficiency Single 167 Single open Cathode 167 Cathode open -40% -20% 0% 20% 40% 60% 80% 100%Fluoxetine(0.6) Meprobamate(0.4) Atenolol( 0.2) Diphenhydramine( 1.1) Amitriptyline( 1.4) Hydrocodone( 1.4) Naproxen(0.4) Ibuprofen(0) Ketoprofen(-0.2) Gemfibrozil( 0.2) Sulfamethoxazole( 0.3) Diclofenac( 0.8) TDCPP(0.9) Propylparaben(0.7) TCPP(0.7) Methylparaben(0.6) Caffeine(0.5) Cimetidine(0.3) Trimethoprim(0.2) Oxybenzone(0.1) Dilantin(0) Carbamazepine(-0.1) Acetaminophen(-0.1) Bisphenol A( 0.3) Atrazine( 0.6) DEET( 0.6) Triclosan( 0.7) Diazepam( 0.9) Triclocarban(-1.0) Primidone(-1.1) Norfluoxetine TCEP Removal Efficiency Anode 167 Anode openA B positive negative neutral positive negative neutral

PAGE 120

104 Therefore, a variety of TOrCs can be effectively re moved in MFCs, but the removal mechanisms were complex referring to many f actors. In the future study, NaN3 will be added in the reactor to kill bacteria and r ule out the possibility of biodegradation in order to better understand TOrCs removal mechani sms.

PAGE 121

105 REFERENCES 1. B.E. Logan and K. Rabaey, Conversion of wastes into bioelectricity and chemicals by using microbial electrochemical techno logies. Science, 2012. 337(6095): 686-690. 2. F. Harnisch and U. Schrder, From MFC to MXC: chemical and biological cathodes and their potential for microbial bioelect rochemical systems. Chem. Soc. Rev., 2010. 39(11): 4433-4448. 3. R.A. Rozendal, H.V.M. Hamelers, K. Rabaey, J. Ke ller and C.J.N. Buisman, Towards practical implementation of bioelectrochemi cal wastewater treatment. Trends Biotechnol., 2008. 26(8): 450-459. 4. C.I. Torres, A.K. Marcus, H.-S. Lee, P. Paramesw aran, R. Krajmalnik-Brown and B.E. Rittmann, A kinetic perspective on extracellular electron tra nsfer by anoderespiting bacteria. FEMS Microbiol. Rev., 2010. 34(1): 3-17. 5. H. Liu and B.E. Logan, Electricity generation using an air-cathode single chamber microbial fuel cell in the presence and abs ence of a proton exchange membrane. Environ. Sci. Technol., 2004. 38(14): 4040-4046. 6. Z. Ren, T.E. Ward and J.M. Regan, Electricity production from cellulose in a microbial fuel cell using a defined binary culture. Environ. Sci. Technol., 2007. 41(13): 4781-4786. 7. J.J. Fornero, M. Rosenbaum and L.T. Angenent, Electric power generation from municipal, food, and animal wastewaters using micro bial fuel cells. Electroanalysis, 2010. 22(7-8): 832-843. 8. S. Cheng, D. Xing, D.F. Call and B.E. Logan, Direct biological conversion of electrical current into methane by electromethanoge nesis. Environ. Sci. Technol., 2009. 43(10): 3953-3958. 9. B.E. Logan, D. Call, S. Cheng, H.V.M. Hamelers, T.H.J.A. Sleutels, A.W. Jeremiasse and R.A. Rozendal, Microbial electrolysis cells for high yield hydrogen gas production from organic matter. Environ. Sci. Technol., 2008. 42(23): 8630–8640. 10. H. Liu, H. Hu, J. Chignell and Y. Fan, Microbial electrolysis: novel technology for hydrogen production from biomass. Biofuels, 2010. 1(1): 129-142. 11. K.B. Gregory and D.R. Lovley, Remediation and recovery of uranium from contaminated subsurface environments with electrode s. Environ. Sci. Technol., 2009. 39(22): 8943–8947.

PAGE 122

106 12. C.S. Butler, P. Clauwaert, S.J. Green, W. Verst raete and R. Nerenberg, Bioelectrochemical perchlorate reduction in a micro bial fuel cell. Environ. Sci. Technol., 2010. 44(12): 4685-4691. 13. F. Aulenta, A. Canosa, M. Majone, S. Panero, P. Reale and S. Rossetti, Trichloroethene dechlorination and H2 evolution are alternative biological pathways of electric charge utilization by a dechlo rinating culture in a bioelectrochemical system. Environ. Sci. Technol., 2008. 42(16): 6185–6190. 14. K. Rabaey and R.A. Rozendal, Microbial electrosynthesis--revisiting the electrical route for microbial production. Nat. Rev. Microbiol., 2010. 8(10): 706716. 15. D.R. Lovley and K.P. Nevin, A shift in the current: New applications and concepts for microbe-electrode electron exchange. Curr. Opin. Biotechnol., 2011. 22(3): 441-448. 16. X. Cao, X. Huang, P. Liang, K. Xiao, Y. Zhou, X Zhang and B.E. Logan, A new method for water desalination using microbial desal ination cells. Environ. Sci. Technol., 2009. 43(18): 7148-7152. 17. M. Mehanna, P.D. Kiely, D.F. Call and B.E. Loga n, Microbial electrodialysis cell for simultaneous water desalination and hydrogen ga s production. Environ. Sci. Technol., 2010. 44(24): 9578–9583. 18. K.S. Jacobson, D.M. Drew and Z. He, Use of a liter-scale microbial desalination cell as a platform to study bioelectrochemical desa lination with salt solution or artificial seawater. Environ. Sci. Technol., 2011. 45(10): 4652-4657. 19. H. Luo, P.E. Jenkins and Z. Ren, Concurrent desalination and hydrogen generation using microbial electrolysis and desalin ation cells. Environ. Sci. Technol., 2011. 45(1): 340-344. 20. H. Luo, P. Xu, T.M. Roane, P.E. Jenkins and Z. Ren, Microbial desalination cells for improved performance in wastewater treatment, e lectricity production, and desalination. Bioresour. Technol., 2012. 105: 60-66. 21. M.C. Potter, Electrical effects accompanying the decomposition o f organic compounds. Proc. R. Soc. London Ser. B, 1911. 84(571): 260-276. 22. U. Schrder, Discover the possibilities: microbial bioelectroche mical systems and the revival of a 100-year–old discovery. J. Solid State Electrochem, 2011. 15(78): 1481–1486.

