Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Studies on direct methanol, formic acid and related fuel cells in conjunction with the electrochemical reduction of carbon dioxide
(USC Thesis Other)
Studies on direct methanol, formic acid and related fuel cells in conjunction with the electrochemical reduction of carbon dioxide
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
STUDIES ON DIRECT METHANOL, FORMIC ACID AND RELATED FUEL
CELLS IN CONJUNCTION WITH THE ELECTROCHEMICAL REDUCTION OF
CARBON DIOXIDE
by
Federico Andres Viva
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
May 2009
Copyright 2009 Federico Andres Viva
ii
Epigraph
Science is a collective enterprise which embraces many cultures and spans
the generations, in every age and sometimes in the most unlikely places there
are those who wish with a great deal of passion to understand the world.
Who speaks for Earth?
Carl Sagan
iii
Dedication
To those who are
no longer with us.
iv
Acknowledgements
I would like to begin by giving thanks to Prof. Surya Prakash not only for
giving me the rewarding opportunity to work at the Loker Hydrocarbon Institute
but also for providing me a project in such an interesting field. To Prof. George
Olah for his continues work to provide a research environment of the highest
quality and to Dr. Robert Aniszfeld for his openness to listen. To Dr. Roman
Kultyshev, Dr. Alain Goeppert and Dr. Matthew Moran for their help. Also want
to mention Dr. Akihisa Saito, Dr. Kimberly McGrath, Dr. Suresh Palale, Dr.
Chiradeep Panja, Dr. Sujit Chakco, Bo Yang and Frederick Krause. And to all
the members of the group that I met and interacted with over all the years, and
made my time at the Loker a pleasant one.
Going back in time I realize the people and places that made possible,
from the early beginning, my accomplishments, and the list is very long. I will
not mention any particular person, but I have to recognize the school Otto
Krause as one of the pillars of my education and where I spent so many
enjoyable years and from where I got my best friends.
The other big pillar in is the University of Buenos Aires, where I got not
only the other big part of my basic education, but where I also made a lot of
good friends. I want to thank first Dr. Estela Andrade and Dr. Fernando Molina,
who were always so kind to me and from whom I got the love for
electrochemistry. Regarding my classmates with whom I spent countless
v
hours of study, coffee and drinks I will mention Georgina, Yanina and Vanina,
although there are a number of people that I meet during that time with whom I
am still spending good times although now a days is more fun than study,
amid them I want to mention Diego.
Among the many persons that influenced my life I would like to mention
Ing. Jorge Candiani, who taught me a lot about business relations and set a
high bar for what a boss should be. His passing away before his time was a
big loss to all who knew him.
I want to give thanks to one of the people that has influenced my life on
a daily basis; Sensei Takada. I am and will always be in debt to him for
teaching me so many things. And I cannot help but mention Sensei Matsumae
for always being there, and all the good friends that I made, in all the different
places I trained, and proved to be exceptional people.
Now is the time to mention my friends from Otto Krause, Brian, Ruben
and Maxi. Although we sometimes had and still have our disagreements and
our differences, I feel blessed to have such a good friends, and the recent loss
of Maxi made me realize how important he was and how important they are,
and I hope that we can enjoy each other for a long time.
Finally I want to thank the people that made it possible for me to get to
where I am today. My mom and dad, who made it possible for me to have
everything I would need, and in doing so, gave me one of the best educations
that anyone can have. To my sister and to my grandmother for whom there
vi
are no words to describe her, I learned a lot from her and I want to mention
how I admired her resilience. I am compelled to mention that during my time in
the U.S. I had to endure the loss, besides of my good friend Maxi, of my father
and my grandmother. Last but not least, I want to thanks Dolores who has
been with me for the past 8 years and next to me in every decision I had to
make and whom I have to say has spoiled me for the last 8 years, and among
many things contributed to helping me finish my studies. To her again, thanks.
To all, I will be always in debt.
vii
Table of Contents
Epigraph ii
Dedication iii
Acknowledgements iv
List of Tables ix
List of Figures xi
List of Schemes xix
Abstract xx
Chapter 1: Introduction 1
Chapter 1: References 15
Chapter 2: Catalysts for Direct Formic Acid Fuel Cells (DFAFC) 16
2.1 Chapter 2: Introduction 16
2.1.1 Electrochemical oxidation of formic acid on noble
metals. 18
2.1.2 Direct Formic Acid Fuel Cells 26
2.2 Chapter 2: Results and Discussion 33
2.2.1 Catalyst preparation characterization 33
2.2.2 Direct Formic Acid Fuel Cell testing 44
2.3 Chapter 2: Conclusion 57
2.4 Chapter 2: References 58
Chapter 3: Electrochemical reduction of carbon dioxide 64
3.1 Chapter 3: Introduction 64
3.1.1 Mechanism of electrochemical reduction of carbon
dioxide 65
3.1.2 Electrochemical reduction of carbon dioxide to formic
acid 70
3.2 Chapter 3: Results and Discussion 76
3.2.1 Electrochemical measurements using a Sn electrode 76
3.2.2 Electrochemical reduction of carbon dioxide 84
3.3 Chapter 3: Conclusion 88
3.4 Chapter 3: References 90
viii
Chapter 4: Pt supported over Carbon mono-fluoride (CFx) as a
catalyst for the Oxygen Reduction Reaction (ORR) 94
4.1 Chapter 4: Introduction 94
4.1.1 The oxygen reduction reaction (ORR) 96
4.2 Chapter 4: Results and Discussions 100
4.2.1 Catalyst surface analysis and quantification 100
4.2.2 Electrochemical Measurements I. Cyclic Voltammetry 102
4.2.3 Electrochemical measurements II. Electrochemical
Impedance Spectroscopy (EIS) 106
4.2.4 Single cell Fuel Cell testing 111
4.2.4.1 Single cell DMFC polarization tests 111
4.2.4.2 Single cell hydrogen fuel cell polarization tests 119
4.3 Chapter 4: Conclusion 121
4.4 Chapter 4: References 123
Chapter 5: Microbial Fuel Cells 126
5.1 Chapter 5: Introduction 126
5.1.1 Microorganism characteristics 128
5.1.2 Microbial fuel cell assembly considerations 130
5.2 Chapter 5: Results and Discussion 132
5.2.1 Microbial inoculation and scanning electron
microscopy (SEM) images 132
5.2.2 MFC polarization measurements 137
5.3 Chapter 5: Conclusion 148
5.4 Chapter 5: References 150
Chapter 6: Experimental set up and procedures 154
6.1 Chapter 6: Powder Catalyst preparation 154
6.1.1 Catalyst surface analysis 155
6.2 Chapter 6: Electrochemical cells and testing methodologies 155
6.3 Chapter 6: Membrane Electrode Assembly preparations 161
6.4 Chapter 6: Microorganism growth 163
6.5 Chapter 6: Scanning Electron Microscopy (SEM) images for
MFC MEAs 164
6.6 Chapter 6: MCF housing, electrochemical measurements and
inoculation procedures 164
6.6.1 Inoculation Procedures 166
6.7 Chapter 6: References 168
Bibliography 169
ix
List of Tables
Table 1.1 Thermodynamic data and thermodynamic efficiencies
( η
thermo
) for the more commons fuels. The values are
under standard condition at 298 K with oxygen as the
oxidant and water produced in the liquid form. 4
Table 1.2 Types of fuel cells, their electrolyte and characteristics
(From references [2, 3]) 5
Table 1.3 Acid media half cell reaction for the most common
compounds used as fuels in the fuel cell anode and
the half cell reaction for the O
2
used in the cathode.
(Standard potential tabulated as reduction potentials
vs. the standard hydrogen electrode (SHE)). 8
Table 1.4 Half cell reactions in the basic media for most
common fuels at the anode and the half cell reaction
for the reduction of oxygen at the cathode. (Standard
potential tabulated as reduction potentials vs SHE). 10
Table 3.1 Standard redox potential for the two-, four-, six- and
eight-electron reduction of carbon dioxide in water
media, calculated from thermodynamic data.
Reproduced from [9, 10]. 66
Table 3.2 Tafel parameters obtained from the Tafel plots. b
c1
represents the slope for the lower overpotential region
and b
c2
for the high overpotential region. 84
Table 3.3 Faradaic efficiencies (ƒ) for formic acid and total
current densities (j) for the electrocatalysts at the
potential applied for the electrolysis. 87
Table 4.1 Physical properties of Advance Research Chemicals
2010 carbon mono-fluoride and Vulcan XC-72®. 96
x
Table 4.2 Tafel parameters, onset potential, and cathodic peak
potential extracted from Figures 4.4 and 4.5. 105
Table 4.3 Values obtained for the elements of the equivalent
circuit of Figure 4.8 from fitting the experimental
Nyquist Plot for the three different catalysts. The
value of Chi
2
fitting parameter is also included. 109
xi
List of Figures
Figure 1.1 Schematic representation of the fuel cell with the
different components and processes. 6
Figure 1.2 Schematic representation of the alkaline fuel (AFC)
cell with the different components and processes. 9
Figure 1.3 Schematic representation of the carbon dioxide
reutilization cycle. 12
Figure 1.4 Schematic representation of a microbial fuel cell
(MFC) with the different components and processes. 13
Figure 2.1 Cyclic voltammogram for Pt electrode in 0.5 M H
2
SO
4
(dashed line), 0.5 M H
2
SO
4
+ 0.1 M HCOOH (solid
line). Sweep rate 140 mV s
-1
. Reproduced from [12]. 20
Figure 2.2 Cyclic voltammogram for Pt electrode in 1 M H
2
SO
4
(pointed line), 1 M H
2
SO
4
+ 0.1 M CH
3
OH (dashed
line). After holding the potential at 0.4V for 300s (solid
line). Sweep speed 100 mV s
-1
. Reproduced from
[21]. 22
Figure 2.3 Cyclic voltammogram for Pd electrode in 0.5 M H
2
SO
4
(dashed line), 0.5 M H
2
SO
4
+ 0.1 M HCOOH (solid
line). Sweep rate 140 mV s
-1
. Reproduced from [12]. 26
Figure 2.4 Polarization and Power density plot. 27
Figure 2.5 (A) fuel cell polarization plots and (B) power density
curves with 3 M formic acid at 30
◦
C for Pd black
(PdBl) and Pd over carbon. Reproduced from [63]. 31
xii
Figure 2.6 Passive air breathing DFAFC vs. the formic acid
concentration at 30
◦
C: (A) cell potential and (B)
power density curves. Reproduced from [62]. 32
Figure 2.7 Cyclic Voltammogram for Pt powder catalyst in 0.1 M
H
2
SO
4
(solid line) and 0.1 M H
2
SO
4
+ 0.1 M HCOOH
(dashed line). Sweep rate 5 mV s
-1
. 35
Figure 2.8 Cyclic Voltammogram for Pd powder catalyst in 0.1 M
H
2
SO
4
(solid line) and 0.1 M H
2
SO
4
+ 0.1 M HCOOH
(dashed line). Sweep rate 5 mV s
-1
. 37
Figure 2.9 Cyclic Voltammogram for Pd/Au powder catalyst in
0.1 M H
2
SO
4
(solid line) and 0.1 M H
2
SO
4
+ 0.1 M
HCOOH (dashed line). Sweep rate 5 mV s
-1
. 40
Figure 2.10 Cyclic Voltammogram for Pd/Sn powder catalyst in
0.1 M H
2
SO
4
(solid line) and 0.1 M HCOOH + 0.1 M
H
2
SO
4
(dashed line). Sweep rate 5 mV s
-1
. Inset
shows the anodic end for the voltammogram of 0.1 M
HCOOH + 0.1 M H
2
SO
4
expanded. 42
Figure 2.11 Cyclic Voltammogram for Pd/Fe/C powder catalyst in
0.1 M H
2
SO
4
(solid line) and 0.1 M HCOOH + 0.1 M
H
2
SO
4
(dashed line). Sweep rate 5 mV s
-1
. Inset
shows the anodic end for the voltammogram of 0.1 M
HCOOH + 0.1 M H
2
SO
4
expanded. 44
Figure 2.12 Polarization plots obtained at 90
o
C using formic acid
1 M on the anode and 1.2 L min
-1
of O
2
at the
cathode. (- ■-) Pt/Ru, (- ●-) Pt, (- ▲-) Pd. 46
Figure 2.13 Polarization plots obtained at 90
o
C using formic acid
1 M on the anode and 0.2 L min
-1
of O
2
at the
cathode. (- ■-) Pt/Ru, (- ●-) Pt, (- ▲-) Pd. 47
xiii
Figure 2.14 Polarization plots obtained at 90
o
C and formic acid
1 M. (- ■-) Pd on teflonized anode and O
2
at
0.2 L min
-1
, (- ●-) Pd on non-teflonized anode and O
2
at 0.2 L min
-1
, (- ▲-) Pd on teflonized anode and O
2
at
1.2 L min
-1
, (- ▼-) Pd at non-teflonized anode and O
2
at 1.2 L min
-1
. 48
Figure 2.15 Polarization plots for Pd in a non teflonized graphite
paper at 90
o
C, formic acid 1 M and 0.2 L min
-1
O
2
at
the cathode. (- ■-) 1:1, (- ●-) 3:1 and (- ▲-) 5:1 Nafion®
to catalyst. 50
Figure 2.16 Power plots for Pd in a non teflonized graphite paper
at 90
o
C, formic acid 1 M and 0.2 L min
-1
O
2
at the
cathode. (- ■-) 1:1, (- ●-) 1:3 and (- ▲-) 1:5 catalyst to
Nafion®. 51
Figure 2.17 Polarization plots for Pd/Au on a non teflonized
graphite anode obtained at 90
o
C with formic acid
1 M. (- ■-) O
2
at 0.2 L min
-1
, (- ●-) O
2
at 0.7 L min
-1
,
(- ▲-) O
2
at 1.2 L min
-1
. 53
Figure 2.18 Polarization plots for Pd/Fe/C on a non teflonized
graphite anode obtained at 90
o
C with formic acid 1
M. (- ■-) O
2
at 0.2 L min
-1
, (- ●-) O
2
at 0.7 L min
-1
, (- ▲-)
O
2
at 1.2 L min
-1
. 54
Figure 2.19 Polarization obtained at 90
o
C, formic acid 1 M and O
2
flow of 1.2 L min
-1
. (- ■-) Pd/Au, (- ▲-) Pd, (- ●-)
Pd/Fe/C and (- ▼-) Pt/Ru. 55
Figure 2.20 Power plot obtained at 90
o
C, formic acid 1 M and O
2
flow of 1.2 L min
-1
. (- ■-) Pd/Au, (- ▲-) Pd, (- ●-)
Pd/Fe/C and (- ▼-) Pt/Ru. 55
Figure 3.1 Distribution of species diagram (alpha plot) for
carbonic acid (H
2
CO
3
). α
i
is the concentration of
specie i divided the total concentration. 72
xiv
Figure 3.2 Tafel plots for the electrochemical reduction of carbon
dioxide over different metals in a pH 5.5 buffer
solution. Reproduced from [15]. 75
Figure 3.3 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn disc
electrode at 100 mV s
-1
. Bubbled with Ar (solid line).
Bubbled with CO
2
(dashed line). 78
Figure 3.4 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn
powder over a graphite disc electrode at 100 mV s
-1
.
Bubbled with Ar (solid line). Bubbled with CO
2
(dashed line). 79
Figure 3.5 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn
powder over a graphite paper GDE at 100 mV s
-1
.
Bubbled with Ar (solid line). Bubbled with CO
2
(dashed line). 80
Figure 3.6 Tafel plots obtained from the corresponding
voltammograms for bulk Sn metal disc ( ■), Sn powder
on a graphite disc ( ●) and Sn powder on a GDE ( ▲). 82
Figure 3.7 Formic acid Faradaic yields vs. potential for the
different electrocatalysts. (- ■-) Sn, (- ●-) In, (- ▲-) Cd
and (- ▼-) Pb. 86
Figure 4.1 Effect of solution purity on O
2
evolution. ( ○) no pre-
electrolysis. ( ●) 40 h of pre-electrolysis. ( ▲) 60 h of
pre-electrolysis. Reproduced from [18]. 98
Figure 4.2 Tafel plot in acid solution. ( ○) Stationary Pt wire
electrode in HClO
4
(pH 1). (+, x) Rotating Pt disk
electrode in H
2
SO
4
(pH 1). Reproduced from [19]. 99
Figure 4.3 TEM images of Pt/CFx at 20K X magnification 101
xv
Figure 4.4 Cyclic Voltammogram of catalyst in aqueous 0.1 M
H
2
SO
4
at 5 mV s
-1
with an O
2
flow rate of 8 mL min
-1
.
(- ■-) Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx. 103
Figure 4.5 Tafel plot obtained from the graphs in Figure 4.4. (- ■-)
Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx. 104
Figure 4.6 Nyquist Plot for the three catalysts at rest potential
with an O
2
flow of 8 mL min
-1
in the frequency range
10
5
to 0.01Hz. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx. 107
Figure 4.7 Kinetic control close-up for the Nyquist Plot of Figure
4.6 for the three catalysts at rest potential with an O
2
flow of 8 mL min
-1
. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-)
Pt/CFx. 108
Figure 4.8 Equivalence circuit used to fit the Nyquist plot. R
s
:
solution resistance, C
//
: parallel capacitance, R
ct
:
charge transfer resistance, C
dl
: double layer
capacitance, W: Warburg impedance. 109
Figure 4.9 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and
Pt/CFx (- ▲-) with 1M methanol at 90
o
C and O
2
flow
at 200 mL min
-1
. 113
Figure 4.10 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and
Pt/CFx (- ▲-) with 1M methanol at 90
o
C and O
2
flow
at 60 mL min
-1
. 113
Figure 4.11 Normalized polarization and power plot for Pt (- ■-),
Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M methanol at 90
o
C
and O
2
flow at 200 mL min
-1
. 114
Figure 4.12 Normalized polarization and power plot for Pt (- ■-),
Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M methanol at 90
o
C
and O
2
flow at 60 mL min
-1
. 115
xvi
Figure 4.13 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and
Pt/CFx (- ▲-) with 1M methanol, O
2
flow at
60
mL min
-1
at 60
o
C. 116
Figure 4.14 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and
Pt/CFx (- ▲-) with 1M methanol, O
2
flow at
60
mL min
-1
at 30
o
C. 116
Figure 4.15 Normalized polarization and power plot for Pt (- ■-),
Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M methanol, O
2
flow
at 60
mL min
-1
at 60
o
C. 117
Figure 4.16 Normalized polarization and power plot for Pt (- ■-),
Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M methanol, O
2
flow
at 60
mL min
-1
at 30
o
C. 117
Figure 4.17 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and
Pt/CFx (- ▲-) with 1M methanol at 60
o
C and air
flow
at 220
mL min
-1
. 118
Figure 4.18 Normalized polarization and power plot for Pt (- ■-),
Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M methanol at
60
o
C and air
flow at 220
mL min
-1
. 119
Figure 4.19 Polarization and power plot for Pt/C (- ●-) and Pt/CFx
(- ▲-) at 90
o
C. H
2
flow of 600 mL min
-1
and
700 mL min
-1
of O
2
. 120
Figure 4.20 Polarization and power plot for Pt/C (- ●-) and Pt/CFx
(- ▲-) at 90
o
C. H
2
flow of 300 mL min
-1
and
400 mL min
-1
of O
2
. 121
Figure 5.1 Schematic representation of the electron movement
from the carbon source to the reduction of oxygen in
microbial metabolism. 127
xvii
Figure 5.2 Anaerobic metabolism of Shewanella MR-1.
Reproduced from [23] 130
Figure 5.3 SEM images for an MEA used in a FC inoculated
using IP1. (A) 12.5 x magnification. (B) 6000 x
magnification. 135
Figure 5.4 (A) Digital photo taken of an MEA used in a FC
inoculated using IP2. (B) SEM image at 2000 x
magnification. (C) SEM image at 5000 x
magnification. 136
Figure 5.5 (A) and (B) OCV for two different MFC inoculated with
IP1. Temperature 28
o
C, O
2
flow rate 7 mL min
-1
. 138
Figure 5.6 (A) and (B) Cell voltage across a 1K Ω resistance for
two different MFCs inoculated with IP2. Temperature
28
o
C, O
2
flow rate 7 mL min
-1
. 140
Figure 5.7 (A) and (B) Polarization and Power density plots for
two different MFCs inoculated with IP1. (A) Plots
measured following the OCV of Figure 5.5(B).
Temperature 28
o
C, O
2
flow rate 7 mL min
-1
. 141
Figure 5.8 (A) and (B) Polarization and Power density plots for
two different MFCs inoculated with IP2. (A) Plots
measured following the OCV of Figure 5.6(A). (B)
( ─■─) correspond to a day after the second
inoculation and ( ─+ ─) correspond to measures
performed 15 days after the second inoculation.
Temperature 28
o
, O
2
flow rate 7 mL min
-1
. 143
Figure 5.9 (A) Cell voltage across a 1 K Ω resistance and (B)
Polarization and Power density plots for a MFC
prepared with Nafion® 117 using IP2 that was
subjected to membrane conditioning after assembly.
Temperature28
o
C, O
2
flow rate 7 mL min
-1
. 146
xviii
Figure 5.10 Polarization and Power density plots for a MFC with a
PVDF-PSSA MEA. Temperature 28
o
C and O
2
flowing
on the cathode at 7 ml min
-1
, using IP2. 147
Figure 6.1 Electrochemical glass cell used in CV and EIS
measurements. 157
Figure 6.2 Holder for the working electrode used for the cell in
Figure 6.1. 158
Figure 6.3 Glass cell used in the electroreduction of carbon
dioxide. 160
Figure 6.4 Digital picture of the assembled microbial fuel cell,
showing the cathode with the window and the
cathode. 166
xix
List of Schemes
Scheme 2.1 Formic acid oxidation mechanism occurring at the
surface of a metal electrode. 19
Scheme 2.2 Methanol oxidation mechanisms occurring at the
surface of a metal electrode. 19
Scheme 2.3 Steps in the oxidation of methanol to carbon dioxide
as presented by Parsons and VanderNoot [9]. 22
Scheme 3.1 Radical formation for adsorbed carbon dioxide at the
electrode. 67
Scheme 3.2 Reaction scheme for the electrochemical reduction of
CO
2
. i) Non-adsorbed CO
2
-•
on a metal electrode in
aqueous media. ii) Non-adsorbed CO
2
-•
on a metal
electrode in non-aqueous media. iii) CO
2
-•
adsorbed
on a metal electrode in aqueous media. iiib) CO
2
-•
adsorbed on a hydrogen-absorbing metal electrode in
aqueous media. iv) CO
2
-•
adsorbed on a metal
electrode in non-aqueous media. Reproduced from
[18]. 69
Scheme 3.3 Methanol formation from formic acid and methanol. 70
Scheme 3.4 Complete mechanism for the reduction of carbon
dioxide to formate ion. 71
Scheme 3.5 Carbon dioxide equilibrium in water. 72
Scheme 4.1 Pathways for the electrochemical reduction of oxygen.
Reproduced from [25, 26]. 97
xx
Abstract
Although fuel cells are based on a concept dated more than 170 years
ago, thorough and focused research did not take place in this field until the
last five decades. Therefore, considering that fuel cells are the main
candidates for portable power for the near future, there are questions that
need to be answered and problems that need to be solved if wide applications
of fuel cells are to be achieved. The research elaborated in this thesis was
carried out to solve some of the problems and overcome some of the
challenges. On the subject of Direct Formic Acid Fuel Cells, a search for a
better and cheaper catalyst than those already known was conducted, and at
the same time conditions for the Membrane Electrode Assembly preparation
were explored. Concomitantly, the electrochemical reduction of carbon dioxide
was investigated under conditions that primarily yield formic acid. The
research was aimed at maximizing the efficiency as well as the rate of the
carbon dioxide reduction.
One of the cathode reactions in fuel cells, which is the reduction of
oxygen to water, is known to have sluggish kinetics relative to most of the fuel
oxidation reactions. This is another key issue, where breakthroughs are
important and there is still room for improvement. A catalyst based on Pt
supported over carbon monofluoride (CFx) was prepared and tested in fuel
cells cathodes using hydrogen and methanol as fuels. In addition to the fuel
xxi
cell performance with the new cathodic catalyst, electrochemical properties of
the catalyst were also investigated.
The final part of the present thesis is related to microbial fuel cells.
These types of fuel cells have been developed in the last decade. For this
reason, the different components of the fuel cell and the assembly procedures
have neither been standardized nor their performances maximized. Therefore,
the research was aimed to test different materials for the fuel cell fabrication
as well as to standardize assembly procedures, while at the same time
improving the power output by application of all the advances that have taken
place in chemical fuel cells.
1
Chapter 1: Introduction
Fuel cells (FC) are electrochemical devices that convert chemical
energy directly into electrical energy and are considered to be the main choice
to replace batteries for delivering power for a wide range of applications. The
advances made in the basic understanding of fuel cells as well as
improvements in fuel cell technology have allowed researchers to realize the
benefits of switching to fuel cells as reliable power sources. At the same time,
the world population has come to the realization that a change in the economy
based on fossil fuels is undeniably necessary. Fossil fuels are finite resources
and their enormous unmitigated use in the past several centuries have led to
their increased cost. Furthermore, energy harnessed by fossil fuel combustion
releases carbon dioxide and other gases, which contribute to global climate
change (global warming) and other problems like acid rain.
The concept of fuel cells dates back to the first half of the 19
th
century,
when Sir William Grove obtained a small voltage by re-oxidizing hydrogen
obtained from water electrolysis [1]. Since then, fuel cells have been
considered as useful devices to convert chemical energy into electrical energy
in convenient and efficient ways. Chemical energy is stored in the bonds of the
molecules. An internal combustion engine operates to convert chemical
energy stored in the bonds to mechanical energy through combustion of a fuel
harnessing heat. The mechanical energy can in turn be converted into
electrical energy. The process of harnessing heat through a heat engine,
2
however is subject to the Carnot cycle through the Carnot theorem, which
states that the efficiency ( η) of an internal combustion engine is given by
equation 1.1, where T
1
is the temperature of the cold reservoir (i.e.
surroundings) and T
2
is the temperature of the hot reservoir (temperature at
which the engine operates), respectively.
2
1
1
T
T
− = η
(1.1)
All internal combustion engines have two reservoirs where one is at a lower
temperature than the other. The quotient between them limits the efficiency of
an engine. For an internal combustion engine, the efficiency is between 18 to
20 %. In a fuel cell, the chemical energy stored in the chemical bonds can be
converted directly into electrical energy. The fuel cells are not subject to the
Carnot cycle, which would allow, ideally, close to 100% efficiency from such a
device. The efficiency in a fuel cell is the product of the thermodynamic
efficiency ( η
thermo
), the voltage efficiency ( η
Voltage
) and the fuel efficiency ( η
fuel
)
(equations 1.2 and 1.3).
fuel voltage thermo
η η η η × × = (1.2)
fuel
v
nF
i
E
V
H
G
× ×
Δ
Δ
= η (1.3)
The thermodynamic efficiency ( η
thermo
) is the maximum work of the reaction
over the totally energy ( ∆H), and given that in a fuel cell the work is electrical,
3
then the maximum work is the Gibbs free energy for the reaction ( ∆G). This is
a fixed value which depends on the reaction thermodynamics and is
commonly used when referring to the fuel cell’s maximum efficiency or
theoretical efficiency. The voltage efficiency ( η
Voltage
) is the quotient between
the operating voltage (V) of the cell and the thermodynamically reversible
voltage (E
o
), and incorporates the total efficiency losses due to kinetics effects
on the electrode. Finally, the fuel efficiency ( η
fuel
) is the ratio of the fuel used
by the cell to generate electric current over the total fuel provided to the fuel
cell (v
fuel
). The numerator is calculated using the current extracted from the
cell (i), the number of electrons (n) for the particular reaction and Faraday’s
constant (F) as shown by the third term in equation 1.3. The voltage efficiency
accounts for the fact that not all the fuel supplied to the fuel cell will participate
in the electrochemical reaction. Some of the fuel flows through the cell without
reacting while some may take part in side reactions or even cross the
membrane to the cathode side as in the case of DMFC (methanol crossover).
Table 1.1 presents thermodynamic data for some of the most common fuels
used in fuel cells. The thermodynamic efficiency is also presented, and it is
interesting to note that for some fuels, like formic acid, the thermodynamics
dictate more than 100 % efficiency. The values presented in the table allow
the calculation of the total efficiency when the experimental voltage, current
and fuel supplied are known. The table presents an appealing panorama in
terms of the fuel thermodynamic properties, particularly the reversible cell
4
potential (E
o
cell
) and efficiency, but as will be shown in the next chapters,
kinetics play a very important role. In theory any reaction with an exothermic
enthalpy can be used in a fuel cell however favorable thermodynamics with
sluggish kinetics leads to inadequate results. In summary, kinetics and fuel
utilization are areas where research is essential in order to maximize the
overall efficiency (reaching closer to the upper limit given by η
thermo
).
Fuel ∆H
o
(KJ/mol) ∆G
o
(KJ/mol) E
o
cell
(V) n η
thermo
(%)
Hydrogen -286 -237 1.229 2 83
Methane -891 -818 1.060 8 92
Methanol -727 -703 1.214 6 97
Formic Acid -270 -286 1.480 2 106
Ammonia -383 -338 1.170 3 88
Hydrazine -311 -312 1.610 4 100
Table 1.1 Thermodynamic data and thermodynamic efficiencies ( η
thermo
) for the more
commons fuels. The values are under standard condition at 298 K with oxygen as
the oxidant and water produced in the liquid form.
There are different types of fuel cells: some are better suited for fixed
applications and others for portable applications. Table 1.2 gives a list of the
different type of fuel cells together with the electrolyte as well as the internal
charge carrier, identified as conducting specie. The charge carriers that flow
5
through the external circuit are always electrons. Optimal operating
temperature and range of power delivered is also presented.