PAGE 123

107 23. A.P. Borole, G. Reguera, B. Ringeisen, Z.-W. Wa ng, Y. Feng and B.H. Kim, Electroactive biolms: current status and future re search needs. Energy Environ. Sci., 2011. 4(12): 4813-4834. 24. U. Schrder, Microbial fuel cells and microbial electrochemistry : into the next century! ChemSusChem, 2012. 5(6): 959-961. 25. T.H.J.A. Sleutels, A.T. Heijne, C.J.N. Buisman and H.V.M. Hamelers, Bioelectrochemical systems: an outlook for practica l applications. ChemSusChem, 2012. 5(6): 1012-1019. 26. H.V.M. Hamelers, A.T. Heijne, T.H.J.A. Sleutels A.W. Jeremiasse, D.P.B.T.B. Strik and C.J.N. Buisman, New applications and performance of bioelectrochemical systems. Appl. Microbiol. Biotechnol., 2010. 85(6): 1673– 1685. 27. D. Pant, G.V. Bogaert, L. Diels and K. Vanbroek hoven, A review of the substrates used in microbial fuel cells (MFCs) for sustainable energy production. Bioresour. Technol., 2010. 101(6): 1533-1543. 28. J. Wei, P. Liang and X. Huang, Recent progress in electrodes for microbial fuel cells. Bioresour. Technol., 2011. 102(20): 9335-9344. 29. B.E. Logan, Exoelectrogenic bacteria that power microbial fuel cells. Nat. Rev. Microbiol., 2009. 7(5): 375-381. 30. D.R. Lovley, Bug juice: harvesting electricity with microorganis ms. Nat. Rev. Microbiol., 2006. 4(7): 497-508. 31. B.-H. Kim, H.-J. Kim, M.-S. Hyun and D.-H. Park Direct electrode reaction of Fe(III)-reduing bacterium, Shewanella putrefaciens. J. Microbiol. Biotechnol. 1999. 9(2): 127-131. 32. K. Tanaka, C.A. Vega and R. Tamamushi, Mediating effects of ferricchelatecompound in microbialfuelcells Bioelectrochem. Bioenerg., 1983. 11(2-3): 135-143. 33. K. Rabaey, P. Clauwaert, P. Aelterman and W. Ve rstraete, Tubular microbial fuel cells for efficient electricity generation. Environ. Sci. Technol., 2005. 39(20): 8077-8082. 34. Z. He, S.D. Minteer and L.T. Angenent, Electricity generation from atificial wastewater using an upflow microbial fuel cell. Environ. Sci. Technol., 2005. 39(14): 5262-5267.

PAGE 124

108 35. Z. He, N. Wagner, S.D. Minteer and L.T. Angenen t, An upflow microbial fuel cell with an interior cathode: assessment of the interna l resistance by impedance spectroscoty. Environ. Sci. Technol., 2006. 40(17): 5212-5217. 36. Y. Feng, H. Lee, X. Wang, Y. Liu and W. He, Continuous electricity generation by a graphite granule baffled air-cathode microbial fuel cell. Bioresour. Technol., 2010. 101(2): 632-638. 37. P. Aelterman, K. Rabaey, H.T. Pham, N. Boon and W. Verstraete, Continuous electricity generation at high voltages and current s using stacked microbial fuel cells. Environ. Sci. Technol., 2006. 40(10): 3388-3394. 38. Y. Zhang and I. Angelidaki, A simple and rapid method for monitoring dissolved oxygen in water with a submersible microbial fuel c ell (SBMFC). Biosens. Bioelectron., 2012. 38(1): 189-194. 39. M.E. Nielsen, C.E. Reimers and H.A.S. III, Enhanced power from chambered benthic microbial fuel cells. Environ. Sci. Technol., 2007. 41(22): 7895–7900. 40. L.M. Tender, S.A. Gray, E. Groveman, D.A. Lowy, P. Kauffman, J. Melhado, R.C. Tyce, D. Flynn, R. Petrecca and J. Dobarro, The first demonstration of a microbial fuel cell as a viable power supply: power ing a meteorological buoy. J. Power Sources, 2008. 179(2): 571-575. 41. Y. Gong, S.E. Radachowsky, M. Wolf, M.E. Nielse n, P.R. Girguis and C.E. Reimers, Benthic microbial fuel cell as direct power source for an acoustic modem and seawater oxygen/temperature sensor system Environ. Sci. Technol., 2011. 45(11): 5047-5053. 42. Y. Zhang and I. Angelidaki, Self-stacked submersible microbial fuel cell (SSMFC) for improved remote power generation from l ake sediments. Biosens. Bioelectron., 2012. 35(1): 265-270. 43. J.M. Morris, S. Jin, B. Crimi and A. Pruden, Microbial fuel cell in enhancing anaerobic biodegradation of diesel. Chem. Eng. J., 2009. 146(2): 161-167. 44. J.R. Kim, S.H. Jung, J.M. Regan and B.E. Logan, Electricity generation and microbial community analysis of alcohol powered mic robial fuel cells. Bioresour. Technol., 2007. 98(13): 2568-2577. 45. H. Pham, N. Boon, M. Marzorati and W. Verstraet e, Enhanced removal of 1,2dichloroethane by anodophilic microbial consortia. Water Res., 2009. 43(11): 2936-2946. 46. C. Zhang, M. Li, G. Liu, H. Luo and R. Zhang, Pyridine degradation in the microbial fuel cells. J. Hazard Mater., 2009. 172(1): 465-471.

PAGE 125

109 47. H. Luo, G. Liu, R. Zhang and S. Jin, Phenol degradation in microbial fuel cells. Chem.Eng. J., 2009. 147(2-3): 259–264. 48. T. Zhang, S.M. Gannon, K.P. Nevin, A.E. Franks and D.R. Lovley, Stimulating the anaerobic degradation of aromatic hydrocarbons in contaminated sediments by providing an electrode as the electron acceptor. Environ. Microbiol., 2010. 12(4): 1011-1020. 49. R. Thorne, H. Hu, K. Schneider, P. Bombelli, A. Fisher, L.M. Peter, A. Dent and P.J. Cameron, Porous ceramic anode materials for photo-microbial fuel cells. J. Mater. Chem., 2011. 21(44): 18055-18060. 50. S. Malik, E. Drott, P. Grisdela, J. Lee, C. Lee D.A. Lowy, S. Gray and L.M. Tender, A self-assembling self-repairing microbial photoele ctrochemical solar cell. Energy Environ. Sci., 2009. 2(3): 292-298. 51. Y.K. Cho, T.J. Donohue, I. Tejedor, M.A. Anders on, K.D. McMahon and D.R. Noguera, Development of a solar-powered microbial fuel cell. J. Appl. Microbiology, 2008. 104(3): 640-650. 52. D.P.B.T.B. Strik, H.V.M. Hamelers and C.J.N. Bu isman, Solar energy powered microbial fuel cell with a reversible bioelectrode. Environ. Sci. Technol., 2010. 44(1): 532-537. 53. M. Rosenbaum, U. Schrder and F. Scholz, In situ electrooxidation of photobiological hydrogen in a photobioelectrochemic al fuel cell based on Rhodobacter sphaeroides. Environ. Sci. Technol., 2005. 39(16): 6328-6333. 54. Y. Zou, J. Pisciotta, R.B. Billmyre and I.V. Ba skakov, Photosynthetic microbial fuel cells with positive light response. Biotechnol Bioeng., 2009. 104(5): 939-946. 55. T. Yagishita, S. Sawayama, K.-i. Tsukahara and T. Ogi, Effects of intensity of incidentlight and concentrations of Synechococcus s p. and 2-hydroxy-1,4naphthoquinone on the current output of photosynthe tic electrochemical cell. Sol. Energy, 1997. 61(5): 347–353. 56. F. Qian, G. Wang and Y. Li, Solar-driven microbial photoelectrochemical cells with a nanowire phtotcathode. Nano Lett., 2010. 10(11): 4686-4691. 57. H. Deng, Z. Chen and F. Zhao, Energy from plants and microorganisms: progress in plant–microbial fuel cells. ChemsusChem, 2012. 5(6): 1006-1011. 58. Z. He, J. Kan, F. Mansfeld, L.T. Angenent and K .H. Nealson, Self-sustained phototrophic microbial fuel cells based on the syne rgistic cooperation between

PAGE 126

110 photosynthetic microorganisms and heterotrophic bac teria. Environ. Sci. Technol., 2009. 43(5): 1648–1654. 59. D.P.B.T.B. Strik, H. Terlouw, H.V.M. Hamelers a nd C.J.N. Buisman, Renewable sustainable biocatalyzed electricity production in a photosynthetic algal microbial fuel cell (PAMFC). Appl. Microbiol. Biotechnol., 2008. 81(4): 659-668. 60. B. Erable, L. Etcheverry and A. Bergel, From microbial fuel cell (MFC) to microbial electrochemical snorkel (MES): maximizing chemical oxygen demand (COD) removal from wastewater. Biofouling, 2011. 27(3): 319-326. 61. S. Cheng, B.A. Dempsey and B.E. Logan, Electricity generation from synthetic acid-mine drainage (AMD) water using fuel cell tech nologies. Environ. Sci. Technol., 2007. 41(23): 8149-8153. 62. L. Xiao, E.B. Young, J.A. Berges and Z. He, Integrated photo-bioelectrochemical system for contaminants removal and bioenergy produ ction. Environ. Sci. Technol., 2012. 46(20): 11459–11466. 63. F. Zhang, K.S. Brastad and Z. He, Integrating forward osmosis into microbial fuel cells for wastewater treatment, water extration ad bioelectricity generation. Environ. Sci. Technol., 2011. 45(15): 6690-6696. 64. Y. Kim and B.E. Logan, Microbial reverse electrodialysis cells for synergi stically enhanced power production. Environ. Sci. Technol., 2011. 45(13): 5834–5839. 65. R.D. Cusick, Y. Kim and B.E. Logan, Energy capture from thermolytic solutions in microbial reverse-electrodialysis cells. Science, 2012. 335(6075): 1474-1477. 66. H. Liu, S. Grot and B.E. Logan, Electrochemically assisted microbial production of hydrogen from acetate. Environ. Sci. Technol., 2005. 39(11): 4317-4320. 67. R.A. Rozendal, E. Leone, J. Keller and K. Rabae y, Efficient hydrogen peroxide generation from organic matter in a bioelectrochemi cal system. Electrochem. Commun., 2009. 11(9): 1752-1755. 68. K. Rabaey, S. Butzer, S. Brown, J. Keller and R .A. Rozendal, High current generation coupled to caustic production using a la mellar bioelectrochemical system. Environ. Sci. Technol., 2010. 44(11): 4315-4321. 69. J. Ditzig, H. Liu and B.E. Logan, Production of hydrogen from domestic wastewater using a bioelectrochemically assisted mi crobial reactor (BEAMR). Int. J. Hydrogen Energy, 2007. 32(13): 2296-2304.

PAGE 127

111 70. K.-J. Chae, M.-J. Choi, K.-Y. Kim, F.F. Ajayi, I.-S. Chang and I.S. Kim, A Solarpowered microbial electrolysis cell with a platinum catalyst-free cathode to produce hydrogen. Environ. Sci. Technol., 2009. 43(24): 9525-9530. 71. Y. Kim and B.E. Logan, Hydrogen production from inexhaustible supplies of fresh and salt water using microbial reverse-electr odialysis electrolysis cells. Proc. Natl. Acad. Sci., 2011. 108(39): 16176-16181. 72. R.D. Cusick and B.E. Logan, Phosphate recovery as struvite within a single chamber microbial electrolysis cell. Bioresour. Technol., 2012. 107: 110-115. 73. Y. Zhang and I. Angelidaki, Innovative self-powered submersible microbial electrolysis cell (SMEC) for biohydrogen production from anaerobic reactors. Water Res., 2012. 46(8): 2727-2736. 74. K. Rabaey, P. Girguis and L.K. Nielsen, Metabolic and practical considerations on microbial electrosynthesis. Curr. Opin. Biotechnol., 2011. 22(3): 371-377. 75. K.J.J. Steinbusch, H.V.M. Hamelers, J.D. Schaap C. Kampman and C.J.N. Buisman, Bioelectrochemical ethanol production through media ted acetate reduction by mixed cultures. Environ. Sci. Technol., 2010. 44(1): 513-517. 76. K.P. Nevin, T.L. Woodard, A.E. Franks, Z.M. Sum mers and D.R. Lovley, Microbial electrosynthesis: feeding microbes electr icity to convert carbon dioxide and water to multicarbon extracellular organic comp ounds. mBio, 2010. 1(2): e00103-10. 77. K.P. Nevin, S.A. Hensley, A.E. Franks, Z.M. Sum mers, J. Ou, T.L. Woodard, O.L. Snoeyenbos-West and D.R. Lovley, Electrosynthesis of organic compounds from carbon dioxide Is catalyzed by a diversity of acetogenic microorganisms. Appl. Environ. Microbiol., 2011. 77(9): 2882–2886. 78. Y. Gong, A. Ebrahim, A.M. Feist, M. Embree, T. Zhang, D. Lovley and K. Zengler, Sulde-driven microbial electrosynthesis. Environ. Sci. Technol., 2013. 47(1): 568-573. 79. X. Wang, Y. Feng, J. Liu, H. Lee, C. Li, N. Li and N. Ren, Sequestration of CO2 discharged from anode by algal cathode in microbial carbon capture cells (MCCs). Biosens. Bioelectron., 2010. 25(12): 2639-2643. 80. Y. Kim and B.E. Logan, Simultaneous removal of organic matter and salt ion s from saline wastewater in bioelectrochemical system s. Desalination, 2013. 308: 115-121. 81. B. Zhang and Z. He, Integrated salinity reduction and water recovery in an osmotic microbial desalination cell. RSC Adv., 2012. 2(8): 3265–3269.