Fuel Cell
type
Electrolyte
Conducting
Specie
Operating
temperature
(
o
C)
Power (kW)
Polymer
electrolyte
Polymer
membrane
H
+
60-80 1 – 100
Alkaline
Potassium
hydroxide
OH
-
70-120 1 – 10
Phosphoric
acid
Phosphoric
acid
H
+
160-200 10 – 1000
Molten
carbonate
Li/K carbonate CO
3
2-
650 100 – 10000
Solid oxide
Yttria-
stabilized
zirconia
O
2-
1000 1 – 10000
Table 1.2 Types of fuel cells, their electrolyte and characteristics (From references [2, 3])
The studies reported in this thesis will focus on the type referred to as
the Polymer Electrolyte Fuel Cell (PEFC) or Polymer Electrolyte Membrane
Fuel Cell (PEMFC). There is some confusion with the terms Polymer
Electrolyte Membrane and Proton Exchange Membrane, both abbreviated
PEM. The term Proton Exchange Membrane was used for a Polymer
Electrolyte Membrane Fuel Cell with hydrogen as the fuel. Given that
hydrogen was the first fuel used in the PEMFC, and the membrane in a
PEMFC working at acidic pH is a membrane exchanging H
+
with the media,
made these two terms interchangeable. This concept and others can be better
6
understood by taking a look at the fuel cell and how it operates. Figure 1.1
shows the basic scheme and functioning of a PEM fuel cell.
Figure 1.1 Schematic representation of the fuel cell with the different components and
processes.
The heart of the fuel cell is the Membrane Electrode Assembly (MEA).
The MEA is formed by a polymer membrane flanked by two electrodes. The
membrane acts as the ionic conductor between the two electrodes; the anode,
where the fuel is oxidized, and the cathode, where the oxidant is reduced.
7
These electrodes are made of a highly porous and conducting material and
are known as Gas Diffusion Electrodes (GDE), when reactants are gases
although the same terminology is used for liquid fuels. At the interface
between the GDE and the membrane lies a very thin layer of the catalyst. On
the anode the fuel is oxidized on the catalyst, where electrons are released to
the GDE. One of the products of the oxidation is H
+
, which moves through the
membrane to the cathode. This transport is possible because the membrane
exchanges H
+
with the fuel solution, and allows these protons to travel
through, thus the name proton exchange membrane. Above all, the membrane
works as the electrolyte were the ion, in this case H
+
, travels from anode to
cathode. Table 1.3 shows the half cell reaction for the oxidation of the most
common fuels in acidic media. At the reduction side the H
+
combines with
oxygen and the electrons that have moved through the external circuit,
resulting in the reduction of oxygen to water. The oxygen used in fuel cells is
supplied pure (100%) or as air (20%). For a PEMFC, the fuel can be a variety
of compounds, the most common ones being hydrogen, methanol, formic acid,
methane and ethanol.
The aforementioned confusion regarding the name of the FC that
utilizes hydrogen as a fuel, like “hydrogen fuel cell (HFC)” instead of PEMFC,
might be explained by the fact that when fuel cells were first introduced,
hydrogen was the only fuel. Compounds like methane or methanol could not
be converted into electrical power because of the lack of an appropriate
8
catalyst for the anodic oxidation. Thus reforming of fuels like methane or
methanol into hydrogen was needed in order to be used in a fuel cell. Hence
the fuel cells were originally mostly hydrogen based fuel cells or PEMFC. Later
with the discovery of new anode catalysts, the organic fuels were able to be
used directly, giving birth to terms like Direct Methanol Fuel Cell (DMFC) or
Direct Formic Acid Fuel Cell (DFAFC).
Anode Reaction E
o
(V)
H
2
(g) → 2H
+
(aq) + 2e
-
0.000
HCOOH(aq) → CO
2
(g) + 2H
+
(aq) + 2e
-
-0.110
CH
3
OH(aq) + H
2
O(l) → CO
2
(g) + 6H
+
(aq)+ 6e
-
0.030
CH
4
(g) + 2H
2
O(l) → CO
2
(g) + 8H
+
(aq) + 8e
-
0.170
CH
3
CH
2
OH(aq)+ H
2
O(l) → 2CO
2
(g) + 8H
+
(aq)+ 8e
-
0.480
Cathode Reaction
O
2
(g) + 4H
+
(aq) + 4e
-
→ 2H
2
O(l) 1.229
Table 1.3 Acid media half cell reaction for the most common compounds used as fuels in
the fuel cell anode and the half cell reaction for the O
2
used in the cathode.
(Standard potential tabulated as reduction potentials vs. the standard hydrogen
electrode (SHE)).
A different type of fuel cell described in Figure 1.2 is the alkaline fuel
cell (AFC). This type of fuel cell did not possess a membrane initially. A
membrane was later introduced into the design because of advances in
hydroxide based polymer membranes. As such they may be classified as
PEMFC in the near future. In an alkaline fuel cell, the common fuels used
9
could in theory be the same as mentioned for PEMFC, with the difference that
both the oxidation and reduction occurs in the basic media. In contrast to
PEFMCs, in the basic media the oxygen gets reduced to OH
-
, which then
travels to the anode to react with the fuel and produce electrons.
Figure 1.2 Schematic representation of the alkaline fuel (AFC) cell with the different
components and processes.
10
The process was initially made possible by the introduction of KOH as the
electrolyte to facilitate the reaction. Presently, anionic membranes are being
developed that allow the OH
-
anion to move from cathode to anode. These
polymer exchange membranes for OH
-
conduction would make the AFC a
subset of the PEMFC category. Figure 1.2 shows the schematic
representation for the AFC with PEM and in Table 1.4 the half cell reactions
for different fuels in basic media are shown.
Anode Reaction E
o
(V)
H
2
(g) + 2OH
-
(aq) → 2H
2
O(aq) + 2e
-
-0.830
HCOO
-
(aq) + OH
-
(aq) → CO
2
(g) + H
2
O (aq) + 2e
-
-0.723
CH
3
OH(aq) + 6OH
-
(aq) → CO
2
(g) + 5H
2
O (aq) + 6e
-
-0.818
CH
4
(g) + 8OH
-
(aq) → CO
2
(g) + 6H
2
O (aq) + 8e
-
-0.659
CH
3
CH
2
OH(aq) + 8OH
-
(aq) → CO
2
(g) + 7H
2
O(aq) + 8e
-
-0.806
Cathode Reaction
O
2
(g) + 2H
2
O(l) + 4e
-
→ 4OH
-
(aq) 0.400
Table 1.4 Half cell reactions in the basic media for most common fuels at the anode and the
half cell reaction for the reduction of oxygen at the cathode. (Standard potential
tabulated as reduction potentials vs SHE).
As mentioned above, there is a general consensus that the global
climate is in a warming phase. Carbon dioxide (CO
2
) is known to increase the
greenhouse effect on the atmosphere due to its characteristic absorption in the
infrared (IR) portion of the light spectrum. The current concern regarding
climate change is driving the search for ways to produce energy without
11
increasing the release of carbon dioxide into the atmosphere. By looking at the
process that occurs in a PEMFC, it is readily apparent that the only fuel that
does not produce carbon dioxide is hydrogen. However, hydrogen is difficult to
handle and store as a fuel, and is still derived from a process that produces
carbon dioxide as a byproduct, negating its benefits. On the other hand, if
carbon dioxide is recycle, PEMFCs like the DMFC or the DFAFC could be
considered neutral with respect to carbon dioxide production. Moreover, the
reduction of carbon dioxide could yield products used as a starting materials
for higher value chemicals and not only as a FC fuel. This is the main idea
behind the Methanol Economy® proposed by Olah, Goeppert and Prakash [4]
and described in Figure 1.3.
Methanol is an ideal starting point for the reduction of carbon dioxide.
Methanol is a liquid at ambient pressure and temperature (b.p. 64.6
o
C), and it
has a good volumetric energy density content of 4.8 kWh L
-1
, compared to 2.4
kWh L
-1
for liquid hydrogen. It can also serve as a carrier for hydrogen; one
liter of methanol contains more hydrogen than one liter of liquid hydrogen at -
253
o
C. Methanol is not only useful for FC applications but also for internal
combustion engines as a high octane, low emission fuel. Methanol can also be
converted to dimethyl ether and used as a high cetane replacement for diesel
fuel in diesel engines. Furthermore, it can be used as a feed-stock for the
production of a number of valuable and useful organic compounds [4].
12
Figure 1.3 Schematic representation of the carbon dioxide reutilization cycle.
The electrochemical reduction of carbon dioxide has been studied for
more than 100 years. The electrochemical reduction processes occurring at
the electrode must be understood in order to optimize conditions for large
scale reactions aimed to utilize carbon dioxide. As mentioned above, reduction
of carbon dioxide to methanol is highly desired, but formic acid, not methanol,
is the most favorable two-electron reduction product in aqueous media. Formic
acid is not as good an energy carrier compared to methanol; the energy
density is only 2.1 kWh L
-1
, but still possesses other desired properties. For
13
example, it is a proton conducting liquid that is easy to handle and can be
easily converted to methanol or used directly in fuel cells.
Microbial Fuel Cells (MFC) are the most recent type of fuel cell to be
proposed and the research in this area has accelerated in the last few years.
Figure 1.4 Schematic representation of a microbial fuel cell (MFC) with the different
components and processes.
Although the designs found in the literature present a lot of variation, it
is fundamentally a PEMFC that instead of a metal catalyst for the anode
14
reaction, it uses bacteria as the active component for the fuel oxidation. The
microorganisms feed on a particular carbon source and oxidize it, producing
protons and electrons that are transferred to the electrode and move to the
cathode through the external circuit. Similar to the PEMFC design, H
+
moves
from the anode to the cathode wherein, together with the electrons and
oxygen supplied, they form water. Figure 1.4 shows the schematic
representation of a MFC wherein the similarities with PEMFC can be seen.
This type of fuel cell represents a fairly inexpensive alternative to the other
types because of the fact that the microorganism is less costly than any
catalyst, particularly when compared with PEMFC. Although their power output
is quite low and a standard design has not been achieved for the devices, they
are still in a very early stage of development compared to the other types of
fuel cells, and thus improvements in the near future are likely. Such fuel cells
will also aid in purifying water (i.e. getting rid of toxic organic compounds).
15
Chapter 1: References
1. Grove, W., Philos. Mag., 1839. 14: p. 127-130.
2. Blomen, L.J.M.J. and Mugerwa, M.N., Fuel Cell Systems. 1993, New
York: Plenum Press
3. Larminie, J. and Dicks, A., Fuel Cell Systems Explained. 2nd ed. 2003,
Chichester: J. Wiley
4. Olah, G.A., Goeppert, A., and Prakash, G.K.S., Beyond Oil and Gas:
The Methanol Economy. 2006, Weinheim: John Wiley & Sons, Inc
16
Chapter 2: Catalysts for Direct Formic Acid Fuel Cells
(DFAFC)
2.1 Chapter 2: Introduction
As mentioned in Chapter 1, fuel cells of the PEM type are viable
candidates as power sources for portable applications, i.e. laptops, cell
phones, portable games, etc. However, considering the problems that carrying
a small tank of hydrogen could present for a cell phone or laptop computer,
fuel cells that employ liquid fuels are more attractive for such applications. Like
methanol, formic acid has been considered as a liquid fuel for such fuel cells,
and references can be found for such concepts as early as the 1960s [1, 2].
Past several decades have seen an increased interest in FC, mainly because
the National Aeronautics and Space Administration (NASA) started to use FC
for the Gemini project, using hydrogen as the fuel. The DMFC, however, is a
type of fuel cell that has also attracted a great deal of attention, partly because
of methanol’s favorable properties and also availability of efficient bimetallic
catalysts for methanol oxidation (e.g. Pt-Ru) [3, 4, 5]. Within the last few years,
formic acid has also drawn much attention. Formic acid posses a lower energy
density than methanol. However, for fuel cell purposes, it has been considered
to be a good fuel because it is a liquid at room temperature, non flammable
and has been recognized by the U.S. Food and Drug Administration (FDA) as
a safe food additive [6]. It is also a strong proton conducting electrolyte and
therefore is expected to facilitate ionic conduction. The theoretical standard
17
potential for a DFAFC is 1.43 V, higher than that of a DMFC. It has also been
reported to have a lower crossover through Nafion® [7, 8], one of the most
commonly used PEM, commercially available and manufactured by DuPont.
The crossover is an issue that principally affects DMFC. The PEMs posses
certain permeability toward methanol, allowing methanol molecules to cross
from anode to cathode. The effect of such a transfer of fuel from anode to
cathode is two-fold. First there is an energy loss due to the decrease in the
amount of methanol available at the anode, and second, because the
methanol reacts on the cathode electrode, causing a parasitic current and
poisoning of the catalyst as the reaction proceeds. Excess water formation at
the cathode also results in cathode flooding. As a consequence the
concentration of methanol must be limited to 1 – 2 M. On the contrary,
Nafion® appears to be less permeable to formic acid because formic acid
dissociates in water and the resulting ions do not easily penetrate the
membrane. Higher concentration of formic acid can therefore be used, which
compensates for the lower energy density. Similar to a DMFC, the DFAFC still
presents many challenges that must be overcome if commercial applications
are to be realized. The aim of the research presented in this chapter was to
optimize the conditions for MEA preparation in order for the DFAFC to perform
at the level that DMFC does currently, as well as preparation and testing of a
bimetallic catalyst in order to improve the performance of a DFAFC.
18
2.1.1 Electrochemical oxidation of formic acid on noble metals
In order to review the formic acid oxidation over noble metals, it is
instructive to compare the processes that take place in the methanol oxidation.
Although formic acid, like methanol, is among the simplest organic
compounds, the oxidation mechanism is not as simple as one might expect a
priori. It took decades of work performed by several researchers to determine
some of the simplest features of the mechanism for formic acid oxidation over
metal electrodes [9]. Even today some intermediates are not yet fully
understood [10]. Fundamental work on the oxidation of formic acid over Pt and
other noble metals like Ir, Rh, Pd and Au electrodes was performed in the
1970s by Capon and Parsons [11, 12, 13, 14]. Before this pioneering work, the
electrochemical oxidation mechanism was lacking several key steps. They
proposed a dual path mechanism, shown in Scheme 2.1, which is the currently
accepted scheme. After adsorption of the formic acid over the electrode, there
are two possible paths to product formation; a direct oxidation to carbon
dioxide that proceeds through the aforementioned unknown intermediate or
intermediates [10], and an indirect oxidation path that proceeds through
carbon monoxide. It took some time to identify the carbon monoxide as an
intermediate [9, 15, 16] in the indirect path, previously believed to be a COH
ads
species [11, 12] although the intermediate was branded a ‘poisoning
intermediate’, a quality ascribed to carbon monoxide over Pt [10]. This reaction
19
mechanism was later confirmed by
18
O labeling [17] and in situ IR
spectroscopy [18, 19].
HCOOH HCOOH
ads
CO
2
+ 2H
+
+ 2e
-
-H
2
O
CO
ads
+ OH
ads
+ H
+
+ e
-
CO
2
+ 2H
+
+ 2e
-
k
ads
k
d
k
p
k
ox
Scheme 2.1 Formic acid oxidation mechanism occurring at the surface of a metal
electrode.
The oxidation scheme for methanol is shown In Scheme 2.2 and is
analogous to the formic acid scheme. Though, it appears similar, it is
considered to be fundamentally different. The overall oxidation from methanol
to carbon dioxide is a 6 electron process compared to the 2 electron process
of formic acid. This also indicates that the direct 6 e
-
oxidation, as well as the
formation of the poisonous carbon monoxide (k
p
), must occur in several steps
[20].
CH
3
OH
ads
CO
2
+ 4H
+
+ 6e
-
-H
+
,4e
-
CO
ads
+ OH
ads
+ H
+
+ e
-
CO
2
+ H
+
+ e
-
k
ads
k
d
k
p
k
ox
CH
3
OH
Scheme 2.2 Methanol oxidation mechanisms occurring at the surface of a metal electrode.
One of the most used electrochemical techniques to study redox
mechanisms on electrode surfaces is cyclic voltammetry (CV). Figure 2.1
shows the cyclic voltammogram for a solution of 0.1M HCOOH in 0.5 M H
2
SO
4
20
on a Pt electrode, reproduced from reference [12]. Three anodic peaks and
one cathodic peak showing a large oxidation current are present. The first
anodic peak, at a potential of about 0.5 V, represents the direct oxidation of
bulk formic acid to carbon dioxide. This oxidation occurs only at sites on the
electrode surface that are not blocked by some intermediates not involved in
the direct reaction.
Figure 2.1 Cyclic voltammogram for Pt electrode in 0.5 M H
2
SO
4
(dashed line), 0.5 M H
2
SO
4
+ 0.1 M HCOOH (solid line). Sweep rate 140 mV s
-1
. Reproduced from [12].
As the potential increases, reaching the beginning of the oxidation
region, these residues start to undergo oxidation, resulting in an increase in
current. When the potential reaches approximately 1.2 V, oxidation of formic
acid ceases. After passing that potential, the Pt surface is well oxidized,
forming layers of PtO that become involved in the oxidation of formic acid until
a less reactive, passive oxide is formed. On the back sweep, oxidation of
21
formic acid on the Pt does not commence until the surface oxide reduction
region (~0.85 V). At that point, a large oxidation current is observed on the
freshly reduced surface. The current remains present until the potential has
reached 0.3 V. The width of the oxidation peak is elevated due to the
presence of a shoulder at 0.4 V. At this potential, adsorption of non-reactive
intermediates starts to take place resulting in competition between oxidation
and the blocking of sites. The difference in the hydrogen region between the
base electrolyte and the formic acid solution shows blocking of the surface by
adsorbed intermediates. As the potential reach the cathodic end and starts to
move in the anodic direction, bulk oxidation occurs on the free sites and the
cycle returns to the first anodic peak. The hydrogen region is the area between
potentials at which the hydrogen adsorbs and desorbs, below
0.35 V as shown on Figure 2.1. The voltammogram for methanol over Pt is
presented in Figure 2.2, as reproduced from the literature [21]. The electrolyte
solutions are the same as those used by Capon and Parsons with slightly
different concentrations, but the differences are negligible for the purposes of
comparison. It was suggested that the intermediates of the direct and indirect
oxidations for methanol and formic acid are different since different
electrochemical behavior was shown in polycrystalline and single crystal faces
of Pt [21, 22]. This can be observed in the voltammograms on Figure 2.1 and
2.2 and reflected by the absence of the first anodic peak and the voltage for
the cathodic peak when the surface oxides are being reduced. This difference
22
is real despite the fact that formic acid and formaldehyde are found during
methanol oxidation on Pt in acidic solution [23]. Scheme 2.3 was presented by
Parsons and VanderNoot [9] for the adsorbed species.
Figure 2.2 Cyclic voltammogram for Pt electrode in 1 M H
2
SO
4
(pointed line), 1 M H
2
SO
4
+
0.1 M CH
3
OH (dashed line). After holding the potential at 0.4V for 300s (solid
line). Sweep speed 100 mV s
-1
. Reproduced from [21].
CH
3
OH
ads
H
2
CO
ads
HCOOH
ads
CO
2
k
1 k
2
k
3
Scheme 2.3 Steps in the oxidation of methanol to carbon dioxide as presented by Parsons
and VanderNoot [9].
23
They concluded that if steps 1 and 3 are slower than 2, and 1 is slower than 3,
it would explain methanol behaving differently from formaldehyde and formic
acid, since step 1 is the rate determining step (rds). At the same time,
formaldehyde and formic acid would behave similarly but different than
methanol due to step 3. Still, the adsorption constant will be different, as well
as the surface coverage of the various intermediates. Therefore, it is possible
that similar intermediates are formed but different voltammetric responses are
obtained.
As mentioned earlier, the widespread use of DMFC was possible due to
the finding of Pt-Ru as an efficient bimetallic catalyst for methanol oxidation [3,
5]. Before then, FC prepared with Pt at the anode for DMFC would lose
performance after short periods as a result of catalyst poisoning. The absence
of a peak around 0.4 - 0.5 V in the methanol voltammogram, equivalent to the
first peak for formic acid, shows that on Pt, the oxidation proceeds mainly by
the indirect path through the formation of carbon monoxide, which poisons the
Pt. Formation of Pt oxides does not occur below voltages of 0.7 V. The
addition of Ru was found to facilitate the oxidation of carbon monoxide. At
voltages as low as 0.2 V, OH species would be adsorbed onto Ru sites. These
active OH facilitate the oxidation of carbon monoxide adsorbed onto
neighboring Pt sites [24, 25, 26, 27]. This improvement was maximized for
alloys of Pt-Ru with a 1:1 atomic ratio.
24
Although the voltammogram of formic acid on Pt shows that the
reaction proceed by two different mechanisms, under potentiostatic conditions,
the formation of carbon monoxide leads to the blocking of the reactions sites
and the suppression of the first peak. The Pt-Ru alloy was therefore used for
the formic acid oxidation. Markovic et al [28] showed that the Ru had the same
effect as for the methanol oxidation over Pt. Moreover, they demonstrated that
on Pt-Ru alloy, the indirect path, termed dehydration, would be favored over
the direct path (dehydrogenation), making the oxidation of formic acid similar
to that of methanol. Another interesting bimetallic alloy that has been well
studied for formic acid oxidation is Pt-Bi [29, 30, 31, 32]. Clavilier et al. showed
electrochemically that the addition of Bi adatoms to Pt increased the catalytic
activity towards formic acid oxidation. However, a later work [33] showed that
long term performance between Pt-Bi and Pt had no difference because of a
counter effect between enhancing the adsorption of OH, which increased the
rate of carbon monoxide oxidation, and blocking Pt sites for the formic acid
adsorption (k
ads
).
In the work by Capon and Parsons [12], Pd was shown to be an
interesting electrocatalyst for formic acid oxidation. In Figure 2.3, the
voltammogram for formic acid oxidation over Pd is reproduced from reference
[12]. The voltammogram shows almost no resemblance with the
voltammogram of formic acid on Pt. The second and third anodic peaks are
absent, and the cathodic peak starts after the oxide has been reduced.
25
Another interesting observation is that the anodic and cathodic peaks are of
approximately equal heights. They explained this observation by hypothesizing
that on Pd, the strongly adsorbing intermediates are not formed. Therefore the
oxidation proceeds only by the direct 2 e
-
oxidation path through carbon
dioxide. This should mean no poisoning of the electrocatalyst and a better
kinetic performance. Nevertheless, they rated the catalytic activity of Pt as
higher than that of Pd based on the peak current of the obtained
voltammograms.
Many other metals were studied as adatoms for the preparation of
alloys in order to enhance the catalytic activity of the Pt electrocatalyst [34,
35]. Among the most common metals found to improve performance include
Au [36, 37, 38, 39, 40], Pb [41, 42, 43, 44], Sn [43, 45, 46, 47, 48], Fe [49],
and Sb [43, 50]. Two other bimetallic alloys widely studied are Pt-Pd [14, 51,
52, 53, 54] and Pd-Au [52, 55, 56, 57]. The cyclic voltammogram studies for
both alloys showed that the dual metal site was free of the strongly bound
intermediate, enhancing the oxidation mechanism through the direct
pathways.
26
Figure 2.3 Cyclic voltammogram for Pd electrode in 0.5 M H
2
SO
4
(dashed line), 0.5 M H
2
SO
4
+ 0.1 M HCOOH (solid line). Sweep rate 140 mV s
-1
. Reproduced from [12].
2.1.2 Direct Formic Acid Fuel Cells
Before reviewing the research conducted on DFACF, a brief description
of the formulation behind the experiments performed to assess the fuel cell
performance is presented. The most common way to evaluate the
performance of a fuel cell is by obtaining a polarization curve. This polarization
can be carried out either galvanodynamically, i.e. sweeping the current from
zero to a predetermined value, or potentiodynamically, by sweeping the
potential from the open circuit cell voltage (OCV) to a voltage of nearly zero
(short circuit). The voltage is usually plotted on the Y axis, and the current,
which is converted to current density, (current divided by the active area of the
electrode) is plotted on the X axis. The voltage times the current gives the
power generated by the fuel cell, which is usually plotted in the same graph,
as shown in Figure 2.4.
27
Figure 2.4 Polarization and Power density plot.
The shape of the polarization plot obeys equation (2.1)
conc ohmic act rev
E E η η η + + + =
(2.1)
Where E
rev
is the theoretical cell voltage at open circuit conditions,
denominated reversible potential. The other three terms represent losses to
the voltage known as polarization ( η). These are the activation polarization
( η
act
), resistance polarization ( η
ohmic
) and concentration polarization ( η
conc
),
each of which predominates on a different zone of the plot as shown in Figure
2.6. Under experimental conditions, thermodynamic E
rev
is never observed
due to entropy losses and irreversible losses, only the experimental OCV is
observed under zero current flow, indicated as E
0
in the graph. Apart from the
28
ohmic polarization, related with an energy loss due to the internal resistance of
the cell which plays a predominant role on the central part of the plot, the other
two polarizations can be divided into contributions from the anode and the
cathode. The activation polarization, dominant at low current densities,
depends on both the chemical nature of the reacting compound and the
electrode composition in terms of chemical and physical nature. The
concentration polarization, also referred to as mass transport polarization, is
related to how fast the compounds reach the surface of the electrode, which
explains why it has a dominant contribution at high current densities.
In the previous section, a description of the basic study of the
electrochemical oxidation of formic acid was discussed. Pt, Pd, Pt-Ru and
other related alloys were shown to be, under CV conditions, good catalysts for
the oxidation of formic acid. However, the fuel cell is a more complex system
and a good CV response does not necessarily imply that that compound with a
particular electrocatalyst will perform well in the fuel cell. A good catalytic
activity in the CV is reflected as a high OCV measured in a fuel cell, however,
the kinetics will not necessarily be flawless. Nevertheless the discussed
catalysts with good CV responses for formic acid oxidation were the starting
point for the research conducted on DFAFC for the past twelve years [7, 8, 38,
42, 53, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76,
77, 78, 79, 80]. Initially, Weber and co-workers prepared a DFAFC using Pt
and Pt-Ru as catalysts for the anode [73]. The PEM used was
29
polybenzimidazole (PBI) doped with phosphoric acid. This membrane allowed
the fuel cell to operate at a temperature of 170
o
C. A potential of 100 mV lower
than that attained for methanol under similar conditions was observed which
suggested that Pt-Ru was more active than Pt, as expected from the cyclic
voltammetry studies. Weber also performed real-time mass spectrometry
measurements in the fuel cell and observed that carbon dioxide was the only
reaction product. Although this research ultimately resulted in a functional
DFAFC, the membrane and conditions used are not the most common ones
for PEMFC, a reason why no further work was followed up. In following works
Pt, Pd and Pt-Ru were still the catalysts used for the fuel cell, but Nafion® was
the main choice as the PEM and temperature testing conditions were below 90
o
C down to room temperature. The major focus for catalyst has been on Pt-Ru
and Pd [8, 53, 60, 61, 62, 63, 64, 70, 71, 72, 79, 80, 81, 82] as well as Pd
supported over graphite. Concentrations of formic acid up to 20 M were
reported, and they showed good results with active as well as passive fuel
cells. The term active fuel cell refers to a set up made in such a way that the
fuel is circulated through the anode with a pump, therefore the fuel
concentration is almost constant through the entire test and mass transport
limitations are not present. In contrast, a passive fuel cell possesses a fuel
reservoir at the anode filled with the fuel to be used along the test. Some of
the results are shown in Figure 2.5 and Figure 2.6 [62, 63]. Figure 2.5
corresponds to data of an active fuel cell with a 5 cm
2
electrode area, with dry
30
air circulating through the cathode and a testing temperature of 30
o
C with 3 M
formic acid. Figure 2.6 shows the polarization and power obtained for a
passive fuel cell with a 4 cm
2
electrode area, also tested at 30
o
C with different
concentrations of formic acid.
Another report worth mentioning [77] involved the preparation of a
catalyst composed of Pd with phosphorous supported on graphite. Although
they did not perform fuel cell tests or report any polarization or power plots, the
voltammograms acquired show the same behavior as the Pd electrode, with a
decrease in the peak potential and an increase of the current, indicating that
the performance expected for this catalyst on the fuel cell should be better
than Pd alone. In all the reports mentioned above Pt was used as the cathode
catalyst, whereas MEA preparation (electrode material, size, assembly) varied
greatly from report to report.
31
Figure 2.5 (A) fuel cell polarization plots and (B) power density curves with 3 M formic acid at
30
◦
C for Pd black (PdBl) and Pd over carbon. Reproduced from [63].
32
Figure 2.6 Passive air breathing DFAFC vs. the formic acid concentration at 30
◦
C: (A) cell
potential and (B) power density curves. Reproduced from [62].
33
2.2 Chapter 2: Results and Discussion
2.2.1 Catalyst preparation characterization
Bimetallic powder catalysts were prepared with the goal of improving
formic acid oxidation in a fuel cell environment. Although research performed
on bulk metal and metal alloy electrodes indicate that formic acid oxidation
proceeds mainly by the direct route, it was reported that in fuel cells, after a
long period of operation, a decrease in the performance is observed, most
likely caused by the accumulation of poisonous species in the catalyst layer
[80]. Characterization of the powder catalyst consisted of surface analyses by
Energy Dispersive X-ray Spectrometry (EDS) and X-ray Photoelectron
Spectroscopy (XPS) to assess the composition of the bimetallic catalyst.
Electrochemical measurements were performed to study the oxidation of
formic acid over the prepared catalyst powder mixed with the ionomer used in
MEA preparation over a conductive substrate (i.e. the electrode) in a similar
way to the oxidation of formic acid over bulk metal electrodes as described in
Section 2.1.1. The rationale behind using an electrode with the powder
catalyst mixed with the ionomer is to assess the oxidation of formic acid in an
environment that more closely approximates that of a regular fuel cell. In the
fuel cell electrode, the catalyst is a powder with a high surface area that is in
close contact with the membrane and the ionomer. The ionomer works as an
adhesive between membrane, catalyst and the electrode. In the present study
the membrane and ionomer used in the preparation of the fuel cell is Nafion®.
34
The high surface area and the acidity of the Nafion®, among other
considerations, could affect the response on the voltammogram in a different
manner than plain metal electrodes.
Commercial Pt and Pd black catalysts were also tested in order to
compare with studies carried out previously on bulk electrodes and to use
them as a baseline for the bimetallic catalyst. In all the experiments, the
potential was swept at 5 mV s
-1
and plotted against the SHE. Because of the
high surface electrode area, at sweeping speeds faster than 10 mV s
-1
peaks
broaden too much for a meaningful interpretation. The current was converted
to current density (j) using the geometrical area of the electrode (1 cm
2
).
Figure 2.7 shows the voltammogram of formic acid in H
2
SO
4
for Pt black
powder. The voltammogram of the base electrolyte is also presented.