PAGE 128

112 82. Y. Kim and B.E. Logan, Microbial desalination cells for energy production and desalination. Desalination, 2013. 308: 122-130. 83. C. Forrestal, P. Xu, P.E. Jenkins and Z. Ren, Microbial desalination cell with capacitive adsorption for ion migration control. Bioresour. Technol., 2012. 120: 332–336. 84. A. Morel, K. Zuo, X. Xia, J. Wei, X. Luo, P. Li ang and X. Huang, Microbial desalination cells packed with ion-exchange resin t o enhance water desalination rate. Bioresour. Technol., 2012. 118: 43-48. 85. S. Chen, G. Liu, R. Zhang, B. Qin and Y. Luo, Development of the microbial electrolysis desalination and chemical-production c ell for desalination as well as acid and alkali productions. Environ. Sci. Technol., 2012. 46(4): 2467-2472. 86. C. Forrestal, P. Xu and Z. Ren, Sustainable desalination using a microbial capacitive desalination cell. Energy Environ. Sci., 2012. 5(5): 7161-7167. 87. L. Yuan, X. Yang, P. Liang, L. Wang, Z.-H. Huan g, J. Wei and X. Huang, Capacitive deionization coupled with microbial fuel cells to desalinate lowconcentration salt water. Bioresour. Technol., 2012. 110: 735-738. 88. X. Chen, X. Xia, P. Liang, X. Cao, H. Sun and X Huang, Stacked microbial desalinaiton cells to enhance water desalination ef ficiency. Environ. Sci. Technol., 2011. 45(6): 2465–2470. 89. Y. Qu, Y. Feng, X. Wang, J. Liu, J. Lv, W. He a nd B.E. Logan, Simultaneous water desalination and electricity generation in a microbial desalination cell with electrolyte recirculation for pH control. Bioresour. Technol., 2012. 106: 89–94. 90. H.S. Park, B.H. Kim, H.S. Kim, H.J. Kim, G. Kim M. Kim, I.S. Chang, Y.K. Park and H.I. Chang, A novel electrochemically active and Fe(III)-reduci ng bacterium phylogenetically related to Clostridium b utyricum isolated from a microbial fuel cell. Anaerobe, 2001. 7(6): 297-306. 91. C.I. Torres, R. Krajmalnik-Brown, P. Parameswar an, A.k. Marcus, G. Wanger, Y.A. Gorby and B.E. Rittmann, Selecting anode-respiring bacteria based on anode potential: phylogeneric, electrochemical, and microscopic characterization. Environ. Sci. Technol., 2009. 43(24): 9519-9524. 92. D.R. Lovley, Live wires: direct extracellular electron exchange for bioenergy and the bioremediation of energy-related contamination. Energy Environ. Sci., 2011. 4(12): 4896-4906.

PAGE 129

113 93. Z.M. Summers, H.E. Fogarty, C. Leang, A.E. Fran ks, N.S. Malvankar and D.R. Lovley, Direct exchange of electrons within aggregates of a n evolved syntrophic coculture of anaerobic bacteria. Science, 2010. 330(6009): 1413-1415. 94. H.v. Canstein, J. Ogawa, S. Shimizu and J.R. Ll oyd, Secretion of flavins by Shewanella species and their role in extracellular electron transfer. Appl. Environ. Microbiol., 2008. 74(3): 615-623. 95. Y.A. Gorby, S. Yanina, J.S. McLean, K.M. Rosso, D. Moyles, A. Dohnalkova, T.J. Beveridge, I.S. Chang, B.H. Kim, K.S. Kim, D.E Culley, S.B. Reed, M.F. Romine, D.A. Saffarini, E.A. Hill, L. Shi, D.A. Eli as, D.W. Kennedy, G. Pinchuk, K. Watanabe, S.i. Ishii, B. Logan, K.H. Nealson and J.K. Fredrickson, Electrically conductive bacterial nanowires produce d by Shewanella oneidensis strain MR-1 and other microorganisms. PNAS, 2006. 103(30): 11358-11363. 96. E. Marsili, D.B. Baron, I.D. Shikhare, D. Cours olle, J.A. Gralnick and D.R. Bond, Shewanella secretes flavins that mediate extracellu lar electron transfer. PNAS, 2008. 105(10): 3968-3973. 97. K. Rabaey, N. Boon, M. Hfte and W. Verstraete, Microbial phenazine production enhances electron transfer in biofuel ce lls. Environ. Sci. Technol., 2005. 39(9): 3401–3408. 98. D.H. Park and J.G. Zeikus, Electricity generation in microbial fuel cells usin g neutral red as an electronophore. Appl. Environ. Microbiol., 2000. 66(4): 12921297. 99. C.E. Milliken and H.D. May, Sustained generation of electricity by the sporeforming, Gram-positive, Desulfitobacterium hafniens e strain DCB2. Appl. Microbiol. Biotechnol., 2007. 73(5): 1180-1189. 100. C.F. Thurston, H.P. Bennetto, G.M. Delaney, J. R. Mason, S.D. Roller and J.L. Stirling, Glucose metabolism in a microbial fuel cell stoichi ometry of product formation in a thionine-mediated Proteus vulgaris f uel cell and its relation to coulombic yields. J. Gen. Microbiol., 1985. 131(6): 1391-1401. 101. K. Scott and C. Murano, Microbial fuel cels utilising carbohydrates. J. Chem. Technol. Biotechnol., 2007. 82(1): 92-100. 102. S.B. Velasquez-orta, T.P. Curtis and B.E. Loga n, Energy from algae using microbial fuel cells. Biotechnol. Bioeng., 2009. 103(6): 1068-1076. 103. P. Kuntke, K.M. Smiech, H. Bruning, G. Zeeman, M. saakes, T.H.J.A. Sleutels, H.V.M. Hamelers and C.J.N. Buisman, Ammonium Recovery and energy production from urine by a microbial fuel cell. Water Res., 2012. 46(8): 2627– 2636.

PAGE 130

114 104. K. Rabaey, K.V.D. Sompel, L. Maignien, N. Boon P. Aelterman, P. Clauwaert, L.D. Schamphelaire, H.T. Pham, J. Vermeulen, M. Ver haege, P. Lens and W. Verstraete, Microbial fuel cells for sulfide removal. Environ. Sci. Technol., 2006. 40(17): 5218-5224. 105. Z. Ren, L.M. Steinberg and J.M. Regan, Electricity production and microbial biofilm characterization in cellulose-fed microbial fuel cells. Water Sci. Technol., 2008. 58(3): 623-628. 106. P. Parameswaran, C.I. Torres, H.-S. Lee, R. Kr ajmalnik-Brown and B.E. Rittmann, Syntrophic interactions among anode respiring bacte ria (ARB) and non-ARB in a biofilm anode: electron balances. Biotechnol. Bioeng., 2009. 103(3): 513-523. 107. S. Freguia, K. Rabaey, Z. Yuan and J. Keller, Syntrophic processes drive the conversion of glucose in microbial fuel cell anodes Environ. Sci. Technol., 2008. 42(21): 7937–7943. 108. B.E. Logan, B. Hamelers, R. Rozendal, U. Schr der, J. Keller, S. Freguia, P. Aelterman, W. Verstraete and K. Rabaey, Microbial fuel cells: methodology and technology. Environ. Sci. Technol., 2006. 40(17): 5181–5192. 109. H. Liu, R. Ramnarayanan and B.E. Logan, Production of electricity during wastewater treatment using a single chamber microbi al fuel cell. Environ. Sci. Technol., 2004. 38(7): 2281-2285. 110. C.E. Reimers, L.M. Tender, S. Fertig and W. Wa ng, Harvesting energy from the marine sediment-water interface. Environ. Sci. Technol., 2001. 35(1): 192-195. 111. D.P.B.T.B. Strik, R.A. Timmers, M. Helder, K.J .J. Steinbusch, H.V.M. Hamelers and C.J.N. Buisman, Microbial solar cells: applying photosynthetic and electrochemically active organisms. Trends Biotechnol., 2011. 29(1): 41-49. 112. H. Liu, S. Cheng and B.E. Logan, Production of electricity from acetate or butyrate using a single-chamber microbial fuel cell Environ. Sci. Technol., 2005. 39(2): 658-662. 113. K. Rabaey, G. Lissens, S.D. Siciliano and W. V erstraete, A microbial fuel cell capable of converting glucose to electricity at hig h rate and efficiency. Biotechnol. Lett., 2003. 25(18): 1531-1535. 114. X. Zhang, S. Cheng, X. Huang and B.E. Logan, The use of nylon and glass ber lter separators with different pore sizes in air-c athode single-chamber microbial fuel cells. Energy Environ. Sci., 2010. 3(5): 659-664.

PAGE 131

115 115. Z. Ren, Chapter 19: the principle and applications of bioel ectrochemical systems, in Biofuel Technol., V.K. Gupta and M.G. Tuohy, Edi tors. 2013, Springer. 501527. 116. S.T. Oh, J.R. Kim, G.C. Premier, T.H. Lee, C. Kim and W.T. Sloan, Sustainable wastewater treatment: how might microbial fuel cell s contribute. Biotechnol. Adv., 2010. 28(6): 871-881. 117. M. Huggins, P. Fallgren and Z. Ren, Energy and performance comparison of microbial fuel cell and conventional aeration treat ing of wastewater. J. Microbial. Biochem. Technol. In process, 2013. 118. Y. Fan, S.-K. Han and H. Liu, Improved performance of CEA microbial fuel cells with increased reactor size. Energy Environ. Sci., 2012. 5(8): 8273-8280. 119. C. Donovan, A. Dewan, H. Peng, D. Heo and H. B eyenal, Power management system for a 2.5 W remote sensor powered by a sedim ent microbial fuel cell. J. Power Sources, 2011. 196(3): 1171-1177. 120. L.M. Tender, C.E. Reimers, H.A.S. III, D.E. Ho lmes, D.R. Bond, D.A. Lowy, K. Pilobello, S.J. Fertig and D.R. Lovley, Harnessing microbially generated power on the seafloor. Nat. Biotechnol., 2002. 20(8): 821-825. 121. J.M. Morris and S. Jin, Feasibility of using microbial fuel cell technology for bioremediation of hydrocarbons in groundwater. J. Environ. Sci. Health A Tox. Hazard Subst. Environ. Eng. 2008. 43(1): 18-23. 122. Y. Yuan, S. Zhou and L. Zhuang, A new approach to in situ sediment remediation based on air-cathode microbial fuel cells. J. Soils Sediments, 2010. 10(7): 14271433. 123. D. Huang, S. Zhou, Q. Chen, B. Zhao, Y. Yuan a nd L. Zhuang, Enhanced anaerobic degradation of organic pollutants in a so il microbial fuel cell. Chem. Eng. J., 2011. 172(2-3): 647-653. 124. X. Wang, Z. Cai, Q. Zhou, Z. Zhang and C. Chen Bioelectrochemical stimulation of petroleum hydrocarbon degradation in saline soil using u-tube microbial fuel cells. Biotechnol. Bioeng., 2012. 109(2): 426-433. 125. G. Wang, L. Huang and Y. Zhang, Cathodic reduction of hexavalent chromium [Cr(VI)] coupled with electricity generation in mic robial fuel cells. Biotechnol. Lett., 2008. 30(11): 1959-1966. 126. K.H. Williams, K.P. Nevin, A. Franks, A. Engle rt, P.E. Long and D.R. Lovley, An Electrode-based approach for monitoring in sit u microbial activity during subsurface bioremediation. Environ. Sci. Technol., 2010. 44(1): 47-54.

PAGE 132

116 127. D.P.B.T.B. Strik, H.V.M.H. (Bert), J.F.H. Snel and C.J.N. Buisman, Green electricity production with living plants and bacte ria in a fuel cell. Int. J. Energy Res., 2008. 32(9): 870-876. 128. L.D. Schamphelaire, L.V.D. Bossche, H.S. Dang, M. Hofte, N. Boon, K. Rabaey and W. Verstraete, Microbial fuel cells generating electricity from rh izodeposits of rice plants. Environ. Sci. Technol., 2008. 42(8): 3053-3058. 129. M. Helder, D.P.B.T.B. Strik, H.V.M. Hamelers, A.J. Kuhn, C. Blok and C.J.N. Buisman, Concurrent bio-electricity and biomass production i n three plantmicrobial fuel cells using spartina anglica, Arundi nella anomala and Arundo donax. Bioresour. Technol., 2010. 101(10): 3541-3547. 130. R.A. Timmers, D.P.B.T.B. Strik, H.V.M. Hameler s and C.J.N. Buisman, Longterm performance of a plant microbial fuel cell wit h Spartina anglica. Appl. Microbiol. Biotechnol., 2010. 86(3): 973–981. 131. X. Cao, X. Huang, N. Boon, P. Liang and M. Fan Electricity generation by an enriched phototrophic consortium in a microbial fue l cell. Electrochem. Commun., 2008. 10(9): 1392–1395. 132. L.D. Schamphelaire and W. Verstraete, Revival of the biological sunlight-tobiogas energy conversion system. Biotechnol Bioeng., 2009. 103(2): 296-304. 133. R.A. Rozendal, H.V.M. Hamelers, G.J.W. Euverin k, S.J. Metz and C.J.N. Buismana, Principle and perspectives of hydrogen production t hrough biocatalyzed electrolysis. Int. J. Hydrogen Energy, 2006. 31(12): 1632–1640. 134. H.-S. Lee, W.F.J. Vermaas and B.E. Rittmann, Biological hydrogen production: prospects and challenges. Trends Biotechnol., 2010. 28(5): 262–271. 135. H. Hu, Y. Fan and H. Liu, Hydrogen production using single-chamber membrane-free microbial electrolysis cells. Water Res., 2008. 42(15): 4172– 4178. 136. M. Sun, G. Sheng, L. Zhang, C. Xia, Z. Mu, X. Liu, H. Wang, H. Yu, R. Qi, T. Yu and M. Yang, An MEC-MFC-coupled system for biohydrogen productio n from acetate. Environ. Sci. Technol., 2008. 42(21): 8095-8100. 137. C.W. Marshall, D.E. Ross, E.B. Fichot, R.S. No rman and H.D. May, Electrosynthesis of commodity chemicals by an autot rophic microbial community. Appl. Environ. Microbiol., 2012. 78(23): 8412-8420.