Comparison with the Pt bulk electrode voltammogram (Figure 2.1) shows
differences in the formic acid oxidation. Even the voltammogram in the base
electrolyte presents differences, which include a higher capacitive current, due
to the surface area of the catalyst in powder form and the shift to a more
cathodic potential for the oxide reduction peak. On the voltammogram for the
oxidation of formic acid several differences are also observed. The three
anodic peaks seen in Figure 2.1 are not easily visible. All the anodic peaks
have shifted to more anodic potentials. Furthermore, the decrease in current
between the anodic peaks is much less pronounced. This decrease might be
35
due to the fact that there is still oxidation occurring between the peaks, or to a
higher capacitive current.
Figure 2.7 Cyclic Voltammogram for Pt powder catalyst in 0.1 M H
2
SO
4
(solid line) and
0.1 M H
2
SO
4
+ 0.1 M HCOOH (dashed line). Sweep rate 5 mV s
-1
.
The third anodic peak can be seen starting at the end of the sweeping
cycle, and it was more noticeable when the anodic limit was increased to
1.7 V. Two other important differences are i) that the first anodic peak presents
a current similar to the second peak. The first anodic peak is the oxidation of
formic acid by the direct pathway, meaning that in a fuel cell the current
expected by the direct oxidation should be higher than one would expect
36
based on measurements in bulk electrodes. The other difference is that the
cathodic peak has a lower anodic current compared to the bulk metal, most
likely due to the fact that fresh reduced sites are present in less amount
compared to the bulk metal electrode and it has also shifted to more cathodic
potentials. However, even with these differences the catalyst more or less
behaves in the same way as the bulk electrode.
Figure 2.8 shows the voltammogram for formic acid in H
2
SO
4
over a Pd
electrode. The voltammogram in the base electrolyte differs with the one of the
bulk electrode. The forward and back scan at a potential higher than 0.8 V is
similar with the one presented in Figure 2.3, however for cathodic potentials
the differences are considerable. The reduction of the oxide peak has also
moved to lower potentials, as in the case for Pt, and the hydrogen region as
well as the double layer region is on the negative current densities. However
the peaks in the hydrogen region are still well defined indicating catalytic
activity. The negative behavior could be explained due to the higher hydrogen
overpotential that Pd presents but also for the type of electrode used in this
study due to the presence of the Nafion® in contact with the catalyst. The
voltammogram for the formic acid oxidation is very similar to the one acquired
for bulk metal. Particularly the forward scan, although there are up to four
anodic peaks above 0.8 V. Capon and Parson noted that the absence of
peaks at these potentials in comparison with Pt was important. Why they were
not observed in the bulk electrode is difficult to answer. Capon and Parson
37
discussed that this situation could arise if there was some Pt contamination on
the Pd [12].
Figure 2.8 Cyclic Voltammogram for Pd powder catalyst in 0.1 M H
2
SO
4
(solid line) and
0.1 M H
2
SO
4
+ 0.1 M HCOOH (dashed line). Sweep rate 5 mV s
-1
.
This contamination is not likely in the present case. Care was taken to avoid
any contamination, and the measurements were repeated several times. In
these experiments the more striking difference, besides the high surface area
of the catalyst, is again the presence of Nafion® which influences the surface
behavior. On the backward scan, the formic acid oxidation cathodic peak is
almost concomitant with the oxide reduction peak, which resembles more the
38
behavior on bulk Pt than that of bulk Pd. Again contamination with Pt is
disregarded not only for the reason given above but also because the
voltammogram does not present all the features of a Pd electrode with a small
amount of Pt [14]. In spite of the mentioned differences between the
voltammograms of powder and bulk metals electrodes, Figures 2.7 and Figure
2.8 show that the peak value for the oxidation current of formic acid on Pd is
higher than that on Pt. This value indicates a higher reaction rate on Pd than
on Pt, considering that the amount of catalyst over the electrode is
approximately equal.
The first bimetallic catalyst tested in this work is Pd/Au. In Section 2.1.1
it was mentioned that Au was used as an adatom for Pt [39] and that the
electrochemical oxidation of formic acid was also studied over Pd/Au.
Although Au is considered to be a poor electrocatalyst [12], it was tried not
only in the oxidation of organic molecules but also as an electrocatalyst for the
reduction of oxygen [83]. EDS analysis showed that the catalyst composition
obtained was a 3:1 Pd-Au atomic ratio, which is roughly 60 % Pd – 40 % Au
by weight. Figure 2.9 shows the voltammogram for the Pd/Au powder. As in
the case of Pd, the hydrogen region and double-layer region are on the
negative side of the current for the voltammogram on the base electrolyte but
not in the presence of formic acid. The voltammogram is similar to the ones
obtained on bulk electrode with some notable differences. In the
voltammogram obtained by Beden and co-workers [55] the oxide reduction
39
peaks observed at 1.1 V and 0.55 V show inverted values of current, with the
more anodic peak being the larger one. Also, another anodic peak is observed
at approximately 0.7 V in their voltammogram and the anodic peak at 0.95 V
appeared at more positive voltage. In the hydrogen region they did not see the
adsorption/desorption peaks for hydrogen. Kibler and co-workers also
observed the two cathodic peaks, but these peaks were found to be closer to
each other. They studied the peaks heights and found that they were directly
related to the amount of Au and Pd in the mixture [56], which agrees with the
results presented here. In terms of the voltammogram in formic acid, the one
presented here also agrees with the one obtained by Kibler, while a
comparison with Beden could not be made due to the fact that they performed
the oxidation in a neutral media, using K
2
SO
4
as the electrolyte. The anodic
peak at 0.2 V – 0.3 V corresponds to the direct oxidation of formic acid and is
about the same current as that for just Pd, being a few mV more cathodic. In
the back scan, the oxidation peak is not as high as for the Pd, and also start
when the oxide begins to reduce, meaning that the Au forms oxides more
readily than Pd and agrees with the observation by Capon and Parson that Au
forms stronger chemisorption bonds with formic acid intermediates than Pd.
Kiebler and co-workers as well as Baldauf and co-workers [52], obtained a
higher current for the back scan than for the forward scan in all cases, but the
electrode that they used was either Pd or Au with monolayers of the other
metal.
40
Figure 2.9 Cyclic Voltammogram for Pd/Au powder catalyst in 0.1 M H
2
SO
4
(solid line) and
0.1 M H
2
SO
4
+ 0.1 M HCOOH (dashed line). Sweep rate 5 mV s
-1
.
The other bimetallic catalyst prepared was Pd/Sn. It was also
mentioned that Sn was used as an adatom for Pt for the formic acid oxidation,
but it was never combined with Pd for the same purpose. Sn has a high
overpotential for the hydrogen reduction (see Chapter 3) and is considered an
oxygen-adsorbing adatom [43]. EDS analysis showed a composition of 2:3 Pd
to Sn, corresponding to a 40% Pd 60% Sn by weight. As can be observed in
Figure 2.10, in the base electrolyte, the oxide reduction peak is on the
cathodic end of the cycle and adsorption/desorption peaks for the hydrogen
41
are not observed. On the voltammogram in formic acid, a small peak
corresponding to the direct oxidation at around 0.2 V can be observed,
although the current is still negative. The anodic peak at 0.8 V observed in
pure Pd is also still observed and on the back scan there are two cathodic
peaks at around 0.65 V with oxidation currents. This agrees with the published
results presented for Pt/Sn in terms of how the Sn promotes the oxidation of
the formic acid, by forming oxides that would react with the strongly bonded
intermediates, seen in the voltammogram as two peaks occurring at the onset
of the oxide reduction and at the decrease in the reduction current. Based on
the current values shown in the voltammogram, good performance of the
catalyst in a fuel cell is not expected. The possible reason of the unwanted
results in terms of the voltammogram shape and currents is probably due to
the high amount of adatom (Sn) present in the bimetallic catalyst, which is
contrary to the “adatom” idea.
42
Figure 2.10 Cyclic Voltammogram for Pd/Sn powder catalyst in 0.1 M H
2
SO
4
(solid line) and
0.1 M HCOOH + 0.1 M H
2
SO
4
(dashed line). Sweep rate 5 mV s
-1
. Inset shows
the anodic end for the voltammogram of 0.1 M HCOOH + 0.1 M H
2
SO
4
expanded.
The last bimetallic catalyst prepared was Pd/Fe but in this case
supported over carbon. The reasons which lead to the preparation of the
catalyst over carbon will be discussed later. Figure 2.11 shows the
voltammograms for the Pd/Fe/C in formic acid and base electrolyte. Very few
studies are present where Fe was used as an adatoms and in all cases was
only used to modify Pt. Nevertheless Watanabe and coworkers [84, 85, 86]
found that Pt-Fe alloys presented a high CO tolerance toward the oxidation of
hydrogen or the reduction of oxygen. The voltammogram in the base
43
electrolyte shown here is very similar with the one on Pd shown in Figure 2.8
having the oxide reduction peak at lower potentials and the hydrogen region
on the negative current densities side. The voltammogram on formic acid
presents the three anodic peaks expected for the formic acid oxidation, but no
positive cathodic current representing the direct oxidation in the back scan is
shown. The voltammogram obtained by Chen and coworkers for formic acid
on Pt-Fe in a 0.1 M HClO
4
solution resembles very closely to the one
presented in Figure 2.1 on pure Pt with the difference that the first anodic peak
is almost noticeable. The voltammogram on the in house catalyst shows an
improvement in terms of the direct oxidation of formic acid on the forward scan
seen by the fact that the current density of the first peak is similar to the
second peak indicating probably that less CO is adsorbed at lowers potentials.
This is also indicated by the negative current below 0.1V on the forward scan.
However the fact that no positive current is seen on the back scan might
indicate that a complete oxidation of poisonous intermediates is not
accomplished. Based on the voltammogram, the Pd/Fe/C is expected to
perform better than Pd/Sn but probably less than Pd/Au based on the attained
current shown by the peaks.
44
Figure 2.11 Cyclic Voltammogram for Pd/Fe/C powder catalyst in 0.1 M H
2
SO
4
(solid line) and
0.1 M HCOOH + 0.1 M H
2
SO
4
(dashed line). Sweep rate 5 mV s
-1
. Inset shows
the anodic end for the voltammogram of 0.1 M HCOOH + 0.1 M H
2
SO
4
expanded.
2.2.2 Direct Formic Acid Fuel Cell testing
Experiments performed on the various fuel cells were initially designed
to find the best condition for the MEA fabrication. This included the choice of
electrode material, amount of catalyst, amount of Nafion® ionomer and
membrane thickness as well as the temperature, pressure and duration of
pressing process that would give the best performance and or maximum
power.
45
At the outset, the methodology chosen was to use preexisting materials
and conditions that gave the best performances for methanol fuel cells. From
these results, changes to the materials and methods were tried and compared
with literature reports.
Figure 2.12 shows a polarization plot for the formic acid fuel cell using
commercial Pt, Pt/Ru and Pd as anode catalysts. The MEA preparation used
was the one that renders the best result for DMFC, namely; graphite paper as
GDE with a Teflon® coating (i.e. teflonized) on both the anode and the
cathode, Nafion® 117 as the membrane and a ratio of 1:1 for catalyst and
Nafion® ionomer in the paint used to prepare the electrode. The electrode size
was 25 cm
2
. Pd, although yielded an OCV 100 mV higher than the other two
catalysts, did not perform as well as one would expect from the literature data
[70, 71]. This is because there are a lot of empirical steps in the MEA
fabrication process which are not typically reported and may contribute to the
overall performance. Pt/Ru performed better than Pt demonstrating that an
oxygen-adsorbing adatom like Ru is useful in facilitating the oxidation of formic
acid.
46
Figure 2.12 Polarization plots obtained at 90
o
C using formic acid 1 M on the anode and
1.2 L min
-1
of O
2
at the cathode. (- ■-) Pt/Ru, (- ●-) Pt, (- ▲-) Pd.
Figure 2.13 shows the same fuel cell tested at a lower oxygen flow on
the cathode. In DMFC, 1.3 L min
-1
of oxygen gives the best performance as
mentioned above, but for Pt and Pd improvements were observed at lower
flow. Moreover, Pd shows an OCV 120 mV higher than Pt and Pt/Ru at that
flow. In DMFC the aforementioned flow gives the best performance because it
gives the necessary amount of oxygen that allows the methanol to react at a
high rate, and avoids the flooding of the cathode, which is something that is
more likely to occur at lower flow rates. A possible explanation for why the
47
formic acid seems to have better performance at lower flow is the possibility
that not enough water is reaching the cathode (less water drag).
Figure 2.13 Polarization plots obtained at 90
o
C using formic acid 1 M on the anode and
0.2 L min
-1
of O
2
at the cathode. (- ■-) Pt/Ru, (- ●-) Pt, (- ▲-) Pd.
In DMFC and DFAFC, the oxygen or air circulating through the cathode
is dry. Dry oxygen or air is used because an aqueous solution is circulated
through the anode and while the teflonized graphite electrode repels water,
sufficient amount of water can still reach the cathode. An aqueous solution of
formic acid differs from an aqueous solution of methanol. Pure formic acid is
denser and more viscous than pure methanol, and formic acid ionizes in
48
water. For these reasons it is possible that the Teflon coating on the graphite
electrode repels the formic acid solution more efficiently than the methanol
solution, thus giving a better performance with lower oxygen flow rates.
Figure 2.14 Polarization plots obtained at 90
o
C and formic acid 1 M. (- ■-) Pd on teflonized
anode and O
2
at 0.2 L min
-1
, (- ●-) Pd on non-teflonized anode and O
2
at
0.2 L min
-1
, (- ▲-) Pd on teflonized anode and O
2
at 1.2 L min
-1
, (- ▼-) Pd at
non-teflonized anode and O
2
at 1.2 L min
-1
.
Figure 2.14 shows a comparison of two MEA using Pd on the anode.
One of the MEA was prepared using a teflonized graphite paper, while the
other was prepared using a graphite paper with no Teflon treatment. Non-
teflonized graphite paper was used at the cathode in both cases. The data for
49
oxygen flows of 1.2 and 0.2 L min
-1
are shown for comparison. It can be seen
that with graphite paper as the electrode without Teflon coatings the
performance is improved considerably for both oxygen flow levels. However,
the cell tested at lower oxygen flow still performs better, but still not at the level
of performance of the Pt/Ru. Another observation is that the polarization
curves have a high negative slope, indicating a high internal resistance;
however the resistance of each cell was measured after assembly with a
milliohmmeter and all the cells that provided useful data showed resistance in
the same range.
The next step was to vary the amount of Nafion® ionomer employed in
the catalyst paint preparation. Typically the catalyst ionomer mixture is
prepared based on 200 mg of catalyst, a fixed amount of water and an amount
of ionomer in a 1:1 catalyst/ionomer ratio. A detailed explanation is given in
Chapter 6. The 1:1 ratio of catalyst/ionomer was found to give the best result
for DMFC as mentioned earlier. However, the proportion is purely empirical
and was found to work well for methanol. At the same time, based on
experimental results obtained in our laboratory, it is known that other
proportions may be equally effective. In fact, using a ratio of 1:5
catalyst/ionomer with a different press condition yields similar results for
DMFC. These results, and the fact that other authors have achieved success
with different amounts of ionomer [63, 70], prompted an investigation where
three different ratios of catalyst/ionomer were compared; 1:1, 1:3 and 1:5.
50
Figure 2.15 illustrates the polarization plots for Pd on non-teflonized graphite
paper with an oxygen flow of 0.2 L min
-1
on the cathode. The data used for the
1:1 ratio is the same for Pd on non teflonized anode presented in Figure 2.14.
Although in every case different flows were used, a low flow rate of
0.2 L min
-1
yielded the best result.
Figure 2.15 Polarization plots for Pd in a non teflonized graphite paper at 90
o
C, formic acid 1
M and 0.2 L min
-1
O
2
at the cathode. (- ■-) 1:1, (- ●-) 3:1 and (- ▲-) 5:1 Nafion® to
catalyst.
This might not be apparent in Figure 2.15 due to the fact that the
limiting current is the same for the 1:3 and 1:5 ratio. Therefore the power was
plotted and is shown in Figure 2.16. The power plot at 1:5 ratio shows higher
51
performance and from Figure 2.15, the OCV is 860 mV, a high value and very
difficult to obtain in DMFC with Nafion®. Also the limiting current is close to the
one shown by Pt/Ru in Figure 2.13.
Figure 2.16 Power plots for Pd in a non teflonized graphite paper at 90
o
C, formic acid 1 M
and 0.2 L min
-1
O
2
at the cathode. (- ■-) 1:1, (- ●-) 1:3 and (- ▲-) 1:5 catalyst to
Nafion®.
In summary, the MEA that gives the best results for Pd black as the
anode catalyst consisted of a non-teflonized graphite paper as the anode
electrode, and the catalyst mixture used to paint the electrodes has a 1:5 ratio
52
of catalyst to ionomer, and in terms of low oxygen flow conditions gave better
performance.
In terms of the press settings for MEA preparation, the temperature
used in all cases was 150
o
C when working with Nafion®, which is the glass
transition temperature (Tg) for the polymer. What is intended is to get the
Nafion® close to a fusion state, at which point it should bind better to the
catalyst, this is the main reason why Nafion® ionomer is added to the catalyst
mixture used to coat the graphite electrode. The pressure used, however,
varies based on the amount of Nafion® ionomer present in the catalyst
mixture. Experiments performed using different ratios of catalyst/ionomer
showed that increasing the amount of Nafion® required the pressure values to
be lowered. As yet there is no explanation at an elementary level why this is
the case.
Having optimized the MEA preparation and conditions, the catalysts
prepared in house were tested. Figure 2.17 shows the performance for Pd/Au.
An OCV of ca. 0.9 V was obtained at three different oxygen flows, and was the
highest obtained for formic acid in our laboratory. The low flow of 0.2 L min
-1
,
which gave the best performance during the optimization as was described,
gave the higher performance for Pd/Au as well. This performance was also
superior to the MEAs prepared using commercial Pd, Pt and Pt/Ru.
53
Figure 2.17 Polarization plots for Pd/Au on a non teflonized graphite anode obtained at
90
o
C with formic acid 1 M. (- ■-) O
2
at 0.2 L min
-1
, (- ●-) O
2
at 0.7 L min
-1
, (- ▲-) O
2
at 1.2 L min
-1
.
Figure 2.18 shows the results for the Pd/Fe/C catalysts. The OCV
showed to be high for the different oxygen flows and the lowest flow of
0.2 L min
-1
presented a slightly better performance as it was the case with
Pd/Au. Again, it is difficult to assess the best performance from the
polarization plot due to the similarity of the three catalysts. As was expected
from the voltammograms the performance is lower than the one shown by
Pd/Au.
54
Figure 2.18Polarization plots for Pd/Fe/C on a non teflonized graphite anode obtained at
90
o
C with formic acid 1 M. (- ■-) O
2
at 0.2 L min
-1
, (- ●-) O
2
at 0.7 L min
-1
, (- ▲-) O
2
at 1.2 L min
-1
.
Figure 2.19 shows the performance, while Figure 2.20 shows the power
plot for Pd, Pt/Ru, Pd/Au and Pd/Fe/C. Again the Pd/Au electrode shows not
only the highest OCV but also the highest limiting current density and power
density. In perspective, the limiting current density of 440 mA cm
-2
shown for
Pd/Au represent a total current of 11 A for the tested fuel cell. For comparison
the data extracted from references [62, 63] and presented in Figures 2.5 and
2.6 correspond to 4.4 A and 2, respectively due to the electrode size that were
employed.
55
Figure 2.19 Polarization obtained at 90
o
C, formic acid 1 M and O
2
flow of 1.2 L min
-1
. (- ■-)
Pd/Au, (- ▲-) Pd, (- ●-) Pd/Fe/C and (- ▼-) Pt/Ru.
Figure 2.20 Power plot obtained at 90
o
C, formic acid 1 M and O
2
flow of 1.2 L min
-1
.
(- ■-) Pd/Au, (- ▲-) Pd, (- ●-) Pd/Fe/C and (- ▼-) Pt/Ru.
56
While Pd/Au and Pd/Fe/C performed well, the corresponding Pd/Sn
catalyst did not yield good results. The best OCV obtained was ca. 0.8 V and
the limiting current density did not exceed 1.2 A, representing just a few
mA cm
-2
. During MEA preparation and testing it was realized that the particle
size of the catalysts prepared presented some complications. During the paint
preparation, a homogeneous suspension was difficult to obtain, which led to
complication for the painting procedure. Obtaining a uniform layer of catalyst
over the graphite paper was compromised. Also, after testing, when the fuel
cell was taken apart, leaching of the catalyst from the electrode was observed.
This is why the third catalyst, Pd/Fe was prepared on carbon support. The
suspension obtained with the supported catalyst for applying it to the MEA was
more uniform than the unsupported ones and application of such catalyst over
the electrode presented less complications. The limiting current density was
extended to 4 A for Pd/Sn supported on carbon. These results will be
presented elsewhere [87]. Another issue that must be addressed relates to the
Pd catalysts: the redox potential of oxygen in the fuel cell approaches the
redox potential of the Pd. This could cause an oxidation of the Pd and the
consequent loss of catalyst to the fuel solution. Sn and Fe would present a
similar problem, but is not the case for Pt or Au, whose redox potentials are
1.2 V and 1.5 V, respectively.
57
2.3 Chapter 2: Conclusion
The data presented demonstrates that the direct formic acid fuel cell is
a good candidate as an energy source for portable applications. Pd and Pd
containing catalysts both performed well. The OCV was more than 100 mV
higher when compared with the Pt catalyst, and also when compared with
DMFC using Nafion®. This agrees with the fact that the theoretical value of
OCV is higher for DFAFC than DMFC.
The research presented has shown an increase in the fuel cell
performance by tuning the different parameters in the MEA preparation and
assembly when the commercial catalysts were used. Moreover, the
performance was further improved with one of the catalysts prepared in house.
However Pd is a metal not as noble as Pt or Au, indicated by the standard
redox potential, which creates some problems related to the re-dissolution of
the Pd (due to oxidation) from the catalyst layer to the fuel solution.
58
2.4 Chapter 2: References
1. Breiter, M.W., Electrochemical Processes in Fuel Cells. 1969, New
York: Springer-Verlag
2. Wynn, J.E., Proc., Ann. Power Sources Conf., 1960. 14: p. 52-55.
3. Jansen, M.M.P. and Moolhuysen, J., Electrochim. Acta, 1976. 21(1): p.
869.
4. Surampudi, S., Narayanan, S.R., Vamos, E., Frank, H., Halpert, G.,
Laconti, A., Kosek, J., Prakash, G.K.S., and Olah, G.A., J. Power
Sources, 1994. 47(3): p. 377-385.
5. Watanabe, M., Uchida, M., and Motoo, S., J. Electroanal. Chem., 1987.
229(1-2): p. 395-406.
6. U.S. Code of Federal Regulations - FDA, CFR Part 21 - 186.1316,
172.515
7. Jeong, K.-J., Miesse, C.M., Choi, J.-H., Lee, J., Han, J., Yoon, S.P.,
Nam, S.W., Lim, T.-H., and Lee, T.G., J. Power Sources, 2007. 168(1):
p. 119-125.
8. Rhee, Y.W., Ha, S.Y., and Masel, R.I., J. Power Sources, 2003. 117(1-
2): p. 35-38.
9. Parsons, R. and Vandernoot, T., J. Electroanal. Chem., 1988. 257(1-2):
p. 9-45.
10. Markovic, N.M. and Ross, P.N., Surf. Sci. Rep. , 2002. 45(4-6): p. 121-
229.
11. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 45(2): p. 205-
231.
12. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 44(2): p. 239-
254.
13. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 44(1): p. 1-7.
14. Capon, A. and Parsons, R., J. Electroanal. Chem., 1975. 65(1, Part 1):
p. 285-305.
59
15. Beden, B., Lamy, C., Bewick, A., and Kunimatsu, K., J. Electroanal.
Chem., 1981. 121: p. 343-347.
16. Tacconi, N.R.d., Leger, J.M., Beden, B., and Lamy, C., J. Electroanal.
Chem., 1982. 134(1): p. 117-130.
17. Wolter, O., Willsau, J., and Heitbaum, J., J. Electrochem. Soc., 1985.
132(7): p. 1635-1638.
18. Beden, B., Bewick, A., and Lamy, C., J. Electroanal. Chem., 1983.
148(1): p. 147-160.
19. Kunimatsu, K. and Kita, H., J. Electroanal. Chem., 1987. 218(1-2): p.
155-172.
20. Lamy, C., Electrochim. Acta, 1984. 29(11): p. 1581-1588.
21. Pletcher, D. and Solis, V., Electrochim. Acta, 1982. 27(6): p. 775-782.
22. Lamy, C., Leger, J.M., Clavilier, J., and Parsons, R., J. Electroanal.
Chem., 1983. 150(1-2): p. 71-77.
23. Iwasita, T. and Vielstich, W., J. Electroanal. Chem., 1986. 201(2): p.
403-408.
24. Clavilier, J., J. Electroanal. Chem., 1987. 236(1-2): p. 87-94.
25. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J. Phys.
Chem. , 1993. 97(46): p. 12020-12029.
26. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J.
Electrochem. Soc., 1994. 141(7): p. 1795-1803.
27. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J. Phys.
Chem. , 1994. 98(2): p. 617-625.
28. Markovic, N.M., Gasteiger, H.A., Ross, P.N., Jiang, X., Villegas, I., and
Weaver, M.J., Electrochim. Acta, 1995. 40(1): p. 91-98.
29. Adzic, R.R., Tripkovic, A.V., and Markovic, N.M., J. Electroanal. Chem.,
1983. 150(1-2): p. 79-88.
30. Clavilier, J., Femandez-Vega, A., Feliu, J.M., and Aldaz, A., J.
Electroanal. Chem., 1989. 261(1): p. 113-125.
60
31. Clavilier, J., Fernandez-Vega, A., Feliu, J.M., and Aldaz, A., J.
Electroanal. Chem., 1989. 258(1): p. 89-100.
32. Fernandez-Vega, A., Feliu, J.M., Aldaz, A., and Clavilier, J., J.
Electroanal. Chem., 1989. 258(1): p. 101-113.
33. Schmidt, T.J., Behm, R.J., Grgur, B.N., Markovic, N.M., and Ross Jr.,
P.N., Langmuir, 2000. 16: p. 8159.
34. Beden, B., Kadirgan, F., Lamy, C., and Leger, J.M., J. Electroanal.
Chem., 1982. 142(1-2): p. 171-190.
35. Chen, S.P., Huang, T., Zhen, C.H., Zhang, Q., Gong, H., and Sun,
S.G., Spectrosc. Spectra Anal., 2003. 23(2): p. 273-275.
36. Beltowska-Brzezinska, M. and Heitbaum, J., J. Electroanal. Chem.,
1985. 183(1-2): p. 167-181.
37. Chen, S.L., Wu, B.L., and Cha, C.S., J. Electroanal. Chem., 1997.
431(2): p. 243-247.
38. Choi, J.H., Jeong, K.J., Dong, Y., Han, J., Lim, T.H., Lee, J.S., and
Sung, Y.E., J. Power Sources, 2006. 163(1): p. 71-75.
39. Choi, J.H., Park, K.W., Park, I.S., Kim, K., Lee, J.S., and Sung, Y.E., J.
Electrochem. Soc., 2006. 153(10): p. A1812-A1817.
40. McGrath, K.M., Electrochemical studies of membranes, catalysts, and
fuels for direct oxidation fuel cells University of Southern California, Ph.
D., Thesis, 2005.
41. Roychowdhury, C., Matsumoto, F., Zeldovich, V.B., Warren, S.C.,
Mutolo, P.F., Ballesteros, M., Wiesner, U., Abruna, H.D., and DiSalvo,
F.J., Chem. Mater., 2006. 18(14): p. 3365-3372.
42. Uhm, S.Y., Chung, S.T., and Lee, J.Y., Electrochem. Commun., 2007.
9(8): p. 2027-2031.
43. Watanabe, M., Shibata, M., and Motoo, S., J. Electroanal. Chem., 1985.
187(1): p. 161-174.
44. Zhang, L.J., Wang, Z.Y., and Xia, D.G., J. Alloys Compd., 2006. 426(1-
2): p. 268-271.
45. Casella, I.G. and Contursi, M., J. Electroanal. Chem., 2006. 588(1): p.
147-154.
61
46. Colmati, F., Antolini, E., and Gonzalez, E.R., Electrochim. Acta, 2005.
50(28): p. 5496-5503.
47. Colmati, F., Antolini, E., and Gonzalez, E.R., Appl. Catal., B, 2007.
73(1-2): p. 106-115.
48. Kelaidopoulou, A., Abelidou, E., and Kokkinidis, G., J. Appl.
Electrochem., 1999. 29(11): p. 1255-1261.
49. Chen, W., Kim, J., Sun, S.H., and Chen, S.W., Phys. Chem. Chem.
Phys., 2006. 8(23): p. 2779-2786.
50. Yang, Y.Y. and Sun, S.G., J. Phys. Chem. B, 2002. 106(48): p. 12499-
12507.
51. Arenz, M., Stamenkovic, V., Schmidt, T.J., Wandelt, K., Ross, P.N., and
Markovic, N.M., Phys. Chem. Chem. Phys., 2003. 5(19): p. 4242-4251.
52. Baldauf, M. and Kolb, D.M., J. Phys. Chem. , 1996. 100(27): p. 11375-
11381.
53. Larsen, R. and Masel, R.I., Electrochem. Solid-State Lett., 2004. 7(6):
p. A148-A150.
54. Li, X.G. and Hsing, I.M., Electrochim. Acta, 2006. 51(17): p. 3477-3483.
55. Beden, B., Lamy, C., and Leger, J.M., Electrochim. Acta, 1979. 24(11):
p. 1157-1166.
56. Kibler, L.A., El-Aziz, A.M., and Kolb, D.M., J. Mol. Catal. A: Chem. ,
2003. 199(1-2): p. 57-63.
57. Vela, M.E., Lezna, R.O., Detacconi, N.R., Arvia, A.J., Beden, B., Hahn,
F., and Lamy, C., J. Electroanal. Chem., 1992. 323(1-2): p. 289-302.
58. Chen, F.L., Chang, M.H., and Lin, M.K., Electrochim. Acta, 2007. 52(7):
p. 2506-2514.
59. Chen, W., Tang, Y., Bao, J., Gao, Y., Liu, C., Xing, W., and Lu, T., J.
Power Sources, 2007. 167(2): p. 315-318.