PAGE 133

117 138. Y. Kim and B.E. Logan, Series assembly of microbial desalination cells containing stacked electrodialysis cells for partia l or complete seawater desalination. Environ. Sci. Technol., 2011. 45(13): 5840-5845. 139. H. Luo, P. Xu, P.E. Jenkins and Z. Ren, Ionic composition and transport mechanisms in microbial desalination cells. J. Membrane Sci., 2012. 409-410: 16-23. 140. H. Luo, P. Xu and Z. Ren, Long-term performance and characterization of microbial desalination cells in treating domestic w astewater. Bioresour. Technol., 2012. 120: 187-193. 141. R.D. Cusick, B. Bryan, D.S. Parker, M.D. Merri ll, M. Mehanna, P.D. Kiely, G. Liu and B.E. Logan, Performance of a pilot-scale continuous flow microb ial electrolysis cell fed winery wastewater. Appl. Microbiol. Biotechnol., 2011. 89(6): 2053-2063. 142. J. Keller and K. Rabaey. Experiences from MFC pilot plant operation: how to get the technology market ready? in Microbial Fuel Cells First International Symposium. 2008. State College, PA, USA. 143. B.E. Logan, Scaling up microbial fuel cells and other bioelectr ochemical systems. Appl. Microbiol. Biotechnol., 2010. 85(6): 1665–1671. 144. H. Wang, J.-D. Park and Z. Ren, Active energy harvesting from microbial fuel cells at the maximum power point without using resi stors. Environ. Sci. Technol., 2012. 46(9): 5247-5252. 145. J.-D. Park and Z. Ren, Hysteresis controller based maximum power point tra cking energy harvesting system for microbial fuel cells. J. Power Sources, 2012. 205(9): 151-156. 146. J.M. Foley, R.A. Rozendal, C.K. Hertle, P.A. L ant and K. Rabaey, Life cycle assessment of high-rate anaerobic treatment, microb ial fuel cells, and microbial electrolysis cells. Environ. Sci. Technol., 2010. 44(9): 3629-3637. 147. D. Pant, A. Singh, G.V. Bogaert, Y.A. Gallego, L. Diels and K. Vanbroekhoven, An introduction to the life cycle assessment (LCA) of bioelectrochemical systems (BES) for sustainable energy and product generation :Relevance and key aspects. Renewable Sustainable Energy Rev., 2011. 15(2): 1305-1313. 148. X. Wang, S. Cheng, Y. Feng, M.D. Merrill, T. S aito and B.E. Logan, Use of carbon mesh anodes and the effect of different pret reatment methods on power production in microbial fuel cells. Environ. Sci. Technol., 2009. 43(17): 68706874.

PAGE 134

118 149. B.E. Logan, Microbial fuel cell. 2008, New York: John Wiley&Sons. 150. Y. Fan, H. Hu and H. Liu, Enhanced coulombic efficiency and power density of air-cathode microbial fuel cells with an improved c ell configuration. J. Power Sources, 2007. 171(2): 348-354. 151. S. Cheng and B.E. Logan, Ammonia treatment of carbon cloth anodes to enhance power generation of microbial fuel cells. Electrochem. Commun., 2007. 9(3): 492-496. 152. B. Logan, S. Cheng, V. Watson and G. Estadt, Graphite fiber brush anodes for increased power production in air-cathode microbial fuel cells. Environ. Sci. Technol., 2007. 41(9): 3341-3346. 153. R.M. Association, Scrap tire markets in the united states 2009: United States. 105. 154. R. Siddique and T.R. Naik, Properties of concrete containing scrap-tire rubber an overview. Waste Manag., 2004. 24(6): 563-569. 155. Z. Tang, M.A. Butkus and Y.F. Xie, Crumb rubber filtration: a potential technology for ballast water treatment. Mar. Environ. Res., 2006. 61(4): 410-423. 156. Y. Xie, Filter media: crumb rubber for wastewater filtratio n. Filtr. Separat., 2007. 44(5): 30-32. 157. J. Park, T.G. Ellis and M. Lally. Evaluation of tire derived rubber particles for biofiltration media. in WEF Proc. 2006: Water Environment Foundations. 158. Z. Ren, R.P. Ramasamy, S.R. Cloud-Owen, H. Yan M.M. Mench and J.M. Regan, Time-course correlation of biofilm properties and e lectrochemical performance in single-chamber microbial fuel cells. Bioresour. Technol., 2011. 102(1): 416-421. 159. S. Cheng, H. Liu and B.E. Logan, Increased performance of single-chamber microbial fuel cells using an improved cathode stru cture. Electrochem. Commun., 2006. 8(3): 489-494. 160. Y. Zuo, S. Cheng, D. Call and B.E. Logan, Tubular membrane cathodes for scalable power generation in microbial fuel cells. Environ. Sci. Technol., 2007. 41(9): 3347-3353. 161. P. Xu, J.E. Drewes, D. Heil and G. Wang, Treatment of brackish produced water using carbon aerogel-based capacitive deionization technology. Water Res., 2008. 42(10-11): 2605-2617.

PAGE 135

119 162. E.P. Barrett, L.G. Joyner and P.P. Halenda, The determination of pore volume and area distributionws in porous substances. I. Comput ations from nitrogen isotherms. J. Am. Chem. Soc., 1951. 73(1): 373-380. 163. F. Zhang, M.D. Merrill, J.C. Tokash, T. Saito, S. Cheng, M.A. Hickner and B.E. Logan, Mesh optimization for microbial fuel cell cathodes constructed around stainless steel mesh current collectors. J. Power Sources, 2011. 196(3): 10971102. 164. N. Chang, Economic and policy instrument analyses in support on the scrap tire recycling program in Taiwan. J. Environ. Manage. 2008. 86(3): 435-450. 165. Y. Fan, E. Sharbrough and H. Liu, Quantification of the internal resistance distribution of microbial fuel cells. Environ. Sci. Technol., 2008. 42(21): 81018107. 166. H. Rismani-Yazdi, S.M. Carver, A.D. Christy an d O.H. Tuovinen, Cathodic limitations in microbial fuel cells: an overview. J. Power Sources, 2008. 180(2): 683-694. 167. F. Zhao, F. Harnisch, U. Schrder, F. Scholz, P. Bogdanoff and I. Herrmann, Challenges and constraints of using oxygen cathodes in microbial fuel cells. Environ. Sci. Technol., 2006. 40(17): 5193-5199. 168. Q. Deng, X. Li, J. Zuo, A. Ling and B.E. Logan Power generation using an activated carbon fiber felt cathode in an upflow mi crobial fuel cell. J. Power Sources, 2010. 195(4): 1130-1135. 169. S. Cheng and B.E. Logan, Increasing power generation for scaling up singlechamber air cathode microbial fuel cells. Bioresour. Technol., 2011. 102(6): 4468-4473. 170. P. Liang, H. Wang, X. Xia, X. Huang, Y. Mo, X. Cao and M. Fan, Carbon nanotube powders as electrode modifier to enhance t he activity of anodic biofilm in microbial fuel cells. Biosens. Bioelectron., 2011. 26(6): 3000-3004. 171. X. Xie, L. Hu, M. Pasta, G.F. Wells, D. Kong, C.S. Criddle and Y. Cui, Threedimensional carbon nanotube-textile anode for highperformance microbial fuel cells. Nano Lett., 2011. 11(1): 291-296. 172. J.L. Lamp, J.S. Guest, S. Naha, K.A. Radavish, N.G. Love, M.W. Ellis and I.K. Puri, Flame synthesis of carbon nanostructures on stainle ss steel anodes for use in microbial fuel cells. J. Power Sources, 2011. 196(14): 5829-5834.

PAGE 136

120 173. H. Tsai, C. Wu, C. Lee and E.P. Shih, Microbial fuel cell performance of multiwall carbon nanotubes on carbon cloth as elect rodes. J. Power Sources, 2009. 194(1): 199-205. 174. A.C. Dillon, T. Gennett, K.M. Jones, E.L. Alle man, P.A. Parilla and M.J. Heben, A simple and complete purification of single-walled carbon nanotube. Adv. Mater., 1999. 11(16): 1354-1358. 175. A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Peti t, J. Robert, C. Xu, Y.H. Lee, S.G. Kim, A.G. Rinzler, D.T. Colbert, G.E. Scuseria, D. Tomanek, J.E. Fischer and R.E. Smalley, Crystalline ropes of metallic carbon nanotubes. Science, 1996. 273(5274): 483-487. 176. C. Chen, M. Chen, H. Yu, S. Lu and C. Chen, Rapid and homogeneous dispersion of Pt catalyst particles on multi-walled carbon nan otubes by temperaturecontrolled microwave polyol method. Jpn. J. Appl. Phys., 2008. 47: 2324-2329. 177. Z. Ren, H. Yan, W. Wang, M.M. Mench and J.M. R egan, Charaterization of microbial fuel cells at microbially and electrochem ically meaningful time scales. Environ. Sci. Technol., 2011. 45(6): 2435-2441. 178. F. Zhang, T. Saito, S. Cheng, M.A. Hickner and B.E. Logan, Microbial fuel cell cathodes with poly(dimethylsiloxane) diffusion laye rs constructed around stainless steel mesh current collectors. Environ. Sci. Technol., 2010. 44(4): 14901495. 179. J.R. Kim, J.-Y. Kim, S.-B. Han, K.-W. Park, G. D. Saratale and S.-E. Oh, Application of co-naphthalocyanine (CoNPc) as alter native cathode catalyst and support structure for microbial fuel cells. Bioresour. Technol., 2011. 102(1): 342347. 180. S. Cheng, P. Kiely and B.E. Logan, Pre-acclimation of a wastewater inoculum to cellulose in an aqueous-cathode MEC improves power generation in air-cathode MFCs. Bioresour. Technol., 2011. 102(1): 367-371. 181. P.D. Kiely, G. Rader, J.M. Regan and B.E. Loga n, Long-term cathode performance and the microbial communities that deve lop in microbial fuel cells fed different fermentation endproducts. Bioresour. Technol., 2011. 102(1): 361366. 182. K. Chung, I. Fujiku and S. Okabe, Effect of formation of biofilms and chemical scale on the cathode electrode on the performance o f a continuous two-chamber microbial fuel cell. Bioresour. Technol., 2011. 102(1): 355-360.

PAGE 137

121 183. O. Lefebvre, W.K. Ooi, Z. Tang, M. Abdullah-Al -Mamun, D.H.C. Chua and H.Y. Ng, Optimization of a Pt-free cathode suitable for prac tical applications. Bioresour. Technol., 2009. 100(20): 4907-4910. 184. S. Yang, B. Jia and H. Liu, Effects of the Pt loading side and cathode-biofilm on the performance of a membrane-less and single-chamb er microbial fuel cell. Bioresour. Technol., 2009. 100(3): 1197-1202. 185. R.C. Wagner, D.F. Call and B.E. Logan, Optimal set anode potentials vary in bioelectrochemical systems. Environ. Sci. Technol., 2010. 44(16): 6036-6041. 186. R.P. Pinto, B. Srinivasan, S.R. Guiot and B. T artakovsky, The efffect of real-time external resistance optimization on microbial fuel cell performance. Water Res., 2011. 45(4): 1571-1578. 187. J.C. Biffinger, J. Pietron, O. Bretschger, L.J Nadeau, G.R. Johnson, C.C. Williams, K.H. Nealson and B.R. Ringeisen, The influence of acidity on microbial fuel cells containing shewanella oneidensis. Biosens. Bioelectron., 2008. 24(4): 906-911. 188. Z. He, Y. Huang, A.K. Manohar and F. Mansfeld, Effect of electrolyte pH on the rate of the anodic and cathodic reactions in an air -cathode microbial fuel cell. Bioelectrochemistry, 2008. 74(1): 78-82. 189. S. Cheng, D. Xing and B.E. logan, Electricity generation of single-chamber microbial fuel cells at low temperatures. Biosens. Bioelectron., 2011. 26(5): 1913-1917. 190. L. Woodward, M. Perrier, B. Srinivasan, R.P. P into and B. Tartakovsky, Comparison of real-time methods for maximizing powe r output in microbial fuel cells. AIChE J., 2010. 56(10): 2742-2750. 191. P. Liang, W. Wu, J. Wei, L. Yuan, X. Xia and X Huang, Alternate charging and discharging of capacitor to enhance the electron pr oduction of bioelectrochemical systems. Environ. Sci. Technol., 2011. 45(15): 6647–6653. 192. C. Donovan, A. Dewan, D. Heo and H. Beyenal, Batteryless, wireless sensor powered by a sediment microbial fuel cell. Environ. Sci. Technol., 2008. 42(22): 8591–8596. 193. Y. Kim, M.C. Hatzell, A.J. Hutchinson and B.E. Logan, Capturing power at higher voltages from arrays of microbial fuel cells without voltage resersal. Energy Environ. Sci., 2011. 4(11): 4662-4667.

PAGE 138

122 194. H. Wang, Z. Wu, A. Plaseied, P. Jenkins, L. Si mpson, C. Engtrakul and Z. Ren, Carbon nanotube modified air-cathodes for electrici ty production in microbial fuel cells. J. Power Sources, 2011. 196(18): 7465-7469. 195. F. Zhang, K.S. Jacobson, P. Torres and Z. He, Effects of anolyte recirculation rates and catholytes on electricity generation in a litre-scale upflow microbial fuel cell. Energy Environ. Sci., 2010. 3(9): 1347-1352. 196. J.D. Park and Z. Ren. Efficient energy harvester for microbial fuel cells using DC/DC converters. in IEEE Energy Conversion Congre. Expos. 2011. Phoenix, Arizona. 197. A. Meehan, H. Gao and Z. Lewandowski, Energy harbesting with microbial fuel cell and power management system. IEEE Trans. Power Electron., 2011. 26(1): 176-181. 198. F. Zhang, L. Tian and Z. He, Powering a wireless temperature sensor using sediment microbial fuel cells with vertical arrange ment of electrodes. J. Power Sources, 2011. 196(22): 9568-9573. 199. J.-D. Park and Z. Ren, High efficiency energy harvesting from microbial fu el cells using a synchronous boost converter. J. Power Sources, 2012. 208: 322-327. 200. P.K. Wu, J.C. Biffinger, L.A. Fitzgerald and B .R. Ringeisen, A low power DC/DC booster circuit designed for microbial fuel cells. Process Biochem., 2012. 47(11): 1620–1626. 201. J.-D. Park and Z. Ren, Hysteresis-controller-based energy harvesting schem e for microbial fuel cells with parallel operation capabi lity. IEEE Trans. Energy Conversion, 2012. 27(3): 715 724. 202. A. Dewan, H. Beyenal and Z. Lewandowski, Intermittent energy harvesting improves the performance of microbial fuel cells. Environ. Sci. Technol., 2009. 43(12): 4600-4605. 203. E.J. Gardel, M.E. nielsen, P.T. Grisdela and P Girguis, Duty cycling influences current generation in multi-anode environmental mic robial fuel cells. Environ. Sci. Technol., 2012. 46(9): 5222-5229. 204. F. Grondin, M. Perrier and B. Tartakovsky, Microbial fuel cell operation with intermittent connection of the electrical load. J. Power Sources, 2012. 208(15): 18-23. 205. P.S. Bonanni, G.D. Schrott, L. Robuschi and J. P. Busalmen, Charge accumulation and electron transfer kinetics in Geob acter Sulfurreducens biofilms. Energy Environ. Sci., 2012. 5(3): 6188-6195.