60. Chu, K.L., Shannon, M.A., and Masel, R.I., J. Electrochem. Soc., 2006.
153(8): p. A1562-A1567.
61. Ha, S., Adams, B., and Masel, R.I., J. Power Sources, 2004. 128(2): p.
119-124.
62
62. Ha, S., Dunbar, Z., and Masel, R.I., J. Power Sources, 2006. 158(1): p.
129-136.
63. Ha, S., Larsen, R., and Masel, R.I., J. Power Sources, 2005. 144(1): p.
28-34.
64. Ha, S., Rice, C.A., Masel, R.I., and Wieckowski, A., J. Power Sources,
2002. 112(2): p. 655-659.
65. Jayashree, R.S., Spendelow, J.S., Yeom, J., Rastogi, C., Shannon,
M.A., and Kenis, P.J.A., Electrochim. Acta, 2005. 50(24): p. 4674-4682.
66. Kang, S., Lee, J., Lee, J.K., Chung, S.Y., and Tak, Y., J. Phys. Chem.
B, 2006. 110(14): p. 7270-7274.
67. Kundu, A., Jang, J.H., Gil, J.H., Jung, C.R., Lee, H.R., Kim, S.H., Ku,
B., and Oh, Y.S., J. Power Sources, 2007. 170(1): p. 67-78.
68. Liu, Z.L., Hong, L., Tham, M.P., Lim, T.H., and Jiang, H.X., J. Power
Sources, 2006. 161(2): p. 831-835.
69. Miesse, C.M., Jung, W.S., Jeong, K.-J., Lee, J.K., Lee, J., Han, J.,
Yoon, S.P., Nam, S.W., Lim, T.-H., and Hong, S.-A., J. Power Sources,
2006. 162(1): p. 532-540.
70. Rice, C., Ha, R.I., Masel, R.I., Waszczuk, P., Wieckowski, A., and
Barnard, T., J. Power Sources, 2002. 111(1): p. 83-89.
71. Rice, C., Ha, S., Masel, R.I., and Wieckowski, A., J. Power Sources,
2003. 115(2): p. 229-235.
72. Waszczuk, P., Barnard, T.M., Rice, C., Masel, R.I., and Wieckowski, A.,
Electrochem. Commun., 2002. 4(7): p. 599-603.
73. Weber, M., Wang, J.T., Wasmus, S., and Savinell, R.F., J.
Electrochem. Soc., 1996. 143(7): p. L158-L160.
74. Yeom, J., Jayashree, R.S., Rastogi, C., Shannon, M.A., and Kenis,
P.J.A., J. Power Sources, 2006. 160(2): p. 1058-1064.
75. Yeom, J., Mozsgai, G.Z., Flachsbart, B.R., Choban, E.R., Asthana, A.,
Shannon, M.A., and Kenis, P.J.A., Sens. Actuators, B 2005. 107(2): p.
882-891.
76. Zhang, L.L., Lu, T.H., Bao, J.C., Tang, Y.W., and Li, C., Electrochem.
Commun., 2006. 8(10): p. 1625-1627.
63
77. Zhang, L.L., Tang, Y.W., Bao, J.C., Lu, T.H., and Li, C., J. Power
Sources, 2006. 162(1): p. 177-179.
78. Zhou, X., Xing, W., Liu, C., and Lu, T., Electrochem. Commun., 2007.
9(7): p. 1469-1473.
79. Zhu, Y.M., Ha, S.Y., and Masel, R.I., J. Power Sources, 2004. 130(1-2):
p. 8-14.
80. Zhu, Y.M., Khan, Z., and Masel, R.I., J. Power Sources, 2005. 139(1-2):
p. 15-20.
81. Chu, K.L., Gold, S., Subramanian, V., Lu, C., Shannon, M.A., and
Masel, R.I., J. Microelectromech. Syst., 2006. 15(3): p. 671-677.
82. Larsen, R., Ha, S., Zakzeski, J., and Masel, R.I., J. Power Sources,
2006. 157(1): p. 78-84.
83. Schmidt, T.J., Stamenkovic, V., Arenz, M., Markovic, N.M., and Ross,
P.N., Electrochim. Acta, 2002. 47(22-23): p. 3765-3776.
84. Igarashi, H., Fujino, T., Zhu, Y.M., Uchida, H., and Watanabe, M., CO
Tolerance of Pt alloy electrocatalysts for polymer electrolyte fuel cells
and the detoxification mechanism, 2001. 3(3): p. 306-314.
85. Toda, T., Igarashi, H., Uchida, H., and Watanabe, M., Enhancement of
the electroreduction of oxygen on Pt alloys with Fe, Ni, and Co, 1999.
146(10): p. 3750-3756.
86. Watanabe, M., Zhu, Y.M., and Uchida, H., Oxidation of CO on a Pt-Fe
alloy electrode studied by surface enhanced infrared reflection-
absorption spectroscopy, 2000. 104(8): p. 1762-1768.
87. Prakash, G.K.S., Krause, F., Viva, F.A., and Olah, G.A., Unpublished
work, 2008.
64
Chapter 3: Electrochemical reduction of carbon
dioxide
3.1 Chapter 3: Introduction
Carbon dioxide is a simple nonpolar molecule which is the end product
of the combustion of any organic molecule with air or oxygen. The reduction of
carbon dioxide has been envisioned as a possible answer to solve three main
global challenges. Two of them which are interconnected are; the diminishing
of fossil fuels, a non-renewable materials on the human timescale, and the
anthropogenic emission of carbon dioxide, which is tied to global climate
change (global warming). The third is the need for food for an ever growing
population. The possibility of reducing carbon dioxide emission on a large
scale concurrently will address many of these problems. The reduction of
carbon dioxide can produce simple molecules, some of which are useful as
fuels (such as methanol, as discussed in the introduction). These simple
molecules can subsequently be used as precursors for building more complex
molecules that can act as fuels of high molecular weight, which are currently
obtained from fossil fuels. The source for carbon dioxide should ultimately be
the atmospheric carbon dioxide; this will decrease the amount of carbon
dioxide in the atmosphere, alleviating climate change caused by human
factors. Moreover, the reduction of carbon dioxide is viewed as a way to store
energy. Electrical energy obtained from sources such hydro, wind, solar and
nuclear can be used to reduce carbon dioxide to fuels which can be stored
65
then recycled when needed, using a fuel cell or some other thermoelectric
devices. The end product will be carbon dioxide, closing the cycle. The
reduction of carbon dioxide by electrochemical means could be more
convenient on a large scale instead of just sequestration and storage. Carbon
dioxide can also be conveniently captured from point sources such as fossil
fuel burning power plants, aluminum plants, fermentation and cement
industries.
The first reports for the electrochemical reduction of carbon dioxide
appeared in the second half of the nineteenth century. It was demonstrated
that the reduction of carbon dioxide to formic acid could be achieved on
mercury or amalgam electrodes [1, 2, 3, 4, 5, 6]. However, research on the
electrochemical reduction of carbon dioxide had practically ceased after these
initial reports. Advances in electrochemical and spectroscopic instrumentation
allowed for a better understanding of the fundamental aspects of the
electrochemical reduction. This understanding, combined with the necessity to
solve some of the challenges that the world is facing, prompted researchers to
reinvestigate the electrochemical reduction of carbon dioxide which have
continued for the past two decades [7, 8].
3.1.1 Mechanism of electrochemical reduction of carbon dioxide
Because carbon dioxide is a gas, the electrochemical reduction
involves a gas dissolved in a solvent that reacts at a solid surface. Research
66
has been conducted using a wide variety of electrocatalysts, including metals,
non metals, semiconductors etc. as well as employing an array of conditions
such as different solvents, electrolytes and pressures. The different
combinations of electrodes and solvents yielded a variety of products. Thus,
metal electrodes were grouped based on the product obtained from reduction,
for example In, Pd and Sn yielded formic acid.
Reactions E
o
(V)
CO
2
(g) + 2H
+
(aq) + 2e
-
→ CO(g) + H
2
O(l) -0.103
CO
2
(g) + 2H
+
(aq) + 2e
-
→ HCOOH(aq) -0.110
2CO
2
(g) + 2H
+
(aq) + 2e
-
→H
2
C
2
O
4
-0.475
CO
2
(g) + 4H
+
(aq) + 4e
-
→ HCHO(aq) + H
2
O(l) -0.071
CO
2
(g) + 6H
+
(aq) + 6e
-
→ CH
3
OH(aq) + H
2
O(l) 0.030
CO
2
(g) + 8H
+
(aq) + 8e
-
→ CH
4
(aq) + 2H
2
O(l) 0.196
Table 3.1 Standard redox potential for the two-, four-, six- and eight-electron reduction of
carbon dioxide in water media, calculated from thermodynamic data. Reproduced
from [9, 10].
In Table 3.1, the products obtained from the direct reduction of carbon
dioxide and their standard potentials are shown [9, 10]. The potential reduction
is pH dependent and, as expected, at lower pH the processes are more
favorable but at the same time the solubility of carbon dioxide decreases,
diminishing the overall product yield. The electrolyte used in the media,
particularly in water, can enhance carbon dioxide solubility. However, the
electrolyte can also enhance the reduction of water (electrolysis) by the
67
electrode (based on the pH), which competes with the carbon dioxide
reduction. With the exception of the reduction to methanol and methane, all
the processes are less favorable than the reduction of H
+
to H
2
. However, the
difficulty of transferring multiple electrons reduces the practical feasibility of
reduction to methanol or methane.
There is consensus that the first step in the carbon dioxide reduction
involves the formation of the CO
2
-•
radical as indicated by Scheme 3.1 [8, 10,
11, 12, 13, 14, 15, 16, 17].
CO
2
CO
2
ads
+ e
-
Scheme 3.1 Radical formation for adsorbed carbon dioxide at the electrode.
The product obtained in the CO
2
reduction will then depend on the
subsequent steps; whether or not the radical anion is adsorbed on the metal
surface and the nature of the solvent (protic or aprotic) as presented by Hori
and co-workers [18]. Scheme 3.2 shows the steps that follow after the
formation of the radical anion. The mechanism identified as i) involves the
formation of formic acid. Such formation occurs when the radical anion is not
adsorbed on the metal electrode in the aqueous media. Water is the proton
donor, although other authors have indicated that the proton could be donated
from other species [10, 14, 15]. Examples of metals used for this pathway
include; Cd, Sn, In, Tl, Hg and Pb [19, 20, 21, 22, 23, 24]. Mechanism ii) is
analogous to i), but occurs in aprotic media. The radical anion can couple with
68
itself or react with a molecule of carbon dioxide to form the oxalate di-anion.
Metals that promote this mechanism include Pb, Tl and Hg [19, 20, 22, 23, 24].
The reaction proceeds by mechanism iii) when the radical anion is adsorbed
on the metal in aqueous media. The end product is carbon monoxide, and
metals that favor this mechanism are Au, Ag and Zn [24, 25, 26]. The
mechanism iiib) is a continuation to the one identified as iiia), with the
difference that the resulting product is a mixture of hydrocarbons (methane
and ethylene) and occurs when the metal electrode can also adsorb hydrogen,
as is the case for Pt, Pd, Cu and Ni [8, 27, 28, 29, 30]. The last possibility is
mechanism iv), which is similar to mechanism ii), but since the radical anion is
adsorbed on the metal electrode the coupling with another molecule of carbon
dioxide proceeds through an oxygen instead through the carbon, producing
carbon monoxide and eliminating the carbonate anion. Metals that facilitate
this pathway are Au, Cu, Zn, Cd, Sn and In [20, 31, 32, 33, 34, 35]. Scheme
3.2 not only shows the mechanism for the carbon dioxide reduction, but
reveals that the most common products for such reduction are carbon
monoxide, formic acid and oxalic acid as well as methane and ethylene.
Glyoxalic acid or the glyoxalate anion can be obtained by coupling two
oxalates [36]. Products like methanol or formaldehyde are not contemplated
from a mechanistic standpoint, and although their formation was reported on
metals like Cu, Ag and Ru [11, 14, 33, 37, 38, 39, 40, 41] albeit in negligible
amounts.
69
C
O O H
2
O
C
H
O O
HCOOH
e
-
OH
-
i)
C
O O C
O O
(COO
-
)
2
e
-
ii)
CO
2
C
O O
C
O O
H
2
O
C
HO O
OH
-
e
-
C
OH
-
O
iiia)
C
O
CH
2
iiib)
H
H
2e
-
CH
4
CH
2
H
2
C
OH
-
C
O O
C
O O
e
-
C
O
iv)
CO
2
C
O
O
CO
3
=
H
+
Scheme 3.2 Reaction scheme for the electrochemical reduction of CO
2
. i) Non-adsorbed
CO
2
-•
on a metal electrode in aqueous media. ii) Non-adsorbed CO
2
-•
on a
metal electrode in non-aqueous media. iii) CO
2
-•
adsorbed on a metal
electrode in aqueous media. iiib) CO
2
-•
adsorbed on a hydrogen-absorbing
metal electrode in aqueous media. iv) CO
2
-•
adsorbed on a metal electrode in
non-aqueous media. Reproduced from [18].
70
3.1.2 Electrochemical reduction of carbon dioxide to formic acid
In order to commercialize applications that utilize carbon dioxide
reduction, current densities above 100 mA cm
-2
are required. To achieve these
current densities not only must the mechanism be understood, but also the
related electrokinetics of the reaction. Carbon monoxide and formic acid are
the most feasible products to be obtained with high selectivity. Carbon
monoxide is of interest because, together with hydrogen, it forms syngas, used
in the methanol synthesis or the Fischer-Tropsch method for hydrocarbon
synthesis. Research has been conducted in our laboratory in order to reduce
carbon dioxide to CO for this purpose [42]. Formic acid is also of interest as it
is stable in the reaction cell, and can be used directly or reacted with methanol
to yield two molecules of methanol through the hydrogenation of methyl
formate according to Scheme 3.3.
HCOOH + CH
3
OH HCOOCH
3
+ 2H
2
2CH
3
OH
Scheme 3.3 Methanol formation from formic acid and methanol.
The complete proposed mechanism for the reduction of CO
2
to formic
acid is presented in reaction Scheme 3.4. This mechanism combines the
previously proposed and accepted mechanisms [8, 10, 21, 43] .
71
CO
2
(ads)
+ e
-
CO
2
ads
(ii)
CO
2
ads
+ H
2
O
HCO
2
ads
+ OH
-
(iii)
HCO
2
ads
+e
-
HCO
2
-
(iv)
CO
2
CO
2
ads
(i)
Scheme 3.4 Complete mechanism for the reduction of carbon dioxide to formate ion.
The first step is the adsorption of carbon dioxide over the electrode.
Although step (i) is generally ignored in electrochemical kinetics, in order to
have the reaction occurring, carbon dioxide must first dissolve in the media so
it can be in contact or near the electrode. It is here that the media and the
electrolyte play significant roles. In organic solvents the solubility of carbon
dioxide can be high. In water, carbon dioxide has a low solubility, particularly
at low pH. At neutral to basic pH, the solubility is increased by the formation of
bicarbonate in equilibrium, which can revert to carbon dioxide. Reaction
Scheme 3.4 shows the carbon dioxide equilibrium in water, while Figure 3.1
presents the distribution of species diagram (alpha plot) for carbonic acid.
Carbonate and bicarbonate solutions have proved to be good electrolytes for
the carbon dioxide electroreduction from the pH standpoint [18, 20, 21, 22, 23,
44, 45]. Bicarbonate solutions with concentrations ranging from 0.1 M to 0.5
M, which are the concentrations most commonly used, would give a pH of ca.
8.5.
72
CO
2
+ H
2
O
H
2
CO
3
H
2
CO
3
H
+
+ HCO
3
-
CO
3
=
+ H
+
HCO
3
-
Scheme 3.5 Carbon dioxide equilibrium in water.
Figure 3.1 Distribution of species diagram (alpha plot) for carbonic acid (H
2
CO
3
). αi is the
concentration of specie i divided the total concentration.
In step (ii) of the mechanism (Scheme 3.4), the adsorbed carbon
dioxide is reduced to the radical anion. In the mechanism presented by Hori,
showed in Scheme 3.1, the radical anion that is not adsorbed on the electrode
73
gets protonated to an HCOO
●
radical, which is then further reduced to formic
acid. However, other authors [8, 15] proposed that the radical anion formed
stays in the adsorbed state. The standard potential for the whole process
shown in Table 1.1, is -0.11 V against the SHE. Although negative, the
potential is low and it would be expected that the reduction would be facile.
However the standard reduction potential for the step (ii) in the mechanism
(Scheme 3.4) has been estimated to be -1.66 V or lower against the SHE [11].
This value is in part the source of the high overpotential required for the
reduction of carbon dioxide in water. It has been reported that the potential
must be below -1.4 V vs the SHE for the reaction to begin [32]. This
overpotential is also related to the energy necessary to break the linearity of
the carbon dioxide molecule. In fact Faradaic yields above 90 % for the
electroreduction of carbon dioxide to formic acid are seen at potentials of ca. -
1.6 V vs. the SHE [18, 20, 22]. This large overpotential also plays a role in the
adsorption of the radical anion intermediate. The metal electrode’s point of
zero charge (PZC) will provide the environment for the radical anion to be
adsorbed. The PZC is a particular potential value of the electrode at which the
charge of the electrode surface and the related interfacial potential are zero. At
potentials more positive than the PZC, the electrode surface is positively
charged and at potentials below the PZC the electrode surface is negatively
charged [46]. In terms of the carbon dioxide electroreduction mechanism,
metals with highly negative PZC will provide better conditions for the
74
adsorption of the radical anion and the subsequent protonation. On the other
hand, the HCOO
●
radical, once formed, will stay adsorbed because of the
unpaired electron until reduced to formate.
The best electrocatalytic metals for the electrochemical reduction of
carbon dioxide to formic acid were found to be In, Pb and Sn. They present
high hydrogen overpotentials, which assure a low competing reaction, and a
low PZC. The electrochemical measurements performed on these metals have
provided some kinetic information. The Tafel plots obtained from polarization
and linear sweep voltammetry yielded two linear regions for various metals
[15] indicating two consecutive steps in the reduction process as the
mechanism shows. These authors obtained values between 80 to 120 mV for
the first region of the Tafel slope, whereas the second region of the slope was
between 300 to 400 mV, which agrees with previous results [21, 43, 47].
Experiments employing different pressures of carbon dioxide demonstrated
that there was no effect on the first region, indicating zero order, while an
order of one was found for the second region [43, 47]. The reaction rate
dependence with the pH was also studied and found to be virtually
independent [12, 21], except that at high pH the reaction rates are slightly
diminished due to the formation of HCO
3
-
and the subsequent decrease of free
carbon dioxide. However with respect to which of the two electron transfer is
the rate determining step (rds) there seem to be no clear agreement. Kapusta
and co-workers indicated that the rds is the first electron transfer (step (ii)) at
75
more positive potentials assuming Langmuir conditions, while the change in
slope at more negative potentials indicates a change in the rds step [21]. They
also presented a two step mechanism in which steps (iii) and (iv) are
combined.
Figure 3.2 Tafel plots for the electrochemical reduction of carbon dioxide over different
metals in a pH 5.5 buffer solution. Reproduced from [15].
On the other hand, Eyring and coworkers [43, 48] assigned the rds for
the more positive potential region to the second electron transfer (step (iv))
76
because they observed an accumulation of the radical intermediate, while the
first electron transfer (step(ii)) was considered to be the rds in the region of
more negative potential. They also proposed a two step mechanism in their
later work where steps (ii) and (iii) were combined [47]. Figure 3.2 shows the
Tafel plot for different metals including In, Pb and Sn [15].
3.2 Chapter 3: Results and Discussion
3.2.1 Electrochemical measurements using a Sn electrode.
Cyclic voltammetry was employed to study and measure the
electrochemical parameters of the carbon dioxide reduction to formic acid over
a Sn electrode. As mentioned in the previous section, this experiment was
already performed by various researchers under a wide variety of conditions. It
was described that the protonation of the radical anion proceeds through a
proton donor. Water is the most common donor as shown in reaction iii on
Scheme 3.3. However it was also mentioned that the proton might come from
other species besides water, and the scheme has been written in a more
generalized way as the species BH donating the proton [11]. On the other
hand, carbonate and bicarbonate are also good electrolytes for the carbon
dioxide reduction. But in solutions of these electrolytes the pH is on the basic
side and the availability of protons is compromised. Therefore, in the research
presented, carbon dioxide reduction to formic acid was studied on metal Sn
powder, as the electrocatalyst, bound with Nafion® on a gas diffusion
77
electrode (GDE). The electrolyte used was sodium bicarbonate. The purpose
behind using Sn powder is to maximize the surface area of the electrode in
order to maximize the amount of formic acid produced as well as to maximize
current density and to have a proton donor in contact with the electrocatalyst
to improve the reaction, while at the same time using an electrolyte that
possesses a reasonable carbon dioxide solubility.
Cyclic voltammograms were obtained for three different kinds of
electrodes; a bulk Sn metal disc, Sn metal powder mixed with Nafion® over a
graphite disc and Sn metal powder mixed with Nafion® over a graphite paper
in a cell designed to have the carbon dioxide flow through the GDE.
Electrochemical parameters were derived from Tafel plots to compare the
three different environments. Figure 3.3 shows the voltammograms for the
bulk Sn metal disc electrode in a solution of 0.5 M NaHCO
3
saturated with Ar
or CO
2
. The two anodic peaks at ca. -0.5V and the cathodic peak at ca.
-1.1 V correspond to the adsorption and desorption of carbonate species, as
was previously observed and measured by a quartz crystal microbalance and
in situ infrared reflection absorption spectroscopy [25, 49]. On the cathodic
end of the voltammogram it can be seen that the current increases in absolute
value. On the voltammogram of Ar this increase is due to hydrogen evolution,
while when CO
2
is bubbled the current represents the reduction to formic acid.
The onset potential for the reduction reaction moves anodically 100 mV while
the current of the onset increases by a factor of 2.
78
Figure 3.3 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn disc electrode at 100 mV s
-1
.
Bubbled with Ar (solid line). Bubbled with CO
2
(dashed line).
The same experiment on Sn powder on a graphite disc is shown in
Figures 3.4 while Sn over graphite paper GDE is shown in Figure 3.5. The
main features are still present, although the peaks corresponding to the
adsorption/desorption of the carbonic species are less defined due to the
increase in the surface area, particularly in the case of GDE where the anodic
peaks shifted to a positive potential when carbon dioxide is bubbled through.
79
Figure 3.4 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn powder over a graphite disc
electrode at 100 mV s
-1
. Bubbled with Ar (solid line). Bubbled with CO
2
(dashed
line).
The reason for the decrease in the carbonic species peak might be that
the electrode surface is occupied by the reaction intermediates. Actually, for all
cases, a shift to more anodic potentials can be observed, but in the GDE case
this is very noticeable due to the high surface area. In terms of the current
arising from the reduction reaction, the onset potential of the curve moves
anodically even further than in the case of the Sn metal electrode. In the case
of Figure 3.4, the shift is ca. 300 to 400 mV while for Figure 3.5 is above 400
mV. At the same time, the current at the onset potential increases 3 to 4 times
80
that of the current when Ar is used. In Figure 3.5, the onset potential has been
displaced to ca. 0.4 V. A shift to more anodic potentials means a decrease in
the overpotential which implies that less energy is required to activate the
reaction.
Figure 3.5 Cyclic voltammogram in 0.5 M NaHCO
3
on a Sn powder over a graphite paper
GDE at 100 mV s
-1
. Bubbled with Ar (solid line). Bubbled with CO
2
(dashed line).
The current density reached with the Sn powder over the graphite disc
is 10 times higher than the one obtained with Sn over the GDE, which is likely
caused by the carbon dioxide flowing through the GDE, which can disrupt the
reduction. However, the current density of the GDE is still higher than that of
81
the metal electrode. Moreover, the Sn powder over the GDE presents the
highest surface area, yielding the highest total current.
From the voltammograms, the corresponding Tafel plots were obtained
as show in Figure 3.6. The linear region of the plot obeys the Tafel equation
(3.1) which relates the overpotential ( η) with the current density (j). The
overpotential is defined as the applied potential less the equilibrium potential,
i.e. when the current or current density is zero. The line’s intercept gives the
exchange current density (j
0
), which is directly related to the standard rate
constant and the activation energy and thus with the magnitude of the
overpotential needed in order to obtain a net current in the desired direction.
The slope gives the transfer coefficient, α, a measure of the energy barrier
symmetry [50], which is limited to values between 0 and 1, but commonly lies
between 0.3 and 0.7. Equation 3.3 represents the slope in Tafel equation for
the cathodic branch while equation 3.4 for the anodic branch as derived from
the Batler-Volmer equation [50].
) log( j b a ⋅ + = η (3.1)
) log(
3 . 2
0
j
F
RT
a
α
= (3.2)
F
RT
b
α
3 . 2 −
= (3.3)
82
F
RT
b
) 1 (
3 . 2
α −
−
= (3.4)
Figure 3.6 Tafel plots obtained from the corresponding voltammograms for bulk Sn metal
disc ( ■), Sn powder on a graphite disc ( ●) and Sn powder on a GDE ( ▲).
Figure 3.6 shows the Tafel plots for the three electrodes, and the data
obtained from these plots are presented in Table 3.2. The bulk metal Sn
electrode is the only one that can be compared with the literature data, since
Tafel plots, and their parameters, are typically obtained for bulk metal
electrodes as they provide the most reliable values. The two linear regions
obtained for the bulk Sn metal were in excellent agreement with the results
83
reported by Kapusta and Vassiliev (Figure 3.2) [15, 21].The other two
electrodes were measured here for comparison with the bulk metal electrode.
The standard potential obtained for the Sn metal disc is in good agreement
with the potential obtained from the literature, as shown in Table 3.1. The
exchange current densities differ by three orders of magnitude; although they
should be closer in value, the electrolyte solution used could account for part
of that difference. The slopes are in good agreement with those published
previously. Both Kapusta [21] and Vassiliev [15] reported a slope of ca.
120 mV for the region of lower overpotential while for the high overpotential
region the reported value was 300 to 350 mV. The low overpotential region
from this study agrees fairly well with the value reported, which correspond to
an α of 0.5. The high overpotential region overlaps with the adsorption peak of
the carbonic species present in the media, which introduces some difficulty in
the slope determination, therefore the value obtained is somewhat higher but
still of the same order.
The values obtained for the other two kinds of electrodes differ with the
bulk metal one, as expected for high surface area electrodes; however,
several observations can be made. The standard redox potential increases for
the powder on graphite disc and for the powder in GDE, indicating that the
reduction requires less energy for these electrodes. This also is seen in the j
0
value, which also increases, meaning a decrease in the activation energy,
facilitating the rate in the desired direction. In terms of the Tafel slopes for Sn
84
powder on the graphite disc, in both regions, the values are slightly higher
compared to the bulk metal, but the values are consistent with each other and
with the bulk metal and they maintain the same trend, meaning that there is no
substantial change in the reduction mechanism. Overall the values indicate a
more facile reduction on the powder electrodes.
Sn metal disc
Sn powder on
graphite disc
Sn powder on
GDE
b
c1
(mV) 116 180 185
b
c2
(mV) 430 480 458
j
o
(A cm
-2
) 1.2 x 10
-6
2.8 x 10
-5
1.6 x 10
-4
E
eq
(V) -0.10 0.030 0.053
Table 3.2 Tafel parameters obtained from the Tafel plots. b
c1
represents the slope for the
lower overpotential region and b
c2
for the high overpotential region.
3.2.2 Electrochemical reduction of carbon dioxide
Electrolysis of carbon dioxide to obtain formic acid was performed on
various metal powders on a GDE bound with Nafion® as described in the
previous section. The electrolysis was performed potentiostatically at
potentials between -0.8 to -2 V vs. the SHE on intervals of 0.2 V. The media
used was the same as for the voltammograms described in the previous
section while a small flow (4 mL min
-1
) of carbon dioxide was bubbled through
the solution and electrode during the entire experiment. The cell, described in
Chapter 6, allows the stream of gas to flow through the electrode and thus is in
85
close contact with the electrocatalyst during the entire reaction. The reduction
was performed potentiostatically since under a constant potential, the reaction
current is set by the system, showing how fast the reaction proceeds. The
reduction was performed on Sn, In, Pb and Cd. The former three are known to
have good Faradaic efficiency (ƒ) for formic acid. Cd has not been widely used
for carbon dioxide reduction but possesses one of the lowest PZC (-0.75 V)
[12] and a chemistry similar to Sn and In. The Faradaic efficiency (equation
3.4) is a means to assess how much of the energy introduced goes to the
desired product and how much is lost. It is the quotient between the obtained
amounts of the desired product, quantified by some analytical method, over
the theoretical amount calculated using Faraday’s law (equation 3.5). It can be
observed that the mass is proportional to the current, therefore the Faradaic
efficiency is also the quotient between the current that is producing the desired
product over the total current circulated through the cell.
l theoretica
m
m
f
exp
= (3.4)
nF
itMw
m = (3.5)
Figure 3.7 plots the Faradaic efficiencies of formic acid obtained from
the reduction of carbon dioxide versus the potential, while Table 3.3 gives the
corresponding values and the total current densities for each experiment.
86
Figure 3.7 Formic acid Faradaic yields vs. potential for the different electrocatalysts. (- ■-) Sn,
(- ●-) In, (- ▲-) Cd and (- ▼-) Pb.
The Faradaic efficiency never reaches 100% because energy is always lost as
heat and through side reactions. In the case of the carbon dioxide
electroreduction in water, the major side reaction is the formation of hydrogen.
For all of the different potentials shown for each metal, hydrogen was
detected. In some cases, carbon monoxide was detected in trace amounts as
well. At the same time, for potentials more negative than -1.4 V, the total
efficiency did not sum up to 100 %, indicating a small percentage of the
energy being dissipated as heat.
87
Pot
(V) vs.
SHE
Sn In Pb Cd
ƒ
(%)
j
(mA cm
-2
)
ƒ
(%)
j
(mA cm
-2
)
ƒ
(%)
j
(mA cm
-2
)
ƒ
(%)
j
(mA cm
-2
)
-0.8 0 1.3 0 1.5 0 1.1 0 1.2
-1.0 2 5.7 6 4.8 4 3.3 0 2.6
-1.2 45 12.4 17 6.9 12 4.3 6 3.2
-1.4 63 17.1 73 16.8 34 9.2 9 4.9
-1.6 70 27.3 85 25.1 62 11.3 48 9.7
-1.8 62 15.0 67 13.7 65 12.5 62 10.3
-2.0 54 11.0 52 9.0 46 7.3 33 6.8
Table 3.3 Faradaic efficiencies (ƒ) for formic acid and total current densities (j) for the
electrocatalysts at the potential applied for the electrolysis.
The values and trends obtained for the Faradaic efficiency versus the
potential agrees with the ones reported in the literature for the different
electrocatalysts [18, 22, 51, 52, 53, 54]. For the electrolysis performed under
similar conditions, Hori reported a value of ƒ of 88 % for Sn, 95 % for In,
97 % for Pb and 78 % for Cd at 5 mA cm
-2
[18] while at 5.5 mA cm
-2
values of
73 % for Sn, 95 % for In, 81 % for Pb and 66 % for Cd [52] were reported,
although these experiments were performed galavanostatically. Noda [54]
performed electrolysis potentiostatically at -1.4 V vs. the SHE in the same
electrolyte and reported the following values: ƒ of 63 % for Sn with a j of
3.8 mA cm
-2
, a value for ƒ of 69 % for In with a j of 2.3 mA cm
-2
, a value for ƒ
of 50 % for Pb with a j of 0.4 mA cm
-2
and a value of ƒ of 39% for Cd with a j of
0.8 mA cm
-2
. Koleli [22] carried out the electrolysis in a GDE-type electrode for
Sn and Pb, and the best Faradaic efficiencies obtained were 74 and 90 %,
88
respectively but for current densities below 1.5 mA cm
-2
. Higher current
densities can be seen only when the electrolysis is performed under pressure.
Hara [20, 55] reported, in experiments conducted at 30 atm of carbon dioxide,
current densities between 90 and 100 mA cm
-2
for Sn, In and Pb with Faradaic
efficiencies above 90 %.
3.3 Chapter 3: Conclusion
The results presented show that an electrode of a powder catalyst
bound with Nafion® on a GDE gives good current densities for the formic acid
formation under potentiostatic conditions from carbon dioxide. Also, it is
interesting to note that Cd, which presents the lowest PZC compared with the
other metals, does not give the highest Faradaic efficiency. Although the PZC
was previously mentioned as an important condition, the results on Cd show
that this is not the most important consideration. The goal of improving the
current density of the reduction of carbon dioxide was achieved.
A fact that has to be mentioned is the preference of Nafion® for cations
like Na
+
for which it has shown more affinity [56] (Chapter 5). Although this fact
might present a problem when using the electrode employed in the
experiments for the reduction of carbon dioxide, it was not a concern in our
study since these experiments were carried over a short period of time,
typically less than 2 hours. Determinations done in our laboratory shown that
more than 3 to 4 hours are needed for a full exchange of the H
+
with Na
+
. The
89
H
+
on the Nafion® can be regenerated by washing the electrode with sulfuric
acid. Nevertheless, it cannot be determined conclusively that the Nafion®
used to bind the catalyst powder to the GDE is a key factor for improving the
reaction rate. To determine Nafion’s® importance in improving the reaction
rate more experiments need to be performed, particularly ones that test the
stability of the electrode, including the aforementioned affinity.
90
3.4 Chapter 3: References
1. Beketov, N.N., Zh. Russ. Fiz. Khim. Ova, 1869. 1: p. 33.
2. Goodridge, F., Lister, K., and Guruswamy, V., J. Electrochem. Soc.,
1978. 125(3): p. C176-C176.
3. Klobukov, N., J. Prakt. Chem., 1887. 34: p. 62.
4. Lieber, A., Monatsh. Chem., 1895. 16: p. 211.
5. Lieber, A., Monatsh. Chem., 1897. 18: p. 528.
6. Royer, E., C.R. Acad.Sci., 1870. 70: p. 73.
7. Gattrell, M., Gupta, N., and Co, A., J. Electroanal. Chem., 2006. 594(1):
p. 1-19.
8. Jitaru, M., Lowy, D.A., Toma, M., Toma, B.C., and Oniciu, L., J. Appl.
Electrochem., 1997. 27(8): p. 875-889.
9. Lamy, E., Nadjo, L., and Saveant, J.M., J. Electroanal. Chem., 1977.
78(2): p. 403-407.
10. Sullivan, B.P., Krist, K., and Guard, H.E., eds. Electrochemical and
Electrocatalytic Reactions of Carbon Dioxide. 1993, Elsevier:
Amsterdam.
11. Ayers, W.M., Spec. Publ.-R. Soc. Chem., 1994(153): p. 365-374.
12. Bagotzky, V.S. and Osetrova, N.V., Russ. J. Electrochem., 1995. 31(5):
p. 409-425.
13. Beden, B., Bewick, A., Razaq, M., and Weber, J., J. Electroanal.
Chem., 1982. 139(1): p. 203-206.
14. Chaplin, R.P.S. and Wragg, A.A., J. Appl. Electrochem., 2003. 33(12):
p. 1107-1123.
15. Vassiliev, Y.B., Bagotsky, V.S., Osetrova, N.V., Khazova, O.A., and
Mayorova, N.A., J. Electroanal. Chem., 1985. 189(2): p. 271-294.
16. Vassiliev, Y.B., Bagotzky, V.S., Khazova, O.A., and Mayorova, N.A., J.
Electroanal. Chem., 1985. 189(2): p. 295-309.
91
17. Vassiliev, Y.B., Bagotzky, V.S., Osetrova, N.V., and Mikhailova, A.A., J.
Electroanal. Chem., 1985. 189(2): p. 311-324.
18. Hori, Y., Wakebe, H., Tsukamoto, T., and Koga, O., Electrochim. Acta,
1994. 39(11-12): p. 1833-1839.
19. Gennaro, A., Isse, A.A., Severin, M.G., Vianello, E., Bhugun, I., and
Saveant, J.M., J. Chem. Soc., Faraday Trans., 1996. 92(20): p. 3963-
3968.
20. Hara, K., Kudo, A., and Sakata, T., J. Electroanal. Chem., 1995. 391(1-
2): p. 141-147.
21. Kapusta, S. and Hackerman, N., J. Electrochem. Soc., 1983. 130(3): p.
607-613.
22. Koleli, F., Atilan, T., Palamut, N., Gizir, A.M., Aydin, R., and Hamann,
C.H., J. Appl. Electrochem., 2003. 33(5): p. 447-450.
23. Mizuno, T., Ohta, K., Sasaki, A., Akai, T., Hirano, M., and Kawabe, A.,
Energy Sources, 1995. 17(5): p. 503-508.
24. Tomita, Y., Teruya, S., Koga, O., and Hori, Y., J. Electrochem. Soc.,
2000. 147(11): p. 4164-4167.
25. Arihara, K., Kitamura, F., Ohsaka, T., and Tokuda, K., J. Electroanal.
Chem., 2001. 510(1-2): p. 128-135.
26. Hori, Y., Ito, H., Okano, K., Nagasu, K., and Sato, S., Electrochim. Acta,
2003. 48(18): p. 2651-2657.
27. Hoshi, N. and Hori, Y., Electrochim. Acta, 2000. 45(25-26): p. 4263-
4270.
28. Hoshi, N., Kawatani, S., Kudo, M., and Hori, Y., J. Electroanal. Chem.,
1999. 467(1-2): p. 67-73.
29. Nakato, Y., Yano, S., Yamaguchi, T., and Tsubomura, H., Denki
Kagaku, 1991. 59(6): p. 491-498.
30. Noda, H., Ikeda, S., Oda, Y., and Ito, K., Chem. Lett., 1989(2): p. 289-
292.
31. Ikeda, S., Hattori, A., Ito, K., and Noda, H., Electrochem., 1999. 67(1):
p. 27-33.
92
32. Ikeda, S., Takagi, T., and Ito, K., Bull. Chem. Soc. Jpn., 1987. 60(7): p.
2517-2522.
33. Ishimaru, S., Shiratsuchi, R., and Nogami, G., J. Electrochem. Soc.,
2000. 147(5): p. 1864-1867.
34. Kaneco, S., Iiba, K., Hiei, N., Ohta, K., Mizuno, T., and Suzuki, T.,
Electrochim. Acta, 1999. 44(26): p. 4701-4706.
35. Ohta, K., Kawamoto, M., Mizuno, T., and Lowy, D.A., J. Appl.
Electrochem., 1998. 28(7): p. 717-724.
36. Eggins, B.R., Ennis, C., McConnell, R., and Spence, M., J. Appl.
Electrochem., 1997. 27(6): p. 706-712.
37. Frese, K.W. and Leach, S., J. Electrochem. Soc., 1985. 132(1): p. 259-
260.
38. Hori, Y., Denki Kagaku, 1990. 58(11): p. 996-1002.
39. Hori, Y., Kikuchi, K., Murata, A., and Suzuki, S., Chem. Lett., 1986(6):
p. 897-898.
40. Hori, Y., Murata, A., and Takahashi, R., J. Chem. Soc., Faraday Trans.,
1989. 85: p. 2309-2326.
41. Hori, Y., Takahashi, I., Koga, O., and Hoshi, N., J. Mol. Catal. A: Chem.
, 2003. 199(1-2): p. 39-47.
42. Prakash, G.K.S., Suresh, P., Viva, F., and Olah, G.A., J. Power
Sources, 2008. 181(1): p. 79-84.
43. Paik, W., Andersen, T.N., and Eyring, H., Electrochim. Acta, 1969.
14(12): p. 1217.
44. Azuma, M., Hashimoto, K., Watanabe, M., and Sakata, T., J.
Electroanal. Chem., 1990. 294(1-2): p. 299-303.
45. Hori, Y., Murata, A., Kikuchi, K., and Suzuki, S., J. Chem. Soc., Chem.
Commun. , 1987(10): p. 728-729.
46. Bockris, J.O.M. and S.U.M., K., Surface Electrochemistry. 1993, New
York: Plenum
47. Ryu, J., Andersen, T.N., and Eyring, H., J. Phys. Chem. , 1972. 76(22):
p. 3278-&.
93
48. Andersen, T.N., Hong, M., Paik, W., and Eyring, H., Ab. Pap. Am.
Chem. Soc., 1969: p. 112.
49. Zhou, A.H., He, D.L., Xie, N.X., Xie, Q.J., Nie, L.H., and Yao, S.Z.,
Electrochim. Acta, 2000. 45(24): p. 3943-3950.
50. Bard, A.J. and Faulkner, L.R., Electrochemical Methods. Fundamentals
and Applications. First ed. 1980: J. Wiley & Sons
51. Azuma, M., Hashimoto, K., Hiramoto, M., Watanabe, M., and Sakata,
T., J. Electrochem. Soc., 1990. 137(6): p. 1772-1778.
52. Hori, Y., Kikuchi, K., and Suzuki, S., Chem. Lett., 1985(11): p. 1695-
1698.
53. Koleli, F., Yesilkaynak, T., and Balun, D., Fresenius Environ. Bull.,
2003. 12(10): p. 1202-1206.
54. Noda, H., Ikeda, S., Oda, Y., Imai, K., Maeda, M., and Ito, K., Bull.
Chem. Soc. Jpn., 1990. 63(9): p. 2459-2462.
55. Hara, K., Kudo, A., Sakata, T., and Watanabe, M., J. Electrochem.
Soc., 1995. 142(4): p. L57-L59.
56. Okada, T., Moller-Holst, S., Gorseth, O., and Kjelstrup, S., J.
Electroanal. Chem., 1998. 442(1-2): p. 137-145.
94
Chapter 4: Pt supported over Carbon mono-fluoride
(CFx) as a catalyst for the Oxygen Reduction Reaction
(ORR)
4.1 Chapter 4: Introduction
As was described in the introduction, the reaction that occurs on the
cathode for the PEMFC using any fuel is the reduction of oxygen to water
(Table 1.2), or in the case of the AFC is the reduction of oxygen to hydroxide
(Table 1.3). Pt is the most common catalyst used in the oxygen reduction
which is commonly dispersed over carbon powder (Pt/C), increasing the
surface area of the catalyst [1, 2].
It was mentioned in Section 2.2.2 that it is important to have the right
amount of water at the cathode, referred to as water management. Water
management is a very important factor in fuel cells [3, 4, 5, 6, 7, 8, 9]. The
amount of water that will reside in the anode or cathode compartments is
crucial for the redox reactions, because, water is necessary for the overall
process; besides its involvement in the reaction, it is essential for ion transport,
as well as avoiding membrane dehydration, which would result in a decrease
of performance. Thus the presence of water affects the overall cell
performance. In the hydrogen fuel cell both hydrogen and oxygen have to be
wet (humidified). Furthermore, the water produced by the oxygen reduction it
is not sufficient to keep the whole cell humid. In the DMFC, on the other hand,
it is not necessary to supply the oxygen humid because an aqueous methanol
95
solution (generally 1 M) is circulated through the anode. As the methanol
solution is aqueous, there is ample water to cross through the membrane to
the cathode. In any case, water management is necessary to avoid what is
known as compartment flooding, or cathode flooding in the DMFC. This is
achieved by choosing some of the fuel cell materials with particular properties,
as exemplified in Chapter 2 (Section 2.2.2), specifically gas diffusion
electrodes (GDE) that contain a hydrophobic Teflon coating that repels water.
The Teflon coating amount can be varied, and although conductivity will
decrease, controlling the water input will keep the fuel cell running.
Fluorocarbons have the property of good oxygen solubility [10, 11, 12]
and among their potential uses they have been proposed as blood substitutes
[11]. Polycarbon mono-fluoride (CFx) is a partially fluorinated graphite used in
Li ion battery cathodes [13, 14, 15]. At low fluorine content this fluorinated
graphite still maintains adequate electrical conductivity. It has already been
mentioned that the cathode catalyst Pt is commonly used dispersed over
graphite in order to enhance the surface area and thus the catalytic activity.
The research presented in this chapter focuses on the preparation of Pt
supported over carbon mono-fluoride for use as a fuel cell cathode catalyst.
Properties of the carbon mono-fluoride employed are shown in Table 4.1
together with Vulcan XC-72 [16]. The latter is the state of the art graphite
currently used as the support for Pt. In Table 4.1 it can be seen that the
difference in the electrical conductivity is small and the surface areas are
96
comparable. At the same time the incorporation of fluorine on the graphite can
improve the oxygen adsorption, facilitating the reduction reaction (enhancing
mass transport kinetics). In order to assess the effect of fluorine on the
graphite, electrochemical measurements were performed, as well as single
fuel cell tests, and compared with commercial Pt black and Pt supported on
graphite (Pt/C).
Product
Advance Research
Chemicals - 2010
Vulcan XC-72
Carbon Source Carbon Black Carbon Black
Total Fluoride (%) 9 - 11 0
Color Black Black
True Density (g ml
-1
) 1.9 1.7 – 1.9
Bulk Density (g ml
-1
) 0.1 0.02-0.55
Median Particle Size ( μ ) <1 <1
Surface Area (m
2
g
-1
) 170 250
Electrical Resistivity ( Ω cm) < 10 >2
Table 4.1 Physical properties of Advance Research Chemicals 2010 carbon mono-fluoride
and Vulcan XC-72®.
4.1.1 The oxygen reduction reaction (ORR)
The groundbreaking work for the oxygen equilibrium, i.e. oxygen
evolution at the anode, as well as the oxygen reduction reaction (ORR) at the
cathode, were carried out by Damjanovic [17, 18, 19, 20, 21, 22] and Bockris
97
[18, 20, 21, 22, 23, 24] in the 1960s, particularly on a Pt single crystal. They
showed the difficulty of studying the oxygen equilibrium as well as the
complexity of the reaction. Scheme 4.1 presents the simplified version for the
oxygen reduction [25, 26]. The scheme shows that the oxygen can be reduced
directly to water through the so called “direct” 4e
-
reduction (k
1
) or through the
formation of a hydrogen peroxide intermediate (k
2
and k
3
). The hydrogen
peroxide can decompose chemically to oxygen (k
4
) or can desorb from the
electrode (k
5
). In acidic media the direct reduction of oxygen to water is the
most likely pathway, while in basic media it is possible to obtain hydrogen
peroxide [21]. The pathway identified as k
1
, although called direct, is a multi-
electron reaction that may include a number of elementary steps.
O
2
O
2ads
H
2
O
2ads
H
2
O
2
H
2
O
2e
-
2e
-
4e
-
k
1
k
2
k
3
k
4
k
5
Scheme 4.1 Pathways for the electrochemical reduction of oxygen. Reproduced from [25,
26].
98
The reduction reaction occurs both on surfaces that are oxide free and
oxide covered [17, 27, 28]. The exchange current density (j
0
) for the reduction
reaction obtained by Damjanovic and Bockis is 10
-10
A cm
-2
[18].
Figure 4.1 Effect of solution purity on O
2
evolution. ( ○) no pre-electrolysis. ( ●) 40 h of pre-
electrolysis. ( ▲) 60 h of pre-electrolysis. Reproduced from [18].
This low value of j
0
represents the sluggishness of the reduction compared to
the fuel oxidation. Moreover, their experiments showed that in order to obtain
99
Tafel lines below 10
-8
A cm
-2
, which allowed them to measure j
0
, extended
periods of solution purification were required. Figure 4.1 shows Tafel plots and
the effect of pre-electrolysis purification on the Tafel lines at low current
densities. Only with long periods (>60 h) of anodic pre-electrolysis did the
Tafel plot show an asymptotic slope approaching 1.23 V.
Figure 4.2 Tafel plot in acid solution. ( ○) Stationary Pt wire electrode in HClO
4
(pH 1). (+, x)
Rotating Pt disk electrode in H
2
SO
4
(pH 1). Reproduced from [19].
Another important observation derived from such studies was a change
in the Tafel slope on the Tafel plots. This change can be observed in Figure
4.2. The slope changed from a value of RT F
-1
at low current densities to a
value of 2RT F
-1
at high current densities with respect to ln j. This was
100
interpreted as a change between kinetic under Temking conditions at low
current density to Langmuir conditions at high current density. In both stages,
it was shown that the rate determining step is the first electron transfer to the
adsorbed oxygen [17, 19, 24, 29]. At high current densities the oxygen
coverage is negligible and the kinetics proceed through Langmuir conditions,
as the voltage reaches equilibrium voltage and as the current density
decreases, the oxygen coverage begins to accumulate causing the kinetics to
change to a Temkin condition.
4.2 Chapter 4: Results and Discussions
4.2.1 Catalyst surface analysis and quantification
Surface characterization of Pt dispersed over carbon mono-fluoride
(Pt/CFx) was performed by similar methods used for the catalyst
characterization of the formic acid fuel cell (Section 2.2.1). Since in this
catalyst system the Pt is supported over carbon particles, the Transmission
Electron Microscopy (TEM) images were also valuable. Figure 4.3 presents
the TEM images of Pt/CFx. Some nucleation of the Pt over the CFx can be
seen, but overall the Pt is well dispersed. XPS and EDS analysis were
performed in order to confirm the presence of Pt and to quantify the amount on
the catalyst. Only Pt, C and F were present in the XPS spectrum while only Pt
is detectable in the EDS besides the Ni used for the sample support grid. Pt
black was used as the standard in the EDS for quantification purpose.
101
(A)
(B)
Figure 4.3 TEM images of Pt/CFx at 20K X magnification
102
From XPS and EDS analyses the amount of Pt calculated to be bound
to the carbon support (CFx) was 30% by weight. This amount of Pt on the
catalyst was in agreement with the surface area measured by cyclic
voltammetry, vide infra.
4.2.2 Electrochemical Measurements I. Cyclic Voltammetry
Electrochemical measurements were performed on Pt black, Pt/C and
Pt/CFx powders mixed with Nafion® on a graphite disc electrode in the same
manner as performed for the DFAFC catalyst (Section 2.2.1). Cyclic
voltammograms of the three catalysts were performed in aqueous 0.1M H
2
SO
4
previously purged with Ar for 30 min. The hydrogen adsorption charge was
used to calculate the real area of the catalyst based on the value of
210 µC cm
-2
[30] and current (i) was then converted to current density (j).
After the voltammograms were measured under Ar, fresh solutions
were introduced into the cell, saturated with oxygen for 30 min, and then the
voltammograms were performed at 5 mV s
-1
with an oxygen flow rate of 8 mL
min
-1
. The voltammograms obtained are shown in Figure 4.4. The reverse
scan was used to derive the Tafel plots (E vs log j) and calculate the Tafel
parameters from the linear regions as shown in Figure 4.5.
103
Figure 4.4 Cyclic Voltammogram of catalyst in aqueous 0.1 M H
2
SO
4
at 5 mV s
-1
with an O
2
flow rate of 8 mL min
-1
. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx.
The Tafel slopes for low (b
c1
) and high (b
c2
) current density, the
exchange current density (j
0
), the equilibrium potential (E
eq
), and the onset
potential obtained from Figure 4.4 and 4.5 are presented in Table 4.2. The
voltammograms for the three catalysts show that the absolute current density
value for the oxygen
reduction peak is higher for Pt/CFx, indicating a higher
activity for the process. The cathodic peak potential for Pt/CFx is more anodic
than Pt/C and Pt, with values of 0.41, 0.40 and 0.36 V, respectively. This shift
to anodic potentials also indicates a higher catalytic activity for Pt/CFx. This
104
activity is also demonstrated by the onset potential, where Pt/CFx gives a
value that is ca. 100 mV more anodic than the other two catalysts. Another
characteristic of the voltammograms is that the peaks in the hydrogen
adsorption/desorption region are better defined for Pt/CFx and Pt as opposed
to Pt/C, indicating that the polycrystalline structure of the Pt on CFx is more
similar to the Pt black than to the Pt on carbon.
Figure 4.5 Tafel plot obtained from the graphs in Figure 4.4. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-)
Pt/CFx.
105
b
c1
(mV)
low j
b
c2
(mV)
high j
j
o
(A cm
-2
)
E
eq
(V)
Onset
potential (V)
Epc
(V)
Pt -51 -153 1.50x10
-0.7
1.28 0.87 0.36
Pt/C -46 -133 3.30x10
-0.7
1.23 0.87 0.40
Pt/CFx -48 -148 4.80x10
-07
1.25 1.00 0.46
Table 4.2 Tafel parameters, onset potential, and cathodic peak potential extracted from
Figures 4.4 and 4.5.
As was described in Section 4.1.1, Damjanovic, Bockris and co-workers
observed a change in the Tafel slope from a value of RT F
-1
at low current
density (zone 1) under Temkin conditions to a value of 2RT F
-1
at high current
density (zone 2) under Langmuir conditions [17, 29]. These were found to be
59 mV and 118 mV, respectively. The values obtained from the Tafel plot for
the slope in this study are in good agreement with the values obtained by
Damjanovic and co-workers, which is surprising given that the electrodes used
are different in morphology. Damjanovic and co-workers demonstrated that the
mechanism proposed for crystalline Pt also applies to polycrystalline Pt [31].
The values measured can depend on the experimental conditions, particularly
for the Tafel slope [32]. The electrodes used in this study, analogous to the
ones used in Chapter 2, are composed of polycrystalline Pt unsupported and
supported on carbon mixed with the ionomer over a conducting substrate
(graphite). This morphology produces a three-dimensional porous high surface
area, where the catalyst and substrate are in close contact with highly acidic
Nafion®. The aforementioned structure can account for the small difference
106
between the values published and the values measured in this study,
particularly at low current densities. In terms of the exchange current density,
the values differ because purification by pre-electrolysis was not performed, as
the intent was to measure the parameters to directly compare the performance
of the three catalysts. From the measured Tafel parameters, the following
observations for the catalysts can be made: 1) the exchange current densities
are small, showing that the process is charge-transfer controlled, but at the
same time the order Pt/CFx > Pt/C > Pt for the exchange current density
values indicates the order in the reaction velocity. 2) The Tafel slope for the
low and high exchange current density zones are similar for the three
catalysts, particularly in zone 1, showing that the electron transfer processes is
the same for the three catalysts, and based on the agreement with the values
published, a similar reaction mechanism. This agreement is important for
Pt/CFx because the electrochemical study shows that the fluorine-on-carbon
does not affect the electron transfer step or the general mechanism of the
oxygen reduction.
4.2.3 Electrochemical measurements II. Electrochemical Impedance
Spectroscopy (EIS)
Electrochemical Impedance Spectroscopy (EIS) measurements were
performed on the same electrodes used for the voltammograms in 0.1 M
H
2
SO
4
after being saturated with oxygen for 30 min. The frequency was
107
scanned from 10
5
Hz to 0.01 Hz, with an AC amplitude of 10 mV. The DC
voltage was set at the open circuit potential (OCP), or rest potential, which lies
between 0.9 and 1.0 V vs. SHE for all three catalysts. This rest potential was
already observed by Damjanovic and co-workers [24], wherein they showed
that said potential depends on the oxygen pressure and is controlled by a
reaction involving dissolved oxygen as a reactant.
Figure 4.6 Nyquist Plot for the three catalysts at rest potential with an O
2
flow of 8 mL min
-1
in
the frequency range 10
5
to 0.01Hz. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx.
108
Figure 4.7 Kinetic control close-up for the Nyquist Plot of Figure 4.6 for the three catalysts at
rest potential with an O
2
flow of 8 mL min
-1
. (- ■-) Pt black, (- ●-) Pt/C, (- ▲-) Pt/CFx.
The value obtained in this study agrees with the reported rest potential.
Figure 4.6 shows the Nyquist plot for the three different catalysts, while Figure
4.7 shows a close up for the high frequency zones. The best-fit lines in both
graphs correspond to result obtained with the equivalent circuit of Figure 4.8.
The fitted values, including the fitting parameter Chi
2
, are provided in Table
4.3.
109
R
s
R
ct
C
dl
C
//
Chi
2
Pt -2.88 11.84 2.94x10
-7
2.25 x10
-7
0.0015
Pt/C -9.20 23.93 1.66 x10
-7
7.87 x10
-9
0.0008
Pt/CFx 9.63 11.58 9.13 x10
-7
1.13 x10
-7
0.003
Table 4.3 Values obtained for the elements of the equivalent circuit of Figure 4.8 from fitting
the experimental Nyquist Plot for the three different catalysts. The value of Chi
2
fitting parameter is also included.
Figure 4.8 Equivalence circuit used to fit the Nyquist plot. R
s
: solution resistance, C
//
: parallel
capacitance, R
ct
: charge transfer resistance, C
dl
: double layer capacitance, W:
Warburg impedance.
The Nyquist plot for the three different catalysts presents a typical
kinetic-diffusion control. At mid-frequencies, a typical 45
o
angle corresponds to
diffusion control and a transition to capacitive behavior at low frequencies,
which is well fitted by the Warburg impedance. For Pt and Pt/C, the capacitive
behavior is higher than that for Pt/CFx, giving 80
o
angles for Pt and Pt/C and
110
70
o
for Pt/CFx. In the kinetic control zone the charge transfer resistance (R
ct
)
is similar for Pt and Pt/CFx, shown by the radius of the semicircle, while for
Pt/C it is much higher, in this respect the Pt/CFx behaves likes Pt. A
capacitance parallel (C
//
) to the solution resistance (R
s
) was introduced in our
equivalent circuit. This element is not commonly found in the simplest
equivalent circuit representations of an electrochemical cell, but can be seen
as another capacitive arc at very high frequency end in Figure 4.7. Although
this arc is mainly visible for Pt/CFx, including this capacitance in the circuit
improved the fitting for the three catalysts in the Nyquist plot (Figure 4.6) by
the decreasing value of the Chi
2
fitting parameter. One possible explanation
for this capacitance is the presence of a double-layer capacitance arising from
the counter electrode, since the electrode area size might not be large enough
when compared to the working electrode area. Another observation that must
be addressed is the fact that for Pt and Pt/C there is negative real impedance
(Z’) at very high frequency. This phenomenon is called electrochemical
oscillation [33] which was observed in several systems including fuel cells, and
one possible explanation for this is: because the purities of the solutions in the
experiment are high, it is believed that the phenomenon in this case could
arise from the Nafion® mixed with the catalyst. The presence of this
phenomenon could mean that the values for the ohmic resistance (R
s
) are
higher than the ones obtained by fitting the data. However the EIS plots and
the fitted values support the findings of the Tafel parameters. The performance
111
of Pt/CFx is similar to Pt and better than Pt/C, as shown in the plot. This
improvement is also seen in the value of the charge transfer resistance
between Pt/CFx and Pt.
4.2.4 Single cell Fuel Cell testing
After assembly and conditioning of the fuel cell and before performing
numerous polarization tests, the cell resistance was measured and yielded
values of the same order for the three different catalysts. These values
indicate that the Pt/CFx does not behave differently from Pt and Pt/C, and the
MEA does not present any problem after the assembly process of the cell.
4.2.4.1 Single cell DMFC polarization tests
MEAs prepared with the three catalysts were tested with methanol (1M)
at 90
o
C, 60
o
C and 30
o
C, with O
2
flow rates ranging from 1.2 L min
-1
to 40 mL
min
-1
and air flow rates ranging from 1.3 L min
-1
to 200 mL min
-1
. The best
performance in terms of polarization and power density plots for Pt black as
the cathode catalyst is obtained for a MEA prepared with Pt/Ru in the anode
feed with 1M methanol at 90
o
C and 1.23 L min
-1
of oxygen flow at the
cathode. When the other two catalysts were tested under the same conditions,
lower performances were observed, as expected, with the one for Pt/C being
higher than that for Pt/CFx. However, when experiments were performed with
lower oxygen flow rates at the cathode the performance of Pt/CFx improved
112
and below a given flow rate was even higher than Pt/C. At 700 mL min
-1
the
Pt/C performance was still better than Pt/CFx, but when the oxygen flow was
decreased to 200 mL min
-1
, Pt/CFx showed better performance than Pt/C, and
this trend was maintained for the other two flow rates of 60 and 40 mL min
-1
respectively. Figures 4.9 and 4.10 illustrate the performance of the three
catalysts at 200 and 60 mL min
-1
of oxygen flow rate at the cathode with 1M
methanol at 90
o
C. It is interesting to note first that Pt/CFx performance is
higher than that of Pt/C, even when there is less Pt on the catalyst, and
second more importantly, the power of Pt/CFx is ca. half that of pure Pt, even
when the amount of Pt is 30 % in the catalyst. Also, the performance of Pt/CFx
decreases to a lesser extent than the performance of Pt due to the decrease in
oxygen flow.