PAGE 139

123 206. V.J. Watson and B.E. Logan, Analysis of polarization methods for elimination of power overshoot in microbial fuel cells. Electrochem. Commun., 2011. 13(1): 5456. 207. Y. Feng, X. Wang, B.E. Logan and H. Lee, Brewery wastewater treatment using air-cathode microbial fuel cells. Appl. Microbiol. Biotechnol., 2008. 78(5): 873880. 208. J.R. Kim, Y. Zuo, J.M. Regan and B.E. Logan, Analysis of ammonia loss mechnisms in microbial fuel cells treating animal w astewater. Biotechnol. Bioeng., 2008. 99(5): 1120-1127. 209. L. Huang and B.E. Logan, Electricity generation and treatment of paper recyc ling wastewater using a microbial fuel cell. Appl. Biochem. Biotechnol., 2008. 80(2): 349-355. 210. B. Min and B.E. Logan, Continuous electricity generation from domestic wastewater and organic substrates in a flat plate m icrobial fuel cell. Environ. Sci. Technol., 2004. 38(21): 5809-5814. 211. Z. Liu, Y. Kanjo and S. Mizutani, Removal mechanisms for endocrine disrupting compounds (EDCs) in wastewater treatment-physical m eans, biodegradation, and chemical advanced oxidation: a review. Sci. Total Environ., 2009. 407(2): 731748. 212. B.J. Vanderford and S.A. Snyder, Analysis of pharmaceuticals in water by isotope dilution liquid chromatography/tandem mass spectrom etry. Environ. Sci. Technol., 2006. 40(23): 7312-7320. 213. W.P. Johnson and B.E. Logan, Enhanced transport of bacteria in porous media by sediment-phase and aqueous-phase natural organic matter. Water Res., 1996. 30(4): 923-931.

PAGE 140

124 Heming Wang, Ph.D Candidate Environmental and Sustainability Engineering Department of Civil Engineering University of Colorado Denver Ph: (720)-325-9456, Email: heming.wang@ucdenver.edu Education Ph.D Candidate, Civil Engineering, University of Co lorado Denver, 2009-present M.E., Environmental Science and Engineering, Harbin Institute of Technology (China), 2008 B.S., Environmental Science, Harbin Institute of Te chnology (China), 2006 Appointments Research Assistant, Civil Engineering, University o f Colorado Denver, 2009-present Refereed Publications Heming Wang Zhiyong Ren and Pei Xu. Removal mechanisms of tra ce organic compounds in microbial fuel cells. In preparation. Heming Wang and Zhiyong Ren. The X factor: a review of bioelec trochemical system as a platform technology. in review. Heming Wang Zhiyong Ren, and Jae-Do Park. Power electronic co nverters for microbial fuel cell energy extraction: Effects of inductance, duty ratio, and switching frequency. J. Power Sources, 2012, 220, 89–94. Heming Wang Jae-Do Park, and Zhiyong Ren. Active energy harve sting from microbial fuel cells at the maximum power point without using resi stors. Environ. Sci. Technol., 2012, 46, 5247–5252. Heming Wang Zhuangchun Wu, Atousa Plaseied, Peter Jenkins, Li n Simpson, Chaiwat Engtrakul, and Zhiyong Ren. Carbon nanotube modifie d air-cathodes for electricity production in microbial fuel cells. J. Power Sources, 2011, 196, 7465-7469. Heming Wang Matthew Davidson, Yi Zuo, and Zhiyong Ren. Recycl ed tire crumb rubber anodes for sustainable power production in microbia l fuel cells. J. Power Sources, 2011, 196, 5863-5866. Xin Wang, Yujie Feng, Heming Wang Youpeng Qu, Yanling Yu, Nanqi Ren, Nan Li, Elle Wang, He Lee and Bruce E. Logan. Bioaugmentation fo r electricity generation from corn stover biomass using microbial fuel cells. Environ. Sci. T echnol., 2009, 43, 6088-6093. Xin Wang, Yujie Feng, Nanqi Ren, Heming Wang He Lee, Nan Li and Qingliang Zhao. Accelerated start-up of two-chambered microbial fue l cells, effect of anodic positive poised potential. Electrochimica Acta, 2009, 54: 1109-1114

PAGE 141

125 Conference Proceedings/Presentations Heming Wang Zhiyong Ren and Pei Xu. Removal mechanisms of tra ce organic compounds in microbial fuel cells. 10th Annual RMSAWWA/RMWEA Stu dent Conference, 2013, Golden, CO, US. Heming Wang Pei Xu, and Zhiyong Ren. Removal mechanisms of tr ace organic compounds in microbial fuel cells. AEESP Education and Research Conference, 2013, Golden, CO, US Heming Wang Jae-Do Park, and Zhiyong Ren. Maximizing microbia l fuel cell energy output by Active harvesting. North American-International Soc iety of Microbial Electrochemical Technologies (NA-ISMET), 2012, Ithaca, NY, US. Heming Wang Zhuangchun Wu, Atousa Plaseied, Peter Jenkins, Li n Simpson, Chaiwat Engtrakul, and Zhiyong Ren. Exploring new electrode materials for sustainable electricity production in microbial fuel cells. 242th American Chemical Society (ACS) National Meeting, 2011, Denver, CO, US. Heming Wang Zhuangchun Wu, Atousa Plaseied, Peter Jenkins, Li n Simpson, Chaiwat Engtrakul, and Zhiyong Ren. Carbon nanofiber modifi ed air cathodes for improving electricity production in microbial fuel cells. 239th American Chemical Society (ACS) National Meeting, 2010, San Francisco, CA, US. Xin Wang, Yujie Feng, Heming Wang Yanling Yu, Youpeng Qu and He Lee. Electricity generation from corn stover in microbial fuel cells 2nd Symposium on Microbial Fuel Cells, 2009, Beijing, China. Heming Wang Yujie Feng, Xin Wang, He Lee and Nanqi Ren. Treat ment of exploded-cornstover washing wastewater using air-cathode single chamber bottle-microbial fuel cell. The 6th IWA World Congress and Exhibition, 2008. Vienna, Au stria. Yujie Feng, Heming Wang Xin Wang and He Li. Electricity generation from c orn stover using air-cathode single chamber microbial fuel cell. The 1st International MFC Symposium, 2008, University Park, PA, US. Honors and Awards Gold Prize of Excellent Master Thesis, 2008 Excellent Graduate, 2008 Special-class Scholarship, 2007~2008 First-class Scholarship, 2006~2007 Outstanding Award in “Environment and Development” Writing Contest, 2006 Excellent Graduate of Commission of Science Technol ogy and Industry for National Defense, 2006 Excellent Graduate, 2006 HIT Special Scholarship, 2005 First Prize of the National Investigation on the Wa stewater Treatment, 2004 The Prize of Number One, 2004 Triple-A Student, 2004~2007 People’s Scholarship, 2002~2006