113
Figure 4.9 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-) with
1M methanol at 90
o
C and O
2
flow at 200 mL min
-1
.
Figure 4.10 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-) with
1M methanol at 90
o
C and O
2
flow at 60 mL min
-1
.
114
In order to better visualize the results for the three catalysts, the data
was normalized by the amount (in mg) of Pt on each catalyst. These results
are presented in Figures 4.11 and 4.12. The resulting current density and the
power density obtained per mg of Pt are higher for Pt/CFx than for the other
two catalysts. As previously mentioned, the same trend was obtained for the
lowest oxygen flow rate tested (40 mL min
-1
). At that flow rate, the
performance of Pt and Pt/CFx are comparable even when the relative amount
of Pt is ignored.
Figure 4.11 Normalized polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-)
with 1M methanol at 90
o
C and O
2
flow at 200 mL min
-1
.
115
Figure 4.12 Normalized polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-)
with 1M methanol at 90
o
C and O
2
flow at 60 mL min
-1
.
Measurements at 60
o
C and 30
o
C for Pt/CFx showed the same
performance as Pt/C, even at oxygen flow rates of 1.2 L min
-1
, in contrast with
90
o
C, and again showed better performance below 700 mL min
-1
. Figures
4.13 and 4.14 shows the performance for the three catalysts at 60
o
C and 30
o
C, respectively, at an oxygen flow rate of 60 mL min
-1
. Again, the
performance of Pt/CFx is better than that of Pt/C and, in the case of 60
o
C, the
performance of Pt/CFx and Pt are very similar. Figures 4.15 and 4.16 provide
the data normalized to the mass of Pt, which again demonstrates the better
performance of Pt/CFx, particularly at 60
o
C.
116
Figure 4.13 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-) with
1M methanol, O
2
flow at 60
mL min
-1
at 60
o
C.
Figure 4.14 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-) with
1M methanol, O
2
flow at 60
mL min
-1
at 30
o
C.
117
Figure 4.15 Normalized polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-)
with 1M methanol, O
2
flow at 60
mL min
-1
at 60
o
C.
Figure 4.16 Normalized polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-)
with 1M methanol, O
2
flow at 60
mL min
-1
at 30
o
C.
118
The last set of data presented was determined for the catalysts tested
with 1M methanol and air flow rates of 220 mL min
-1
at a temperature of
60
o
C. Figure 4.17 shows the raw data while in Figure 4.18 the normalized
results are presented. The values of current and power are low compared with
the previous plots since air and not pure oxygen was used. Nevertheless the
Pt/CFx again gave a better performance than Pt/C and Pt when the relative
amount of Pt is considered.
Figure 4.17 Polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-) with 1M
methanol at 60
o
C and air
flow at 220
mL min
-1
.
119
Figure 4.18 Normalized polarization and power plot for Pt (- ■-), Pt/C (- ●-) and Pt/CFx (- ▲-)
with 1M methanol at 60
o
C and air
flow at 220
mL min
-1
.
4.2.4.2 Single cell hydrogen fuel cell polarization tests
MEAs prepared for hydrogen fuel cells were tested at 90
o
C with
varying flow rates of humid hydrogen at the anode and humid oxygen at the
cathode. The comparison of the catalysts on the hydrogen fuel cell was
performed only for Pt/C and Pt/CFx because it is most common to use Pt
supported on carbon for both the anode and the cathode [34, 35]. This data is
presented in Figures 4.19 and 4.20. Normalization by weight was not done
since the MEAs have roughly the same amount of Pt on the cathode. The data
in Figure 4.19 corresponds to flow rates of 600 mL min
-1
for hydrogen and 700
mL min
-1
for oxygen. In Figure 4.20 the flow rates were 300 mL min
-1
for
120
hydrogen and 400 mL min
-1
for oxygen, and were the lowest flow rates tested
for both gases. The performance for both cathode catalysts was similar in both
cases. For the higher flow rates the polarization plot is similar for both
catalysts with the power at the top of the parabola being slightly higher for Pt/C
than for Pt/CFx. In the case of the lower flow, the polarization and power plots
are slightly better for Pt/CFx than for Pt/C. Two conclusions that can be drawn
from both Figures are that Pt/CFx presents a slightly higher OCV (ca. 100
mV), and that in the activation overpotential zone of the polarization plot, the
Pt/CFx performs slightly better, which is also in agreement with the results
obtained from the electrochemical measurements.
Figure 4.19 Polarization and power plot for Pt/C (- ●-) and Pt/CFx (- ▲-) at 90
o
C. H
2
flow of 600
mL min
-1
and 700 mL min
-1
of O
2
.
121
Figure 4.20 Polarization and power plot for Pt/C (- ●-) and Pt/CFx (- ▲-) at 90
o
C. H
2
flow of 300
mL min
-1
and 400 mL min
-1
of O
2
.
4.3 Chapter 4: Conclusion
Pt supported on CFx as a catalyst for the electrochemical reduction of
oxygen was prepared and its performance analyzed. The electrochemical
measurements, CV and EIS show that there is no change in the mechanism of
oxygen reduction and demonstrates improvements in comparison with Pt and
particularly with Pt supported on a non-fluorinated graphite. These
improvements were shown in practice by performing single cell fuel cell test.
When comparing the catalysts on the DMFCs, for a given set of
conditions, the performance in terms of the polarization and power plots were
better for Pt/CFx than the other two catalysts tested. This improvement can be
122
seen against the Pt/C in the raw data, and against Pt black in the data
normalized by the amount of Pt present in the catalyst.
When comparing the catalysts on the hydrogen fuel cells, improvement
of Pt/CFx over Pt/C is not as readily apparent, but there are some indications
that the Pt/CFx catalyst does perform better under the right conditions. The
fuel cell data presented shows that Pt/CFx has a good performance at low flow
rates of oxygen, and that CFx is an interesting option as an ORR catalyst
support, particularly for the purpose of decreasing the amount of Pt used on
fuel cell cathodes.
The increase in performance is explained by the better adsorption of
oxygen by the fluorinated carbon due to the affinity of oxygen for fluorine,
which makes the oxygen more available for the reduction to water on Pt, and
without compromising the electrical conductivity of the graphite support.
123
4.4 Chapter 4: References
1. Kim, M., Park, J.-N., Kim, H., Song, S., and Lee, W.-H., J. Power
Sources, 2006. 163(1): p. 93-97.
2. Surampudi, S., Narayanan, S.R., Vamos, E., Frank, H., Halpert, G.,
Laconti, A., Kosek, J., Prakash, G.K.S., and Olah, G.A., J. Power
Sources, 1994. 47(3): p. 377-385.
3. Huang, J., Yang, H., Huang, Q., Tang, Y., Lu, T., and Akins, D.L., J.
Electrochem. Soc., 2004. 151(11): p. A1810-A1815.
4. Kunimatsu, M. and Okada, T., Electrochem. Solid-State Lett., 2004.
7(11): p. A389-A390.
5. Lu, G.Q., Liu, F.Q., and Wang, C.-Y., Electrochem. Solid-State Lett.,
2005. 8(1): p. A1-A4.
6. Scott, K. and Argyropoulos, P., J. Power Sources, 2004. 137(2): p. 228-
238.
7. Wang, G., Sun, G., Zhou, Z., Liu, J., Wang, Q., Wang, S., Guo, J.,
Yang, S., Xin, Q., and Yi, B., Electrochem. Solid-State Lett., 2005. 8(1):
p. A12-A16.
8. Xie, F., Tian, Z., Meng, H., and Shen, P.K., J. Power Sources, 2005.
141(2): p. 211-215.
9. Yang, L.X., Allen, R.G., Scott, K., Christenson, P., and Roy, S., J.
Power Sources, 2004. 137(2): p. 257-263.
10. Hamza, M.A., Serratrice, G., Stebe, M.J., and Delpuech, J.J., J. Am.
Chem. Soc., 1981. 103(13): p. 3733-3738.
11. Lawson, D.D., Moacanin, J., Scherer, K.V., Terranova, T.F., and
Ingham, J.D., J. Fluorine Chem. , 1978. 12(3): p. 221-236.
12. Wesseler, E.P., Iltis, R., and Clark, L.C., J. Fluorine Chem. , 1977. 9(2):
p. 137-146.
13. Leger, V.Z., Nonaqueous cell having a MnO2/poly-carbon fluoride
cathode, 1982, US 4327166.
14. Watanabe, N. and Morigaki, K., Denki Kagaku, 1979. 47(3): p. 169-173.
124
15. Whitacre, J.F., West, W.C., Smart, M.C., Yazami, R., Prakash, G.K.S.,
Hamwi, A., and Ratnakumar, B.V., Electrochem. Solid-State Lett., 2007.
10(7): p. A166-A170.
16. Cabot Corporation - http://w1.cabot-corp.com/index.jsp - (1995-2008).
17. Damjanovic, A. and Brusic, V., Electrochim. Acta, 1967. 12(6): p. 615-
628.
18. Damjanovic, A., Dey, A., and Bockris, J.O.M., J. Electrochem. Soc.,
1966. 113(7): p. 739.
19. Damjanovic, A. and Genshaw, M.A., Electrochim. Acta, 1970. 15(7): p.
1281-1283.
20. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Phys. Chem. ,
1966. 70(11): p. 3761.
21. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Electrochem.
Soc., 1967. 114(11): p. 1107.
22. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Electrochem.
Soc., 1967. 114(5): p. 466.
23. Bockris, J.O.M. and Huq, A.K.M.S., Proc. R. Soc. , 1956. 237A: p.
1733.
24. Wroblowa, H., Rao, M.L.B., Damjanovic. A, and Bockris, J.O.M., J.
Electroanal. Chem., 1967. 15(2-3): p. 139.
25. Markovic, N.M. and Ross, P.N., Surf. Sci. Rep. , 2002. 45(4-6): p. 121-
229.
26. Wroblowa, H.S., Yen Chi, P., and Razumney, G., J. Electroanal.
Chem., 1976. 69(2): p. 195-201.
27. Damjanovic, A., J. Electrochem. Soc., 1991. 138(8): p. 2315-2320.
28. Damjanovic, A. and G., H.P., J. Electrochem. Soc., 1988. 135(9): p.
2269-2273.
29. Damjanovic, A., Sepa, D.B., and Vojnovic, M.V., Electrochim. Acta,
1979. 24(8): p. 887-889.
30. Bockris, J.O.M. and S.U.M., K., Surface Electrochemistry. 1993, New
York: Plenum
125
31. Antoine, O., Bultel, Y., and Durand, R., J. Electroanal. Chem., 2001.
499(1): p. 85-94.
32. Sepa, D.B., Vracar, L.M., Vojnovic, M.V., and Damjanovic, A.,
Electrochim. Acta, 1986. 31(11): p. 1401-1402.
33. Koper, M.T.M. and Sluyters, J.H., J. Electroanal. Chem., 1994. 371(1-
2): p. 149-159.
34. Yu, X.W. and Ye, S.Y., J. Power Sources, 2007. 172(1): p. 133-144.
35. Yu, X.W. and Ye, S.Y., J. Power Sources, 2007. 172(1): p. 145-154.
126
Chapter 5: Microbial Fuel Cells
5.1 Chapter 5: Introduction
As mentioned in Chapter 1, MFCs are the most recent developments in
fuel cells. The first reports that present the idea of using a microorganism for
electricity generation in a fuel cell dates from the early 1980’s [1, 2]. The
definition of a MFC is a device that converts chemical energy into electricity
through the catalytic and metabolic activity of a microorganism. Several
microorganisms have been used in the assembly of MFC, among these, the
most common ones found in the literature are Aeromonas [3], Clostridium [4],
Geobacter [5], Enterococcus [6, 7] and Shewanella [8, 9, 10, 11, 12, 13, 14].
As briefly described in Chapter 1, the microorganism feed from a given carbon
source, i.e. the electron donor, oxidizes it through digestion. As with any other
living organism, the electrons resulting from that oxidation are transported to
the respiration chain by the reduced form of Nicotinamide adenine dinucleotide
(NADH) where it is re-oxidized to Nicotinamide adenine dinucleotide (NAD)
and the electrons go to the ultimate electron acceptor. This electron acceptor
is oxygen when the bacteria is in an aerobic environment, but most
microorganisms, can also release the electrons to a variety of acceptors such
as nitrate, nitrite, sulfite and thiosulfate, and metals such as Fe(III) or Mn(IV).
Figure 5.1 shows the schematic representation of the process described
above. The carbon source supplied depends on the bacteria used, and studies
show that MFC with a consortia of microbes have been assembled [7] allowing
127
for a mix of carbon source to be used as fuels. This last fact has resparked
interest in MFC over the last few years because the generation of electricity
can be coupled with waste-water treatment [11, 15, 16, 17, 18], giving the
ability of the bacteria used in MFC to metabolize several carbon sources [1].
Figure 5.1 Schematic representation of the electron movement from the carbon source to the
reduction of oxygen in microbial metabolism.
The final electron acceptor is an important characteristic of the
microorganism. Initially the MFC needed a mediator, or electron shuttle, to
transport the electrons from the microorganism to the electrode in the fuel cell.
Many different bacteria produce electric current in this way [2, 19, 20], but the
need for such a mediator represents an extra step in the electron transfer that
decreases the overall energy generated. More recently, it was found that
certain bacteria can transfer the electrons from the oxidized fuel directly into
the electrode of a fuel cell [10, 12, 13]. This direct transfer simplifies the
process in the MFC due to the absence of the mediator and increases the
128
efficiency in the electron transport path. The microorganisms that allow this
kind of electron process are named electrochemically active bacteria (EAB)
and also show activity through the presence of a peak in cyclic voltammetry
measurements [21]. Among the microorganisms capable of this activity include
the aforementioned Aeromonas, Clostridium and Shewanella.
In the present work, the fuel cell hardware used for the DMFC, i.e. two
chambers separated by an MEA (where one of the chambers is the dry
cathode), was adapted for MFC. MEAs using Nafion® or PVDF-PSSA IPN as
membranes were prepared and tested with Shewanella oneidensis MR-1 as a
single culture microorganism in the anode and lactate as the electron donor
(fuel).
5.1.1 Microorganism characteristics
The Shewanella oneidensis MR-1 strain, formerly known as Shewanella
putrefaciens MR-1, is a bacillus Gram negative anaerobe facultative, first
isolated from the anaerobic sediments of Oneida Lake, New York in the late
1980’s . MR-1, as well as other strains, was characterized as redox interface
organisms because of their abundance in oxic/anoxic interfaces in different
water bodies [22, 23, 24]. The presence of Shewanella in those interfaces was
explained as a result of the microorganisms’ ability to exploit a wide range of
terminal electron acceptors under anaerobic conditions, as well as oxygen
under aerobic conditions. The many electrons acceptor used by Shewanella
129
MR-1 are Fe(III), Mn(IV), S
0
S
2
O
3
2-
, NO
3
-
, NO
2
-
, trimethylamine oxide, dimethyl
sulfoxide, glycine and fumarate during anaerobic respiration as well as growth.
With regard to electron donors (fuels), MR-1 can use hydrogen plus a plethora
of carbon sources which include glucose, lactate, pyruvate, propionate,
ethanol and formate, besides a number of amino acid such as serine [23].
Figure 5.2 shows a simplified version of the anaerobic metabolism of
Shewanella [23]. In addition to the aforementioned carbon sources and
electrons acceptors, which give an idea of why Shewanella can transfer the
electrons directly into an electrode, Shewanella, under anaerobic conditions,
has the ability to allow the migration of cytochromes to the surface of the
membrane, which increases the interaction between the bacteria and the
electron acceptors. More than 39 c-type cytochromes have been identified in
Shewanella MR-1 [25, 26]. Moreover, the presence of a particular pili was
observed [27], and was recently confirmed to play a role in the transport of
electrons [28]. These pili, also termed bacterial nanowire, appear in response
to oxygen limitation and present on the pili exterior are a number of
cytochromes. Scanning electron microscopy (SEM) images show the
presence of these pili between the bacteria and the substrate, as well as
between bacteria. Scanning tunneling microcopy (STM) measurements
confirmed these bacterial nanowires to be electrically conductive, and are
considered to be one of the ways in which the bacteria transfer the electrons
directly to the electrode.
130
Figure 5.2 Anaerobic metabolism of Shewanella MR-1. Reproduced from [23]
5.1.2 Microbial fuel cell assembly considerations
The choice of electrode material, membrane and the cell design of a
MFC influence the overall power output. The plethora of MFC varieties found
in the literature [16, 17, 29, 30, 31, 32, 33, 34, 35, 36, 37] shows that an
optimal design has not been found. Some of the MFCs use both single
chamber as well as dual chamber cell designs in which the anode and cathode
are either free-standing in solution or separated by a proton-permeable
membrane [37, 38, 39, 40]. In Polymer Electrolyte Membrane Fuel Cells (PEM
131
FC) with H
2
as a fuel, or in Direct Methanol Fuel Cell (DMFC) with methanol as
a fuel, there is consensus that the way to obtain the maximum power is by
constructing a Membrane Electrode Assembly (MEA). As was described in
previous Chapters, the MEA is a thin array where a polymeric membrane,
which serves as the ionic conductor, is placed between two gas diffusion
electrodes (GDE), the anode and the cathode. The catalysts for the anodic
and cathodic reactions are located on the inner face of the GDE, in direct
contact with the membrane. One of the MEA preparation process, described
in Chapter 6, is by hot pressing, in order to obtain a close contact between the
GDE, the catalyst and the membrane. In the case of MFC where the anode
catalyst is the microorganism, the MEA has to be assembled before the
bacteria are adsorbed to the anode electrode otherwise the pressing process
will kill the bacteria. Another issue in MFC is the use of water in the cathode
compartment [13, 29, 31]. In this particular design the cathode is filled with an
electrolyte solution, which could be convenient in terms of a more simple
construction, but the low solubility of O
2
results in mass transport limitation at
the cathode resulting in low power output. Probably one of the most discussed
issues is the use of a membrane like Nafion® to separate the compartments
[14, 37, 38, 39, 41, 42]. Nafion® is the most used membrane in PEMFC and
has also been used as a membrane for MFC. It has been pointed out that the
use of Nafion® in MFC is problematic due to the high concentration of ions like
Na
+
, a common cation in the buffers used in MFC and present at
132
concentrations which are nonexistent in PEMFC. Moreover, Nafion® has more
affinity to anions like Na
+
and K
+
than H
+
[43] exacerbating the aforementioned
problem. The price of Nafion® has also been indicated as a drawback for MFC
[37]. The MFC cost has been related to the price of Nafion, particularly
because electrodes with big cross sections would have to be used in order to
obtain significant power, which would require a significant amount of
membrane. Also the acidity of Nafion® has been identified as a problem for
the growth of the bacteria. However, membranes like the interpenetrated
networks of polyvinylidene fluoride and polystyrene sulfonic acid (IPN PVDF-
PSSA) [44] are inexpensive and have shown improvement with respect to
Nafion® in DMFC [45, 46], and membranes of this type, as well as new
engineered membranes, can solve some of the aforementioned issues.
5.2 Chapter 5: Results and Discussion
5.2.1 Microbial inoculation and scanning electron microscopy (SEM)
images
The microorganism inoculation procedure was modified at some point
during testing of the several MFCs assembled from a more direct process,
identified as inoculation procedure 1 (IP1), to a process that spans a period of
three days, identified as inoculation procedure 2 (IP2). Descriptions of both
procedures are provided in Chapter 6. This change from the most direct IP1
procedure to IP2 came with the goal of obtaining a dense biofilm over the
133
electrode and ultimately improving the power output of the MFC. Initially, after
testing some MFCs using IP1, direct observation of the electrodes as well as
scanning electron microscopy (SEM) images showed low amounts of
microbes and poor biofilms, as can be seen in Figure 5.3. These results were
consistently obtained after several attempts, which prompted a change to the
inoculation procedure essentially in three ways. The first change consisted of
filling the cell with buffer and lactate immediately after autoclaving. This step
was a way to prevent contamination since sometimes the microbial culture did
not reach the required optical density (OD) for inoculation on time. Also adding
the lactate solution was considered as a way to keep the cell and membrane
humid until inoculation; even though the MEA’s membrane does not dry out
during autoclaving. Another relevant change consisted of the two subsequent
microbial inoculations with the resting period in between, which was expected
to increase the amount of microbes forming a biofilm. The final change
consisted of the connection of a 1 K Ω resistance from the first inoculation,
between anode and cathode, in order to have the microbes in an active
metabolism under anaerobic conditions from the moment they were introduced
into the anode compartment. Otherwise, in case of circuit open conditions, i.e.
no current flowing, the bacteria will be in a dormant state. These changes, i.e.
IP2, proved to be a much more efficient way to obtain a dense biofilm, as seen
in Figure 5.4. By comparing the SEMs in Figures 5.3 and 5.4, it can be
observed that after fixation of the biofilm a magnification of 12.5 x was
134
necessary in order to see the biofilm in Figure 5.3 (A) using IP1 while in Figure
5.4 (A) using IP2 a digital picture of the MEA without magnification showed
considerable biofilm coverage. In Figure 5.3(B), which corresponds to an SEM
obtained with 6000 x magnification, very few microorganisms appear to be
present in the biofilm. This low density biofilm was observed on the whole
electrode surface. On the other hand, as seen in Figures 5.4(B), obtained with
2000 x magnification, and Figure 5.4(C), obtained with 5000 x magnification,
the images give a clear idea not only of the very densely populated biofilm, but
also the better overall coverage. These studies also confirm that Nafion® and
related proton exchange membranes are excellent hosts for the bacteria
resulting in good biofilm formation.
135
(A)
(B)
Figure 5.3 SEM images for an MEA used in a FC inoculated using IP1. (A) 12.5 x
magnification. (B) 6000 x magnification.
136
(A)
(B)
137
(C)
Figure 5.4 (A) Digital photo taken of an MEA used in a FC inoculated using IP2. (B) SEM
image at 2000 x magnification. (C) SEM image at 5000 x magnification.
5.2.2 MFC polarization measurements
As stated in the previous section, the rationale for changing the
inoculation procedure was to obtain a better biofilm. Apart from the SEM
images, the other indication that a better biofilm is formed is by monitoring of
the actual cell voltage. In the early experiments, using IP1, the OCV was
monitored immediately after inoculation, and polarization measurements were
performed once a constant OCV was obtained. This usually occurred after a
period of 2 to 3 days. Periods of 1 to 2 days were previously reported by other
authors [47].
138
(A)
(B)
Figure 5.5 (A) and (B) OCV for two different MFC inoculated with IP1. Temperature 28
o
C, O
2
flow rate 7 mL min
-1
.
139
Figures 5.5 (A) and (B) show the OCV vs. time for two different MFC
inoculated with IP1. The voltage increased slowly and was unstable for a
period of time and then decreased again. At that point, after the addition of
sodium lactate, the voltage rises and maintains a constant value, which
allowed polarization measurements to be performed. Again, this unsteady
OCV during the initial days of the fuel cell testing was an indication that the
microbes needed some period of adaptation inside the cell.
When IP2 was used, the cell voltage rather than the OCV was
monitored from the second inoculation (see Chapter 6), with the 1K Ω
resistance connected between the anode and the cathode. Figures 5.6 (A)
and (B) show the voltage vs. time across the resistance for two different cells
using IP2. It can be seen that from the moment the cell is connected to the
potentiostat, the voltage indicates a high value. That value was maintained
without change, and as the figures show, in most of the cases the voltage
across the resistance was higher than the OCV values obtained for MFC using
IP1. As mentioned before, after a constant voltage was obtained, polarization
measurements were performed.
140
(A)
(B)
Figure 5.6 (A) and (B) Cell voltage across a 1K Ω resistance for two different MFCs
inoculated with IP2. Temperature 28
o
C, O
2
flow rate 7mL min
-1
.
141
(A)
(B)
Figure 5.7 (A) and (B) Polarization and Power density plots for two different MFCs inoculated
with IP1. (A) Plots measured following the OCV of Figure 5.5(B). Temperature 28
o
C, O
2
flow rate 7 mL min
-1
.
142
Figures 5.7 (A) and (B) show the polarization and power plot obtained
for two cells using IP1. The plots presented in Figure 5.7(A) were measured
after the OCV shown in Figure 5.5(B), the OCV reached at the end of Figure
5.5(B) is almost the same as the starting point in the polarization in Figure
5.7(A). The maximum power density (P
max
), i.e. the top of the parabola, and
the limiting current density (j
lim
), i.e. the last current value in the polarization,
obtained in Figures 5.7 (A) and (B) were satisfactory, however the plots of
Figure 5.7(B) were the best values obtained for IP1, and were shown to be low
when compared to the results using IP2.
Figures 5.8 (A) and (B) show the polarization and power plots for two
MFC using IP2. In Figure 5.8(A), for which the plots were obtained after the
voltage measurements in Figure 5.6(A). it is clear how the cell voltage, which
was ca. 430 mV, increased to an OCV of 530 mV, measured by the
potentiostat right before starting the polarization. In terms of P
max
and j
lim
,
Figure 5.8(A) and (B) show big improvements compared to the MFC using
IP1. The P
max
went from average values of 120 mW m
-2
for IP1 to values
always above 200 mW m
-2
for IP2. The j
lim
, also showed an increase by a
factor of two in Figures 5.7(A) and (B) compared to Figures 5.8(A) and (B).
These results can be directly related to the biofilm density of the
microorganism and show that the steps taken in order to improve the
inoculation procedure were beneficial.
143
(A)
(B)
Figure 5.8 (A) and (B) Polarization and Power density plots for two different MFCs inoculated
with IP2. (A) Plots measured following the OCV of Figure 5.6(A). (B) ( ─■─)
correspond to a day after the second inoculation and ( ─+ ─) correspond to
measures performed 15 days after the second inoculation. Temperature 28
o
, O
2
flow rate 7 mL min
-1
.
144
In Figure 5.8(B), a value of 260 mW m
-2
represented by ( ─■─) can be
seen which was obtained one day after the second inoculation, while the plot
( ─+ ─) which yielded a P
max
value of 290 mW m
-2
corresponds to
measurements performed 15 days after the second inoculation. On the MFC
inoculated using IP1, measurements performed on subsequent days never
yielded better results than measurements performed during the initial 2-3
days. This lack of improvement can be explained by the fact that the biofilm
obtained using IP1 is poor compared to that obtained using IP2, again related
to the inoculation procedure, and therefore the direct inoculation does not
allow a substantial colony of bacteria to develop at the membrane interface.
An important step in DMFC and related fuel cells is the conditioning of
the membrane on the assembled fuel cell; which is accomplished by forcing
protons to move from anode to cathode through the membrane by drawing
current in a non-steady state manner. This procedure allows protons to move
easily through the membrane when polarization measurements are performed;
in other words, it reduces the internal resistance by reducing the opposition to
proton movement. In MFC, since the microbes are the “catalysts” on the
anode side, this method would be impractical. In order to force protons from
the anode to the cathode, water was electrochemically split to hydrogen and
oxygen in the MFC after the cell was assembled and before it was autoclaved.
The anode was filled with a 0.1 M H
2
SO
4
solution. Since the anode lacks a
catalyst, a Pt wire electrode was introduced and used as anode for the
145
oxidation of water to oxygen, while in the MEA cathode protons got reduced to
hydrogen. The redox reaction for water was only possible because the protons
moved through the membrane to reach the cathode. The electrolysis of water
was performed for 6 h at a current were the production of oxygen at the anode
was noticeable and steady. After this membrane conditioning, the MFC was
autoclaved and inoculated via IP2. Figure 5.9 shows the result for the
assembled MFC. Figure 5.9(A) shows the cell voltage after the second
inoculation with the 1K Ω resistance until it was stopped for testing, while
Figure 5.9(B) shows the polarization and power plots. For this MFC, the cell
voltage across the resistance was measured to increase from a value of 500
mV, to an OCV of 570 mV. Both the j
lim
and the P
max
increased when
compared to MFCs using IP2 that did not undergo the membrane conditioning,
as presented in Figures 5.8(A) and (B). The P
max
showed an increase in the
maximum power to 600 mW m
-2
. These results demonstrate that this kind of
MEA conditioning is not only useful, but also shows that considerable power is
lost by the internal resistance.
146
(A)
(B)
Figure 5.9 (A) Cell voltage across a 1 K Ω resistance and (B) Polarization and Power density
plots for a MFC prepared with Nafion® 117 using IP2 that was subjected to
membrane conditioning after assembly. Temperature 28
o
C, O
2
flow rate
7 mL min
-1
.
147
Figure 5.10 Polarization and Power density plots for a MFC with a PVDF-PSSA MEA.
Temperature 28
o
C and O
2
flowing on the cathode at 7 ml min
-1
, using IP2.
MEAs were also prepared with PVDF-PSSA, conditioned as previously
described and inoculated using IP2. Polarization and power results are
showed in Figure 5.10. Firstly there is a noticeably low OCV (ca. 300 mV).
This low OCV could be a result of the membrane thickness, which is a little bit
higher than the Nafion® 117, and to the fact that this membrane was prepared
to withstand higher currents. However, even with a low OCV, the P and j
lim
are
comparable with those of Nafion® MEAs presenting higher OCVs. The P
max
of
245 mW m
-2
is comparable with the P
max
for the MFC shown in Figure 5.8(B),
while the j
lim
is slightly above the j
lim
of the MEA in Figure 5.9(B). That value of
j
lim
for the PVDF-PSSA’s MEA represents a total of 7 mA for the MFC which is
148
again among the highest value obtained presently. This kind of membrane is
promising because if the OCV could be increased (e.g. by reduction of the
thickness), a much higher power should be expected.
5.3 Chapter 5: Conclusion
If MFCs are to be commercialized, a simple and useful design is
needed. Construction of MEA, using the same procedure as used for PEMFC,
for MFC was carried out. A novel way of conditioning the membranes was also
found. The MEA presents the best contact between electrodes and
membrane, thereby reducing the internal resistance, and also the use of a dry
cathode moves away from the complication of having air or oxygen bubbling
into the cathode solution. The results presented here for the MFC built in
house not only increased the power output compared to the more simple fuel
cell designs [48, 49, 50] by 2 to 3 orders of magnitude but these results also
show improvements in power density from early to later tests performed in this
study. The results also demonstrated the importance of microbial inoculation
or microbial accommodation times in the anode chamber for the formation of
adequate biofilm and how it impacts the overall power output. Furthermore,
the results indicate that although a limit on the power output exists for MEAs
using Nafion® due to its affinity for the Na
+
ion, which decreases proton
conductivity, this limit has not yet been reached at the realized power
densities. A more economical option than Nafion® was employed using
149
PVDF-PSSA membrane. The results obtained however did not show a
tangible improvement. On the other hand, we believe that there is much room
for improvement since this particular PVDF-PSSA membrane was initially
designed for DMFC.
150
5.4 Chapter 5: References
1. Akiba, T., Bennetto, H.P., Stirling, J.L., and Tanaka, K., Biotechnol. Lett
. 1985. 9(11): p. 616.
2. Stirling, J.L., Bennetto, H.P., Delaney, G.M., Mason, J.R., Roller, S.D.,
Tanaka, K., and Thurston, C.F., Biochem. Soc. Trans., 1983. 11(4): p.
451-453.
3. Pham, C.A., Jung, S.J., Phung, N.T., Lee, J., Chang, I.S., Kim, B.H., Yi,
H., and Chun, J., FEMS Microbiol. Lett., 2003. 223(1): p. 129-134.
4. Park, H.S., Kim, B.H., Kim, H.S., Kim, H.J., Kim, G.T., Kim, M., Chang,
I.S., Park, Y.K., and Chang, H.I., Anaer., 2001. 7(6): p. 297-306.
5. Bond, D.R. and Lovley, D.R., Appl. Environ. Microbiol., 2003. 69(3): p.
1548-1555.
6. Kim, G.T., Hyun, M.S., Chang, I.S., Kim, H.J., Park, H.S., Kim, B.H.,
Kim, S.D., Wimpenny, J.W.T., and Weightman, A.J., J. Appl. Microbiol.,
2005. 99(4): p. 978-987.
7. Rabaey, K., Boon, N., Siciliano, S.D., Verhaege, M., and Verstraete,
W., Appl. Environ. Microbiol., 2004. 70(9): p. 5373-5382.
8. Hernandez, M.E. and Newman, D.K., Cell. Mol. Life Sci., 2001. 58(11):
p. 1562-1571.
9. Kim, B.H., Ikeda, T., Park, H.S., Kim, H.J., Hyun, M.S., Kano, K.,
Takagi, K., and Tatsumi, H., Biotechnol. Tech., 1999. 13(7): p. 475-478.
10. Kim, B.H., Kim, H.J., Hyun, M.S., and Park, D.H., J. Microbiol.
Biotechnol., 1999. 9(2): p. 127-131.
11. Kim, B.H., Park, H.S., Kim, H.J., Kim, G.T., Chang, I.S., Lee, J., and
Phung, N.T., Appl. Microbiol. Biotechnol., 2004. 63(6): p. 672-681.
12. Kim, H.J., Hyun, M.S., Chang, I.S., and Kim, B.H., J. Microbiol.
Biotechnol., 1999. 9(3): p. 365-367.
13. Kim, H.J., Park, H.S., Hyun, M.S., Chang, I.S., Kim, M., and Kim, B.H.,
Enzyme Microb. Technol., 2002. 30(2): p. 145-152.
14. Kim, J.R., Cheng, S., Oh, S.E., and Logan, B.E., Environ. Sci. Technol.,
2007. 41(3): p. 1004-1009.
151
15. Jang, J.K., Pham, T.H., Chang, I.S., Kang, K.H., Moon, H., Cho, K.S.,
and Kim, B.H., Process. Biochem., 2004. 39(8): p. 1007-1012.
16. Liu, H., Ramnarayanan, R., and Logan, B.E., Environ. Sci. Technol.,
2004. 38(7): p. 2281-2285.
17. Min, B. and Logan, B.E., Environ. Sci. Technol., 2004. 38(21): p. 5809-
5814.
18. Moon, H., Chang, I.S., and Kim, B.H., Bioresour. Technol., 2006. 97(4):
p. 621-627.
19. Delaney, G.M., Bennetto, H.P., Mason, J.R., Roller, S.D., Stirling, J.L.,
and Thurston, C.F., J. Chem. Technol. Biotechnol., 1984. 34(1): p. 13-
27.
20. Roller, S.D., Bennetto, H.P., Delaney, G.M., Mason, J.R., Stirling, J.L.,
and Thurston, C.F., J. Chem. Technol. Biotechnol., 1984. 34(1): p. 3-
12.
21. Chang, I.S., Moon, H., Bretschger, O., Jang, J.K., Park, H.I., Nealson,
K.H., and Kim, B.H., J. Microbiol. Biotechnol., 2006. 16(2): p. 163-177.
22. Brettar, I. and Hofle, M.G., Mar. Ecol. Prog. Ser., 1993. 94(3): p. 253-
265.
23. Scott, J.H. and Nealson, K.H., J. Bacteriol., 1994. 176(11): p. 3408-
3411.
24. Myers, C.R. and Nealson, K.H., Science, 1988. 240(4857): p. 1319-
1321.
25. Meyer, T.E., Tsapin, A.I., Vandenberghe, I., De Smet, L., Frishman, D.,
Nealson, K.H., Cusanovich, M.A., and Van Beeumen, J.J., OMICS A: J.
Integr. Biol., 2004. 8(1): p. 57-77.
26. Schwalb, C., Chapman, S.K., and Reid, G.A., Biochem., 2003. 42(31):
p. 9491-9497.
27. Reguera, G., McCarthy, K.D., Mehta, T., Nicoll, J.S., Tuominen, M.T.,
and Lovley, D.R., Nature, 2005. 435(7045): p. 1098-1101.
152
28. Gorby, Y.A., Yanina, S., McLean, J.S., Rosso, K.M., Moyles, D.,
Dohnalkova, A., Beveridge, T.J., Chang, I.S., Kim, B.H., Kim, K.S.,
Culley, D.E., Reed, S.B., Romine, M.F., Saffarini, D.A., Hill, E.A., Shi,
L., Elias, D.A., Kennedy, D.W., Pinchuk, G., Watanabe, K., Ishii, S.,
Logan, B., Nealson, K.H., and Fredrickson, J.K., Proc. Nat. Acad. Sci.
U.S.A., 2006. 103(30): p. 11358-11363.
29. Aelterman, P., Rabaey, K., Pham, H.T., Boon, N., and Verstraete, W.,
Environ. Sci. Technol., 2006. 40(10): p. 3388-3394.
30. He, Z., Minteer, S.D., and Angenent, L.T., Environ. Sci. Technol., 2005.
39(14): p. 5262-5267.
31. Logan, B.E., Murano, C., Scott, K., Gray, N.D., and Head, I.M., Water
Res., 2005. 39(5): p. 942-952.
32. Min, B.K., Cheng, S.A., and Logan, B.E., Water Res., 2005. 39(9): p.
1675-1686.
33. Rabaey, K., Boon, N., Hofte, M., and Verstraete, W., Environ. Sci.
Technol., 2005. 39(9): p. 3401-3408.
34. Rabaey, K., Clauwaert, P., Aelterman, P., and Verstraete, W., Environ.
Sci. Technol., 2005. 39(20): p. 8077-8082.
35. Rabaey, K., Ossieur, W., Verhaege, M., and Verstraete, W., Water Sci.
Technol., 2005. 52(1-2): p. 515-523.
36. Rosenbaum, M., Schroder, U., and Scholz, F., Environ. Sci. Technol.,
2005. 39(16): p. 6328-6333.
37. Liu, H. and Logan, B.E., Environ. Sci. Technol., 2004. 38(14): p. 4040-
4046.
38. Oh, S.E. and Logan, B.E., Appl. Microbiol. Biotechnol., 2006. 70(2): p.
162-169.
39. Rozendal, R.A., Hamelers, H.V.M., and Buisman, C.J.N., Environ. Sci.
Technol., 2006. 40(17): p. 5206-5211.
40. Liu, H., Cheng, S.A., and Logan, B.E., Environ. Sci. Technol., 2005.
39(14): p. 5488-5493.
41. Liu, H., Cheng, S.A., and Logan, B.E., Environ. Sci. Technol., 2005.
39(2): p. 658-662.
153
42. Pham, T.H., Jang, J.K., Moon, H.S., Chang, I.S., and Kim, B.H., J.
Microbiol. Biotechnol., 2005. 15(2): p. 438-441.
43. Okada, T., Moller-Holst, S., Gorseth, O., and Kjelstrup, S., J.
Electroanal. Chem., 1998. 442(1-2): p. 137-145.
44. Halpert, G., Narayanan, S.R., Prakash, G.K.S., Olah, G.A., Smart,
M.C., Wang, Q.J., and Surumpudi, S., Polymer electrolyte membranes
for use in fuel cells, 2002, US 6444343.
45. Prakash, G.K.S., Smart, M.C., Wang, Q.J., Atti, A., Pleynet, V., Yang,
B., McGrath, K., Olah, G.A., Narayanan, S.R., Chun, W., Valdez, T.,
and Surampudi, S., J. Fluorine Chem., 2004. 125(8): p. 1217-1230.
46. Prakash, G.K.S., Smart, M.C., Olah, G.A., Narayanan, S.R., Chun, W.,
Surampudi, S., and Halpert, G., J. Power Sources, 2007. 173: p. 102-
109.
47. Oh, S., Min, B., and Logan, B.E., Environ. Sci. Technol., 2004. 38(18):
p. 4900-4904.
48. Bretschger, O., Obraztsova, A., Sturm, C.A., Chang, I.S., Gorby, Y.A.,
Reed, S.B., Culley, D.E., Reardon, C.L., Barua, S., Romine, M.F.,
Zhou, J., Beliaev, A.S., Bouhenni, R., Saffarini, D., Mansfeld, F., Kim,
B.H., Fredrickson, J.K., and Nealson, K.H., Appl. Environ. Microbiol.,
2007. 73(21): p. 7003-7012.
49. Manohar, A.K., Bretschger, O., Nealson, K.H., and Mansfeld, F.,
Bioelectrochem., 2008. 72(2): p. 149-154.
50. Manohar, A.K., Bretschger, O., Nealson, K.H., and Mansfeld, F.,
Electrochim. Acta, 2008. 53(9): p. 3508-3513.
154
Chapter 6: Experimental set up and procedures
6.1 Chapter 6: Powder Catalyst preparation
The powder catalysts for the DFAFC presented in Chapter 2 and for the
ORR presented in Chapter 4 were prepared by the same method. For the
catalysts to be supported over carbon, like in the case of Pt/CFx in Chapter 4,
a slurry of the carbon powder was prepared by adding the desired amount of
the carbon support in milli-Q (Direct-Q3 system, Millipore) and the resulting
mixture was vigorously stirred and sonicated. Once the suspension of the
carbon support was obtained, a solution containing the desired amount of the
metal salt or metal salts, in the case of a bimetallic catalyst, was added to the
slurry and adjusted to pH 8 with NaOH (pellets AR, Mallinckrodt) and heated
to 90
o
C. The salts used to prepare the catalyst in Chapter 2 were PdCl
2
(99.9
%, Alfa Aesar), H
2
AuCl
4
(trihydrate, 99.99 % Alfa Aesar), SnSO
4
(95 %, Sigma
Aldrich) and FeCl
3
(hexahydrate, 98% Alfa Aesar) while for the Pt/CFx in
Chapter 4 the salt used was H
2
PtCl
6
(hexahydrate, 99.9 % Alfa Aesar). Once
the temperature was reached, NaBH
4
(granular 98 %, Sigma Aldrich) was
added and the resulting solution was maintained under heat and rapid stirring.
After adding the NaBH
4
, heat was maintained for 2 more hours followed by
stirring for 12 h. The liquid and precipitate were then transferred to a centrifuge
tube, centrifuged and the solid successively washed and centrifuged until the
155
supernatant presented a neutral pH. The solid was then separated and dried
in a vacuum oven at 60
o
C overnight.
6.1.1 Catalyst surface analysis.
The composition of the catalyst prepared in Chapter 2 for DFAFC and
the Pt/CFx prepared in Chapter 4 for the ORR were analyzed by X-ray
photoelectron spectroscopy (XPS) using a VJ XPS, and by Energy Dispersive
X-ray Spectrometry (EDS) coupled to a Transmission Electron Microscope
(TEM) Philips EM420. The TEM images presented in Chapter 4 for Pt/CFx
were acquired using a Transmission Electron Microscope JEOL 100CX.
6.2 Chapter 6: Electrochemical cells and testing methodologies.
Two electrochemical cells were used for all the measurements. The cell
shown in Figure 6.1 was the one used to obtain the cyclic voltammograms for
the catalyst powders in Chapters 2, 3 and 4, as well as the voltammograms for
bulk metal Sn in Chapter 3. The cell was also used to obtain the EIS results for
the powder catalysts in Chapter 4. Figure 6.2 shows the holder for the working
electrode used in the aforementioned cell. The holder was made in house from
Teflon®, and can hold a flat disc exposing one face (1 cm
2
) of the electrode to
the solution. The disc used as the electrode is exchangeable and therefore in
Chapter 3 a Sn (Alpha Aesar 99.9%) disc was used for the Sn bulk metal
measurements. For all other measurements a graphite (Aldrich 99%) disk was
156
used. In the case of the graphite disc, the exposed face was sanded with 200
grit paper to get a rough surface before being covered with the catalyst. A
suspension of the corresponding catalyst powder with Milli-Q water and
Nafion® ionomer (Aldrich 5% in alcohol) in a 1:1:1 proportion was prepared,
spread on the graphite face and dried. The catalysts measured included the
Pd/Au and Pd/Sn prepared in house, as well as Pd black (99.9 % Alfa Aesar);
Chapter 2. In chapter 4 the catalyst used were Pt/CFx prepared in house, Pt
black (Alfa Aesar HiSPEC 1000), and Pt/C 40% (Alfa Aesar HiSPEC 4000).
The amount of dry paint spread over the electrode was ca. 8~9 mg. The cell
design allowed the solution to be saturated with any gas. Carbon dioxide
(99,998 % research grade Gilmore) was used in Chapter 3 and oxygen
(compressed, Gilmore) in Chapter 4; for the measurements of formic acid
oxidation no gas was bubbled. Ar (ultra high pure grade Gilmore) was used
throughout all tests when the solution needed degassing.
In Chapter 2, the base electrolyte was 0.1M H
2
SO
4
(98% ACS grade
EMD) and the measurements were performed in a solution of 0.1M HCOOH
and 0.1M H
2
SO
4
, by diluting pure formic acid (96 % ACS reagent, Sigma
Aldrich). For Chapter 3 the solution used for the carbon dioxide
electroreduction was 0.5 M NaHCO
3
prepared from the solid salt (granular,
ACS reagent, Aldrich). In Chapter 4 only 0.1M H
2
SO
4
was used for all
measurements.
157
Figure 6.1 Electrochemical glass cell used in CV and EIS measurements.
158
Figure 6.2 Holder for the working electrode used for the cell in Figure 6.1.
Figure 6.3 presents the cell used in Chapter 3 for the electrolysis of
carbon dioxide to formic acid, and for the voltammograms acquired for the Sn
metal powder (powder, -100 mesh, 99.5 %, Alfa Aesar) with Nafion® over the
GDE. As mentioned in Chapter 3 the cell allows the gases to flow through the
working electrode (GDE) as can evident from the figure. Besides Sn, the
electrolysis was performed on Pb (powder, -325 mesh, 99.5 %, Alfa Aesar), In
(powder, -325 mesh, 99.99 %, Alfa Aesar) and Cd (powder, -325 mesh, 99.5
%, Alfa Aesar). The cell also possessed a septum port to collect gas samples
to be analyzed by a Gas Chromatograph, and a gas trap that permitted the
gas to flow at ambient pressure. The electrolyte solution used for the
electrolysis at the different potentials was the same as for the electrochemical
measurements performed with the cell in Figure 6.1 (0.5M NaHCO
3
). The gas
samples collected from the cell during and after electrolysis were analyzed by
a Varian CP380 Gas Chromatograph using a Carboxen 1010Plot column (30m
159
x 0.53 mm, Supelco), while the amounts of formic acid obtained by the
reduction of carbon dioxide were analyzed using a High Performance Liquid
Chromatography (HPLC) (Thermo Finnigan Surveyor with a ultra violet (UV)
detector) and using a column to determine low-molecular weight carboxylic
acids, SupelcoGel C-610H (30 cm x 7.8 mm, Supelco)
The reference electrode used in all measurements was Ag/AgCl (ss)
(Metrohm) and all the potentials reported were converted to the standard
hydrogen electrode (SHE) (0.197 V vs Ag/AgCl). The counter electrode used
was a Pt wire coiled with a total area of 2 cm
2
. On both cells, the compartment
for the counter electrode was separated by a glass frit from the main
compartment, while the compartment for the reference electrode was
connected through a Lugging capillary to the main compartment. The cyclic
voltammetry and Electrochemical Impedance Spectroscopy measurements
were performed with a Solartron SI 1287 and SI 1260 Impedance-Phase
Analyzer. Fitting of the data to obtain the electrochemical parameters for the
Tafel and Nyquist plots were performed with the Solartron instrument software.
160
Figure 6.3 Glass cell used in the electroreduction of carbon dioxide.
161
6.3 Chapter 6: Membrane Electrode Assembly preparations
The MEAs whose data is presented in Chapters 2, 4 and 5 were
prepared following the same procedures with minor modifications. The
membranes used were either Nafion® 117 (ion power) in Chapters 2, 4 and 5
or polyvinylidene fluoride - polystyrene sulfonic acid (PPVDF-PSSA) [1] in
Chapter 5. Nafion® membrane was conditioned after purchase by boiling the
membrane in 3 % H
2
O
2
(30 % H
2
O
2
VWR International) for 1 hour followed
by boiling in deionized water for 1 hour, then boiling in 3 % H
2
SO
4
for another
hour and a final boil in deionized water for one more hour. Preparation and
conditioning of PVDF-PSSA has been published elsewhere [1, 2].
The fuel cell current collectors used were of the column type flow with
an area of 25 cm
2
, purchased from Electrochem. The gas diffusion electrode
used in all cases was Toray carbon paper TGP–H60 (Toray Corporation) for
both anode and cathode preparation. When Teflon coating was required,
Toray carbon paper TGP–H60 with 10% wet proofing (E-TEK) was used.
The catalysts were applied over the carbon paper by a direct paint
method. The paint consisted of a suspension of the catalyst in milli-Q water
and Nafion® ionomer, (default proportion is 1:3:1 for catalyst/water/Nafion®).
The catalysts used for Chapter 2, included Pt black on the cathode and Pd
and the prepared Pd/Au and Pd/Sn on the anode, with loadings of 5 to
6 mg cm
-2
. In Chapter 4, the catalysts used included Pt/Ru 50-50 (Alfa Aesar
HiSPEC 6000) for the anode in the DMFCs, and Pt black for the anode of the
162
fuel cells with hydrogen as fuel, while the cathodes were prepared with Pt
black, Pt/C and the in house prepared Pt/CFx. Loadings for DMFC were also 5
to 6 mg cm
-2
, while for the hydrogen fuel cells 3 to 4 mg cm
-2
loadings were
used. For the preparation of the MEAs for MFCs in Chapter 6, either prepared
with Nafion® or PVDF-PSSA as the membrane, the cathodes were prepared
with Pt black using the default proportion for the paint suspension and also a
catalyst loading of 5 to 6 mg cm
-2
. On the anode, no metal catalyst was used;
however a mixture of water and Nafion® containing 400 mg of water and 200
mg of Nafion® was used for binding the carbon paper to the membrane before
pressing. The prepared MEAs were hot pressed at 150
o
C and 500 KgF for 50
min in a press (PHI Corporation). Finally, the fuel cells were assembled by
inserting the MEA between the graphite current collectors and using thin
Teflon films as gaskets. A uniform torque of 36 N m
-1
was applied to each bolt
used to assemble the cell.
For the chemical fuel cells, after pressing and assembling, MEAs were
re-humidified by circulating water in the assembled fuel cell overnight at
80
o
C, and then conditioned by fast galvanodynamic scans. Afterwards, the
resistance was measured with a milliohmmeter (Hewlet Packard) to assess
the MEA preparation process. Polarization data was obtained using the Fuel
Cell Test System 890B from Scribner Associated galvanodynamically by
sweeping the current in 0.5 A steps from the OCV down to 0.1 V. The fuel and
163
oxidant used were methanol (ACS grade, EMD), formic acid, hydrogen
(compress, Gilmore) and oxygen.
6.4 Chapter 6: Microorganism growth
Shewanella oneidensis MR-1 was grown aerobically in batch culture
using a defined medium containing: 18 mM sodium lactate (Sigma-Aldrich) as
the sole carbon source, 50 mM PIPES buffer (C
8
H
18
N
2
O
6
S
2
, Sigma-Aldrich),
7.5 mM NaOH (Sigma-Aldrich), 28 mM NH
4
Cl (99 %, Sigma-Aldrich), 1.3 mM
KCl (EMD Chemicals Inc.), 4.3 mM NaH
2
PO
4
·H
2
O (EMD Chemicals Inc.), 100
mM NaCl (EMD Chemicals Inc.), 10 mL L
-1
of vitamin solution [3], 10 mL L
-1
amino acid solution and trace mineral stock solutions [4]. All solutions were
prepared using deionized water from a NANOPure Infinity model D8961 by
Barnstead. The final optical density at 600 nm (OD
600
) of the culture was
measured using a spectrophotometer (Beckman DU 530), and used to
calculate an experimental dilution of an OD
600
equal to
0.4 in the MFC anode
compartment. The appropriate volume of cells was then injected into each
experimental setup such that approximately 5 x 10
9
cells mL
-1
were present for
every evaluation and testing.
164
6.5 Chapter 6: Scanning Electron Microscopy (SEM) images for MFC
MEAs
For the SEM images, the MEAs with attached biofilms were removed
from fuel cells and immediately fixed in glutaraldehyde (2.5% in DI water,
Electron Microscopy Sciences). Samples were then subjected to a serial
dehydration protocol using increasing concentrations of ethanol (200 proof,
Pharmco-Aaper). After three final changes in 100% ethanol, the samples were
then dried using hexamethyldisilazane (Electron Microscopy Sciences). The
desiccated samples were coated with evaporated carbon and viewed using a
Zeiss-LEO 1550 VP FESEM using an in-lens secondary electron detector.
6.6 Chapter 6: MCF housing, electrochemical measurements and
inoculation procedures
For the MFC housing, a graphite current collector of the column flow
type for DMFC, with an area 25 cm
2
, purchased from Electrochem was
adapted by machining an anode chamber in house from a block of graphite
(density 1.2 g cm
-3
, conductivity 0.22 m Ω cm
-1
) (thegraphitestore.com). The
anode chamber had a volume of 70 cm
3
, and a glass window to allow visual
control of the liquid level. Figure 6.4 presents a digital picture of the assembled
fuel cell with the anode made in house. The obtained MEAs were re-hydrated
by immersing them in DI water at 60
o
C for 6 h. After assembling, the FC with
MEA inside was autoclaved (AMSCO Scientific, model SG-120) followed by
165
microbial inoculation. A PIPES buffer (50 mM PIPES, 7.5 mM NaOH, pH 7.0)
was used in the anode compartment to dilute the solution of bacteria.
Anaerobic conditions were achieved at the anode by sparging the
compartment with sterile, filtered Argon for 30 minutes. Inoculation of the
microorganism was made by two different procedures, vide infra. Sodium
lactate (100 mM) was used as the sole carbon source. After inoculation, the
MFC was placed in such a way that the MEA was in a horizontal position,
allowing the bacteria to sit over the anode by gravity, and maintained in that
way at all times. Oxygen was circulated through the dry cathode. Additional
lactate injections occurred thereafter, depending on when the cell voltage
dropped to baseline levels. MFCs were operated in batch conditions after the
addition of bacteria (no solution exchanges) over a period of several days. The
temperature of MFC was regulated at 28
o
C using heating pads.
Electrochemical measurements were performed using a Solartron
SI 1287 potentiostat. Polarization measurements were performed
galvanodynamically, sweeping the current at 0.1 mA s
-1
from the open circuit
voltage (OCV) until the cell voltage reached 0.01 V.
166
Figure 6.4 Digital picture of the assembled microbial fuel cell, showing the cathode with the
window and the cathode.
6.6.1 Inoculation Procedures
The two different inoculation procedures referred to as inoculation
procedure 1 (IP1) and inoculation procedure 2 (IP2) are described. In IP1,
after autoclaving the MFC with the MEA inside, the anode was filled with
PIPES buffer. The necessary amount of the cultivated microbes was the
added in order to obtain the OD stated in Section 6.6. The microbes were fed
with a solution of sodium lactate for a final concentration of 100 mM on the
167
anode and deaireated with Ar for 30, min while oxygen was circulated through
the cathode. The MFC was then connected to the potentiostat while the OCV
was monitored. For IP2, after the MFC was autoclaved the anode was filled
with PIPES buffer and sodium lactate for a 100 mM concentration on the
anode. The cell was then allowed to sit for 24 h. On the next day the anode
was emptied and the microbes with the desired OD in the buffer previously
deaireated were added followed by sodium lactate addition. A 1 K Ω resistance
was connected between anode and cathode and maintained at all times. The
MFC was not handled for another 24 h, while oxygen circulated through the
cathode. Next, the anode was emptied again and the previous day’s
procedure was repeated. All these steps were performed in a consistent way
for each MFC prepared. Once the second inoculation was completed, the
MFC was connected to the potentiostat and the cell voltage across the
resistance monitored.
168
6.7 Chapter 6: References
1. Prakash, G.K.S., Smart, M.C., Wang, Q.J., Atti, A., Pleynet, V., Yang,
B., McGrath, K., Olah, G.A., Narayanan, S.R., Chun, W., Valdez, T.,
and Surampudi, S., J. Fluorine Chem., 2004. 125(8): p. 1217-1230.
2. Atti, A.R., Development of High performance polymer
electrolytemembrane for direct methanol fuel cells., University of
Souther California, Ph. D., Thesis, 2000.
3. Kieft, T.L., Fredrickson, J.K., Onstott, T.C., Gorby, Y.A., Kostandarithes,
H.M., Bailey, T.J., Kennedy, D.W., Li, S.W., Plymale, A.E., Spadoni,
C.M., and Gray, M.S., Appl. Environ. Microbiol., 1999. 65(3): p. 1214-
1221.
4. Bretschger, O., Obraztsova, A., Sturm, C.A., Chang, I.S., Gorby, Y.A.,
Reed, S.B., Culley, D.E., Reardon, C.L., Barua, S., Romine, M.F.,
Zhou, J., Beliaev, A.S., Bouhenni, R., Saffarini, D., Mansfeld, F., Kim,
B.H., Fredrickson, J.K., and Nealson, K.H., Appl. Environ. Microbiol.,
2007. 73(21): p. 7003-7012.
169
Bibliography
1. Adzic, R.R., Tripkovic, A.V., and Markovic, N.M., J. Electroanal. Chem.,
1983. 150(1-2): p. 79-88.
2. Aelterman, P., Rabaey, K., Pham, H.T., Boon, N., and Verstraete, W.,
Environ. Sci. Technol., 2006. 40(10): p. 3388-3394.
3. Akiba, T., Bennetto, H.P., Stirling, J.L., and Tanaka, K., Biotechnol. Lett
. 1985. 9(11): p. 616.
4. Andersen, T.N., Hong, M., Paik, W., and Eyring, H., Ab. Pap. Am.
Chem. Soc., 1969: p. 112.
5. Antoine, O., Bultel, Y., and Durand, R., J. Electroanal. Chem., 2001.
499(1): p. 85-94.
6. Arenz, M., Stamenkovic, V., Schmidt, T.J., Wandelt, K., Ross, P.N., and
Markovic, N.M., Phys. Chem. Chem. Phys., 2003. 5(19): p. 4242-4251.
7. Arihara, K., Kitamura, F., Ohsaka, T., and Tokuda, K., J. Electroanal.
Chem., 2001. 510(1-2): p. 128-135.
8. Atti, A.R., Development of High performance polymer
electrolytemembrane for direct methanol fuel cells., University of
Souther California, Ph. D. Thesis, 2000.
9. Ayers, W.M., Spec. Publ.-R. Soc. Chem., 1994(153): p. 365-374.
10. Azuma, M., Hashimoto, K., Hiramoto, M., Watanabe, M., and Sakata,
T., J. Electrochem. Soc., 1990. 137(6): p. 1772-1778.
11. Azuma, M., Hashimoto, K., Watanabe, M., and Sakata, T., J.
Electroanal. Chem., 1990. 294(1-2): p. 299-303.
12. Bagotzky, V.S. and Osetrova, N.V., Russ. J. Electrochem., 1995. 31(5):
p. 409-425.
13. Baldauf, M. and Kolb, D.M., J. Phys. Chem. , 1996. 100(27): p. 11375-
11381.
14. Bard, A.J. and Faulkner, L.R., Electrochemical Methods. Fundamentals
and Applications. First ed. 1980: J. Wiley & Sons
170
15. Beden, B., Lamy, C., and Leger, J.M., Electrochim. Acta, 1979. 24(11):
p. 1157-1166.
16. Beden, B., Lamy, C., Bewick, A., and Kunimatsu, K., J. Electroanal.
Chem., 1981. 121: p. 343-347.
17. Beden, B., Bewick, A., Razaq, M., and Weber, J., J. Electroanal.
Chem., 1982. 139(1): p. 203-206.
18. Beden, B., Kadirgan, F., Lamy, C., and Leger, J.M., J. Electroanal.
Chem., 1982. 142(1-2): p. 171-190.
19. Beden, B., Bewick, A., and Lamy, C., J. Electroanal. Chem., 1983.
148(1): p. 147-160.
20. Beketov, N.N., Zh. Russ. Fiz. Khim. Ova, 1869. 1: p. 33.
21. Beltowska-Brzezinska, M. and Heitbaum, J., J. Electroanal. Chem.,
1985. 183(1-2): p. 167-181.
22. Blomen, L.J.M.J. and Mugerwa, M.N., Fuel Cell Systems. 1993, New
York: Plenum Press
23. Bockris, J.O.M. and Huq, A.K.M.S., Proc. R. Soc. , 1956. 237A: p.
1733.
24. Bockris, J.O.M. and S.U.M., K., Surface Electrochemistry. 1993, New
York: Plenum
25. Bond, D.R. and Lovley, D.R., Appl. Environ. Microbiol., 2003. 69(3): p.
1548-1555.
26. Breiter, M.W., Electrochemical Processes in Fuel Cells. 1969, New
York: Springer-Verlag
27. Bretschger, O., Obraztsova, A., Sturm, C.A., Chang, I.S., Gorby, Y.A.,
Reed, S.B., Culley, D.E., Reardon, C.L., Barua, S., Romine, M.F.,
Zhou, J., Beliaev, A.S., Bouhenni, R., Saffarini, D., Mansfeld, F., Kim,
B.H., Fredrickson, J.K., and Nealson, K.H., Appl. Environ. Microbiol.,
2007. 73(21): p. 7003-7012.
28. Brettar, I. and Hofle, M.G., Mar. Ecol. Prog. Ser., 1993. 94(3): p. 253-
265.
29. Cabot Corporation - http://w1.cabot-corp.com/index.jsp - (1995-2008).
171
30. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 44(1): p. 1-7.
31. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 45(2): p. 205-
231.
32. Capon, A. and Parsons, R., J. Electroanal. Chem., 1973. 44(2): p. 239-
254.
33. Capon, A. and Parsons, R., J. Electroanal. Chem., 1975. 65(1, Part 1):
p. 285-305.
34. Casella, I.G. and Contursi, M., J. Electroanal. Chem., 2006. 588(1): p.
147-154.
35. Chang, I.S., Moon, H., Bretschger, O., Jang, J.K., Park, H.I., Nealson,
K.H., and Kim, B.H., J. Microbiol. Biotechnol., 2006. 16(2): p. 163-177.
36. Chaplin, R.P.S. and Wragg, A.A., J. Appl. Electrochem., 2003. 33(12):
p. 1107-1123.
37. Chen, F.L., Chang, M.H., and Lin, M.K., Electrochim. Acta, 2007. 52(7):
p. 2506-2514.
38. Chen, S.L., Wu, B.L., and Cha, C.S., J. Electroanal. Chem., 1997.
431(2): p. 243-247.
39. Chen, S.P., Huang, T., Zhen, C.H., Zhang, Q., Gong, H., and Sun,
S.G., Spectrosc. Spectra Anal., 2003. 23(2): p. 273-275.
40. Chen, W., Kim, J., Sun, S.H., and Chen, S.W., Phys. Chem. Chem.
Phys., 2006. 8(23): p. 2779-2786.
41. Chen, W., Tang, Y., Bao, J., Gao, Y., Liu, C., Xing, W., and Lu, T., J.
Power Sources, 2007. 167(2): p. 315-318.
42. Choi, J.H., Jeong, K.J., Dong, Y., Han, J., Lim, T.H., Lee, J.S., and
Sung, Y.E., J. Power Sources, 2006. 163(1): p. 71-75.
43. Choi, J.H., Park, K.W., Park, I.S., Kim, K., Lee, J.S., and Sung, Y.E., J.
Electrochem. Soc., 2006. 153(10): p. A1812-A1817.
44. Chu, K.L., Shannon, M.A., and Masel, R.I., J. Electrochem. Soc., 2006.
153(8): p. A1562-A1567.
45. Chu, K.L., Gold, S., Subramanian, V., Lu, C., Shannon, M.A., and
Masel, R.I., J. Microelectromech. Syst., 2006. 15(3): p. 671-677.
172
46. Clavilier, J., J. Electroanal. Chem., 1987. 236(1-2): p. 87-94.
47. Clavilier, J., Femandez-Vega, A., Feliu, J.M., and Aldaz, A., J.
Electroanal. Chem., 1989. 261(1): p. 113-125.
48. Clavilier, J., Fernandez-Vega, A., Feliu, J.M., and Aldaz, A., J.
Electroanal. Chem., 1989. 258(1): p. 89-100.
49. Colmati, F., Antolini, E., and Gonzalez, E.R., Electrochim. Acta, 2005.
50(28): p. 5496-5503.
50. Colmati, F., Antolini, E., and Gonzalez, E.R., Appl. Catal., B, 2007.
73(1-2): p. 106-115.
51. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Phys. Chem. ,
1966. 70(11): p. 3761.
52. Damjanovic, A., Dey, A., and Bockris, J.O.M., J. Electrochem. Soc.,
1966. 113(7): p. 739.
53. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Electrochem.
Soc., 1967. 114(5): p. 466.
54. Damjanovic, A. and Brusic, V., Electrochim. Acta, 1967. 12(6): p. 615-
628.
55. Damjanovic, A., Genshaw, M.A., and Bockris, J.O.M., J. Electrochem.
Soc., 1967. 114(11): p. 1107.
56. Damjanovic, A. and Genshaw, M.A., Electrochim. Acta, 1970. 15(7): p.
1281-1283.
57. Damjanovic, A., Sepa, D.B., and Vojnovic, M.V., Electrochim. Acta,
1979. 24(8): p. 887-889.
58. Damjanovic, A. and G., H.P., J. Electrochem. Soc., 1988. 135(9): p.
2269-2273.
59. Damjanovic, A., J. Electrochem. Soc., 1991. 138(8): p. 2315-2320.
60. Delaney, G.M., Bennetto, H.P., Mason, J.R., Roller, S.D., Stirling, J.L.,
and Thurston, C.F., J. Chem. Technol. Biotechnol., 1984. 34(1): p. 13-
27.
61. Eggins, B.R., Ennis, C., McConnell, R., and Spence, M., J. Appl.
Electrochem., 1997. 27(6): p. 706-712.
173
62. Fernandez-Vega, A., Feliu, J.M., Aldaz, A., and Clavilier, J., J.
Electroanal. Chem., 1989. 258(1): p. 101-113.
63. Frese, K.W. and Leach, S., J. Electrochem. Soc., 1985. 132(1): p. 259-
260.
64. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J. Phys.
Chem. , 1993. 97(46): p. 12020-12029.
65. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J. Phys.
Chem. , 1994. 98(2): p. 617-625.
66. Gasteiger, H.A., Markovic, N., Ross, P.N., and Cairns, E.J., J.
Electrochem. Soc., 1994. 141(7): p. 1795-1803.
67. Gattrell, M., Gupta, N., and Co, A., J. Electroanal. Chem., 2006. 594(1):
p. 1-19.
68. Gennaro, A., Isse, A.A., Severin, M.G., Vianello, E., Bhugun, I., and
Saveant, J.M., J. Chem. Soc., Faraday Trans., 1996. 92(20): p. 3963-
3968.
69. Goodridge, F., Lister, K., and Guruswamy, V., J. Electrochem. Soc.,
1978. 125(3): p. C176-C176.
70. Gorby, Y.A., Yanina, S., McLean, J.S., Rosso, K.M., Moyles, D.,
Dohnalkova, A., Beveridge, T.J., Chang, I.S., Kim, B.H., Kim, K.S.,
Culley, D.E., Reed, S.B., Romine, M.F., Saffarini, D.A., Hill, E.A., Shi,
L., Elias, D.A., Kennedy, D.W., Pinchuk, G., Watanabe, K., Ishii, S.,
Logan, B., Nealson, K.H., and Fredrickson, J.K., Proc. Nat. Acad. Sci.
U.S.A., 2006. 103(30): p. 11358-11363.
71. Grove, W., Philos. Mag., 1839. 14: p. 127-130.
72. Ha, S., Rice, C.A., Masel, R.I., and Wieckowski, A., J. Power Sources,
2002. 112(2): p. 655-659.
73. Ha, S., Adams, B., and Masel, R.I., J. Power Sources, 2004. 128(2): p.
119-124.
74. Ha, S., Larsen, R., and Masel, R.I., J. Power Sources, 2005. 144(1): p.
28-34.
75. Ha, S., Dunbar, Z., and Masel, R.I., J. Power Sources, 2006. 158(1): p.
129-136.
174
76. Halpert, G., Narayanan, S.R., Prakash, G.K.S., Olah, G.A., Smart,
M.C., Wang, Q.J., and Surumpudi, S., Polymer electrolyte membranes
for use in fuel cells, 2002, US 6444343.
77. Hamza, M.A., Serratrice, G., Stebe, M.J., and Delpuech, J.J., J. Am.
Chem. Soc., 1981. 103(13): p. 3733-3738.
78. Hara, K., Kudo, A., and Sakata, T., J. Electroanal. Chem., 1995. 391(1-
2): p. 141-147.
79. Hara, K., Kudo, A., Sakata, T., and Watanabe, M., J. Electrochem.
Soc., 1995. 142(4): p. L57-L59.
80. He, Z., Minteer, S.D., and Angenent, L.T., Environ. Sci. Technol., 2005.
39(14): p. 5262-5267.
81. Hernandez, M.E. and Newman, D.K., Cell. Mol. Life Sci., 2001. 58(11):
p. 1562-1571.
82. Hori, Y., Kikuchi, K., and Suzuki, S., Chem. Lett., 1985(11): p. 1695-
1698.
83. Hori, Y., Kikuchi, K., Murata, A., and Suzuki, S., Chem. Lett., 1986(6):
p. 897-898.
84. Hori, Y., Murata, A., Kikuchi, K., and Suzuki, S., J. Chem. Soc., Chem.
Commun. , 1987(10): p. 728-729.
85. Hori, Y., Murata, A., and Takahashi, R., J. Chem. Soc., Faraday Trans.,
1989. 85: p. 2309-2326.
86. Hori, Y., Denki Kagaku, 1990. 58(11): p. 996-1002.
87. Hori, Y., Wakebe, H., Tsukamoto, T., and Koga, O., Electrochim. Acta,
1994. 39(11-12): p. 1833-1839.
88. Hori, Y., Takahashi, I., Koga, O., and Hoshi, N., J. Mol. Catal. A: Chem.
, 2003. 199(1-2): p. 39-47.
89. Hori, Y., Ito, H., Okano, K., Nagasu, K., and Sato, S., Electrochim. Acta,
2003. 48(18): p. 2651-2657.
90. Hoshi, N., Kawatani, S., Kudo, M., and Hori, Y., J. Electroanal. Chem.,
1999. 467(1-2): p. 67-73.
175
91. Hoshi, N. and Hori, Y., Electrochim. Acta, 2000. 45(25-26): p. 4263-
4270.
92. Huang, J., Yang, H., Huang, Q., Tang, Y., Lu, T., and Akins, D.L., J.
Electrochem. Soc., 2004. 151(11): p. A1810-A1815.
93. Igarashi, H., Fujino, T., Zhu, Y.M., Uchida, H., and Watanabe, M., CO
Tolerance of Pt alloy electrocatalysts for polymer electrolyte fuel cells
and the detoxification mechanism, 2001. 3(3): p. 306-314.
94. Ikeda, S., Takagi, T., and Ito, K., Bull. Chem. Soc. Jpn., 1987. 60(7): p.
2517-2522.
95. Ikeda, S., Hattori, A., Ito, K., and Noda, H., Electrochem., 1999. 67(1):
p. 27-33.
96. Ishimaru, S., Shiratsuchi, R., and Nogami, G., J. Electrochem. Soc.,
2000. 147(5): p. 1864-1867.
97. Iwasita, T. and Vielstich, W., J. Electroanal. Chem., 1986. 201(2): p.
403-408.
98. Jang, J.K., Pham, T.H., Chang, I.S., Kang, K.H., Moon, H., Cho, K.S.,
and Kim, B.H., Process. Biochem., 2004. 39(8): p. 1007-1012.
99. Jansen, M.M.P. and Moolhuysen, J., Electrochim. Acta, 1976. 21(1): p.
869.
100. Jayashree, R.S., Spendelow, J.S., Yeom, J., Rastogi, C., Shannon,
M.A., and Kenis, P.J.A., Electrochim. Acta, 2005. 50(24): p. 4674-4682.
101. Jeong, K.-J., Miesse, C.M., Choi, J.-H., Lee, J., Han, J., Yoon, S.P.,
Nam, S.W., Lim, T.-H., and Lee, T.G., J. Power Sources, 2007. 168(1):
p. 119-125.
102. Jitaru, M., Lowy, D.A., Toma, M., Toma, B.C., and Oniciu, L., J. Appl.
Electrochem., 1997. 27(8): p. 875-889.
103. Kaneco, S., Iiba, K., Hiei, N., Ohta, K., Mizuno, T., and Suzuki, T.,
Electrochim. Acta, 1999. 44(26): p. 4701-4706.
104. Kang, S., Lee, J., Lee, J.K., Chung, S.Y., and Tak, Y., J. Phys. Chem.
B, 2006. 110(14): p. 7270-7274.
105. Kapusta, S. and Hackerman, N., J. Electrochem. Soc., 1983. 130(3): p.
607-613.
176
106. Kelaidopoulou, A., Abelidou, E., and Kokkinidis, G., J. Appl.
Electrochem., 1999. 29(11): p. 1255-1261.
107. Kibler, L.A., El-Aziz, A.M., and Kolb, D.M., J. Mol. Catal. A: Chem. ,
2003. 199(1-2): p. 57-63.
108. Kieft, T.L., Fredrickson, J.K., Onstott, T.C., Gorby, Y.A., Kostandarithes,
H.M., Bailey, T.J., Kennedy, D.W., Li, S.W., Plymale, A.E., Spadoni,
C.M., and Gray, M.S., Appl. Environ. Microbiol., 1999. 65(3): p. 1214-
1221.
109. Kim, B.H., Kim, H.J., Hyun, M.S., and Park, D.H., J. Microbiol.
Biotechnol., 1999. 9(2): p. 127-131.
110. Kim, B.H., Ikeda, T., Park, H.S., Kim, H.J., Hyun, M.S., Kano, K.,
Takagi, K., and Tatsumi, H., Biotechnol. Tech., 1999. 13(7): p. 475-478.
111. Kim, B.H., Park, H.S., Kim, H.J., Kim, G.T., Chang, I.S., Lee, J., and
Phung, N.T., Appl. Microbiol. Biotechnol., 2004. 63(6): p. 672-681.
112. Kim, G.T., Hyun, M.S., Chang, I.S., Kim, H.J., Park, H.S., Kim, B.H.,
Kim, S.D., Wimpenny, J.W.T., and Weightman, A.J., J. Appl. Microbiol.,
2005. 99(4): p. 978-987.
113. Kim, H.J., Hyun, M.S., Chang, I.S., and Kim, B.H., J. Microbiol.
Biotechnol., 1999. 9(3): p. 365-367.
114. Kim, H.J., Park, H.S., Hyun, M.S., Chang, I.S., Kim, M., and Kim, B.H.,
Enzyme Microb. Technol., 2002. 30(2): p. 145-152.
115. Kim, J.R., Cheng, S., Oh, S.E., and Logan, B.E., Environ. Sci. Technol.,
2007. 41(3): p. 1004-1009.
116. Kim, M., Park, J.-N., Kim, H., Song, S., and Lee, W.-H., J. Power
Sources, 2006. 163(1): p. 93-97.
117. Klobukov, N., J. Prakt. Chem., 1887. 34: p. 62.
118. Koleli, F., Yesilkaynak, T., and Balun, D., Fresenius Environ. Bull.,
2003. 12(10): p. 1202-1206.
119. Koleli, F., Atilan, T., Palamut, N., Gizir, A.M., Aydin, R., and Hamann,
C.H., J. Appl. Electrochem., 2003. 33(5): p. 447-450.
120. Koper, M.T.M. and Sluyters, J.H., J. Electroanal. Chem., 1994. 371(1-
2): p. 149-159.
177
121. Kundu, A., Jang, J.H., Gil, J.H., Jung, C.R., Lee, H.R., Kim, S.H., Ku,
B., and Oh, Y.S., J. Power Sources, 2007. 170(1): p. 67-78.
122. Kunimatsu, K. and Kita, H., J. Electroanal. Chem., 1987. 218(1-2): p.
155-172.
123. Kunimatsu, M. and Okada, T., Electrochem. Solid-State Lett., 2004.
7(11): p. A389-A390.
124. Lamy, C., Leger, J.M., Clavilier, J., and Parsons, R., J. Electroanal.
Chem., 1983. 150(1-2): p. 71-77.
125. Lamy, C., Electrochim. Acta, 1984. 29(11): p. 1581-1588.
126. Lamy, E., Nadjo, L., and Saveant, J.M., J. Electroanal. Chem., 1977.
78(2): p. 403-407.
127. Larminie, J. and Dicks, A., Fuel Cell Systems Explained. 2nd ed. 2003,
Chichester: J. Wiley
128. Larsen, R. and Masel, R.I., Electrochem. Solid-State Lett., 2004. 7(6):
p. A148-A150.
129. Larsen, R., Ha, S., Zakzeski, J., and Masel, R.I., J. Power Sources,
2006. 157(1): p. 78-84.
130. Lawson, D.D., Moacanin, J., Scherer, K.V., Terranova, T.F., and
Ingham, J.D., J. Fluorine Chem. , 1978. 12(3): p. 221-236.
131. Leger, V.Z., Nonaqueous cell having a MnO2/poly-carbon fluoride
cathode, 1982, US 4327166.
132. Li, X.G. and Hsing, I.M., Electrochim. Acta, 2006. 51(17): p. 3477-3483.
133. Lieber, A., Monatsh. Chem., 1895. 16: p. 211.
134. Lieber, A., Monatsh. Chem., 1897. 18: p. 528.
135. Liu, H., Ramnarayanan, R., and Logan, B.E., Environ. Sci. Technol.,
2004. 38(7): p. 2281-2285.
136. Liu, H. and Logan, B.E., Environ. Sci. Technol., 2004. 38(14): p. 4040-
4046.
137. Liu, H., Cheng, S.A., and Logan, B.E., Environ. Sci. Technol., 2005.
39(14): p. 5488-5493.
178
138. Liu, H., Cheng, S.A., and Logan, B.E., Environ. Sci. Technol., 2005.
39(2): p. 658-662.
139. Liu, Z.L., Hong, L., Tham, M.P., Lim, T.H., and Jiang, H.X., J. Power
Sources, 2006. 161(2): p. 831-835.
140. Logan, B.E., Murano, C., Scott, K., Gray, N.D., and Head, I.M., Water
Res., 2005. 39(5): p. 942-952.
141. Lu, G.Q., Liu, F.Q., and Wang, C.-Y., Electrochem. Solid-State Lett.,
2005. 8(1): p. A1-A4.
142. Manohar, A.K., Bretschger, O., Nealson, K.H., and Mansfeld, F.,
Bioelectrochem., 2008. 72(2): p. 149-154.
143. Manohar, A.K., Bretschger, O., Nealson, K.H., and Mansfeld, F.,
Electrochim. Acta, 2008. 53(9): p. 3508-3513.
144. Markovic, N.M., Gasteiger, H.A., Ross, P.N., Jiang, X., Villegas, I., and
Weaver, M.J., Electrochim. Acta, 1995. 40(1): p. 91-98.
145. Markovic, N.M. and Ross, P.N., Surf. Sci. Rep. , 2002. 45(4-6): p. 121-
229.
146. McGrath, K.M., Electrochemical studies of membranes, catalysts, and
fuels for direct oxidation fuel cells University of Southern California, Ph.
D. Thesis, 2005.
147. Meyer, T.E., Tsapin, A.I., Vandenberghe, I., De Smet, L., Frishman, D.,
Nealson, K.H., Cusanovich, M.A., and Van Beeumen, J.J., OMICS A: J.
Integr. Biol., 2004. 8(1): p. 57-77.
148. Miesse, C.M., Jung, W.S., Jeong, K.-J., Lee, J.K., Lee, J., Han, J.,
Yoon, S.P., Nam, S.W., Lim, T.-H., and Hong, S.-A., J. Power Sources,
2006. 162(1): p. 532-540.
149. Min, B. and Logan, B.E., Environ. Sci. Technol., 2004. 38(21): p. 5809-
5814.
150. Min, B.K., Cheng, S.A., and Logan, B.E., Water Res., 2005. 39(9): p.
1675-1686.
151. Mizuno, T., Ohta, K., Sasaki, A., Akai, T., Hirano, M., and Kawabe, A.,
Energy Sources, 1995. 17(5): p. 503-508.
179
152. Moon, H., Chang, I.S., and Kim, B.H., Bioresour. Technol., 2006. 97(4):
p. 621-627.
153. Myers, C.R. and Nealson, K.H., Science, 1988. 240(4857): p. 1319-
1321.
154. Nakato, Y., Yano, S., Yamaguchi, T., and Tsubomura, H., Denki
Kagaku, 1991. 59(6): p. 491-498.
155. Noda, H., Ikeda, S., Oda, Y., and Ito, K., Chem. Lett., 1989(2): p. 289-
292.
156. Noda, H., Ikeda, S., Oda, Y., Imai, K., Maeda, M., and Ito, K., Bull.
Chem. Soc. Jpn., 1990. 63(9): p. 2459-2462.
157. Oh, S., Min, B., and Logan, B.E., Environ. Sci. Technol., 2004. 38(18):
p. 4900-4904.
158. Oh, S.E. and Logan, B.E., Appl. Microbiol. Biotechnol., 2006. 70(2): p.
162-169.
159. Ohta, K., Kawamoto, M., Mizuno, T., and Lowy, D.A., J. Appl.
Electrochem., 1998. 28(7): p. 717-724.
160. Okada, T., Moller-Holst, S., Gorseth, O., and Kjelstrup, S., J.
Electroanal. Chem., 1998. 442(1-2): p. 137-145.
161. Olah, G.A., Goeppert, A., and Prakash, G.K.S., Beyond Oil and Gas:
The Methanol Economy. 2006, Weinheim: John Wiley & Sons, Inc
162. Paik, W., Andersen, T.N., and Eyring, H., Electrochim. Acta, 1969.
14(12): p. 1217.
163. Park, H.S., Kim, B.H., Kim, H.S., Kim, H.J., Kim, G.T., Kim, M., Chang,
I.S., Park, Y.K., and Chang, H.I., Anaer., 2001. 7(6): p. 297-306.
164. Parsons, R. and Vandernoot, T., J. Electroanal. Chem., 1988. 257(1-2):
p. 9-45.
165. Pham, C.A., Jung, S.J., Phung, N.T., Lee, J., Chang, I.S., Kim, B.H., Yi,
H., and Chun, J., FEMS Microbiol. Lett., 2003. 223(1): p. 129-134.
166. Pham, T.H., Jang, J.K., Moon, H.S., Chang, I.S., and Kim, B.H., J.
Microbiol. Biotechnol., 2005. 15(2): p. 438-441.
167. Pletcher, D. and Solis, V., Electrochim. Acta, 1982. 27(6): p. 775-782.
180
168. Prakash, G.K.S., Smart, M.C., Wang, Q.J., Atti, A., Pleynet, V., Yang,
B., McGrath, K., Olah, G.A., Narayanan, S.R., Chun, W., Valdez, T.,
and Surampudi, S., J. Fluorine Chem., 2004. 125(8): p. 1217-1230.
169. Prakash, G.K.S., Smart, M.C., Olah, G.A., Narayanan, S.R., Chun, W.,
Surampudi, S., and Halpert, G., J. Power Sources, 2007. 173: p. 102-
109.
170. Prakash, G.K.S., Krause, F., Viva, F.A., and Olah, G.A., Unpublished
work, 2008.
171. Prakash, G.K.S., Suresh, P., Viva, F., and Olah, G.A., J. Power
Sources, 2008. 181(1): p. 79-84.
172. Rabaey, K., Boon, N., Siciliano, S.D., Verhaege, M., and Verstraete,
W., Appl. Environ. Microbiol., 2004. 70(9): p. 5373-5382.
173. Rabaey, K., Ossieur, W., Verhaege, M., and Verstraete, W., Water Sci.
Technol., 2005. 52(1-2): p. 515-523.
174. Rabaey, K., Clauwaert, P., Aelterman, P., and Verstraete, W., Environ.
Sci. Technol., 2005. 39(20): p. 8077-8082.
175. Rabaey, K., Boon, N., Hofte, M., and Verstraete, W., Environ. Sci.
Technol., 2005. 39(9): p. 3401-3408.
176. Reguera, G., McCarthy, K.D., Mehta, T., Nicoll, J.S., Tuominen, M.T.,
and Lovley, D.R., Nature, 2005. 435(7045): p. 1098-1101.
177. Rhee, Y.W., Ha, S.Y., and Masel, R.I., J. Power Sources, 2003. 117(1-
2): p. 35-38.
178. Rice, C., Ha, R.I., Masel, R.I., Waszczuk, P., Wieckowski, A., and
Barnard, T., J. Power Sources, 2002. 111(1): p. 83-89.
179. Rice, C., Ha, S., Masel, R.I., and Wieckowski, A., J. Power Sources,
2003. 115(2): p. 229-235.
180. Roller, S.D., Bennetto, H.P., Delaney, G.M., Mason, J.R., Stirling, J.L.,
and Thurston, C.F., J. Chem. Technol. Biotechnol., 1984. 34(1): p. 3-
12.
181. Rosenbaum, M., Schroder, U., and Scholz, F., Environ. Sci. Technol.,
2005. 39(16): p. 6328-6333.
181
182. Roychowdhury, C., Matsumoto, F., Zeldovich, V.B., Warren, S.C.,
Mutolo, P.F., Ballesteros, M., Wiesner, U., Abruna, H.D., and DiSalvo,
F.J., Chem. Mater., 2006. 18(14): p. 3365-3372.
183. Royer, E., C.R. Acad.Sci., 1870. 70: p. 73.
184. Rozendal, R.A., Hamelers, H.V.M., and Buisman, C.J.N., Environ. Sci.
Technol., 2006. 40(17): p. 5206-5211.
185. Ryu, J., Andersen, T.N., and Eyring, H., J. Phys. Chem. , 1972. 76(22):
p. 3278-&.
186. Schmidt, T.J., Behm, R.J., Grgur, B.N., Markovic, N.M., and Ross Jr.,
P.N., Langmuir, 2000. 16: p. 8159.
187. Schmidt, T.J., Stamenkovic, V., Arenz, M., Markovic, N.M., and Ross,
P.N., Electrochim. Acta, 2002. 47(22-23): p. 3765-3776.
188. Schwalb, C., Chapman, S.K., and Reid, G.A., Biochem., 2003. 42(31):
p. 9491-9497.
189. Scott, J.H. and Nealson, K.H., J. Bacteriol., 1994. 176(11): p. 3408-
3411.
190. Scott, K. and Argyropoulos, P., J. Power Sources, 2004. 137(2): p. 228-
238.
191. Sepa, D.B., Vracar, L.M., Vojnovic, M.V., and Damjanovic, A.,
Electrochim. Acta, 1986. 31(11): p. 1401-1402.
192. Stirling, J.L., Bennetto, H.P., Delaney, G.M., Mason, J.R., Roller, S.D.,
Tanaka, K., and Thurston, C.F., Biochem. Soc. Trans., 1983. 11(4): p.
451-453.
193. Sullivan, B.P., Krist, K., and Guard, H.E., eds. Electrochemical and
Electrocatalytic Reactions of Carbon Dioxide. 1993, Elsevier:
Amsterdam.
194. Surampudi, S., Narayanan, S.R., Vamos, E., Frank, H., Halpert, G.,
Laconti, A., Kosek, J., Prakash, G.K.S., and Olah, G.A., J. Power
Sources, 1994. 47(3): p. 377-385.
195. Tacconi, N.R.d., Leger, J.M., Beden, B., and Lamy, C., J. Electroanal.
Chem., 1982. 134(1): p. 117-130.
182
196. Toda, T., Igarashi, H., Uchida, H., and Watanabe, M., Enhancement of
the electroreduction of oxygen on Pt alloys with Fe, Ni, and Co, 1999.
146(10): p. 3750-3756.
197. Tomita, Y., Teruya, S., Koga, O., and Hori, Y., J. Electrochem. Soc.,
2000. 147(11): p. 4164-4167.
198. U.S. Code of Federal Regulations - FDA, CFR Part 21 - 186.1316,
172.515
199. Uhm, S.Y., Chung, S.T., and Lee, J.Y., Electrochem. Commun., 2007.
9(8): p. 2027-2031.
200. Vassiliev, Y.B., Bagotzky, V.S., Osetrova, N.V., and Mikhailova, A.A., J.
Electroanal. Chem., 1985. 189(2): p. 311-324.
201. Vassiliev, Y.B., Bagotzky, V.S., Khazova, O.A., and Mayorova, N.A., J.
Electroanal. Chem., 1985. 189(2): p. 295-309.
202. Vassiliev, Y.B., Bagotsky, V.S., Osetrova, N.V., Khazova, O.A., and
Mayorova, N.A., J. Electroanal. Chem., 1985. 189(2): p. 271-294.
203. Vela, M.E., Lezna, R.O., Detacconi, N.R., Arvia, A.J., Beden, B., Hahn,
F., and Lamy, C., J. Electroanal. Chem., 1992. 323(1-2): p. 289-302.
204. Wang, G., Sun, G., Zhou, Z., Liu, J., Wang, Q., Wang, S., Guo, J.,
Yang, S., Xin, Q., and Yi, B., Electrochem. Solid-State Lett., 2005. 8(1):
p. A12-A16.
205. Waszczuk, P., Barnard, T.M., Rice, C., Masel, R.I., and Wieckowski, A.,
Electrochem. Commun., 2002. 4(7): p. 599-603.
206. Watanabe, M., Shibata, M., and Motoo, S., J. Electroanal. Chem., 1985.
187(1): p. 161-174.
207. Watanabe, M., Uchida, M., and Motoo, S., J. Electroanal. Chem., 1987.
229(1-2): p. 395-406.
208. Watanabe, M., Zhu, Y.M., and Uchida, H., Oxidation of CO on a Pt-Fe
alloy electrode studied by surface enhanced infrared reflection-
absorption spectroscopy, 2000. 104(8): p. 1762-1768.
209. Watanabe, N. and Morigaki, K., Denki Kagaku, 1979. 47(3): p. 169-173.
210. Weber, M., Wang, J.T., Wasmus, S., and Savinell, R.F., J.
Electrochem. Soc., 1996. 143(7): p. L158-L160.
183
211. Wesseler, E.P., Iltis, R., and Clark, L.C., J. Fluorine Chem. , 1977. 9(2):
p. 137-146.
212. Whitacre, J.F., West, W.C., Smart, M.C., Yazami, R., Prakash, G.K.S.,
Hamwi, A., and Ratnakumar, B.V., Electrochem. Solid-State Lett., 2007.
10(7): p. A166-A170.
213. Wolter, O., Willsau, J., and Heitbaum, J., J. Electrochem. Soc., 1985.
132(7): p. 1635-1638.
214. Wroblowa, H., Rao, M.L.B., Damjanovic. A, and Bockris, J.O.M., J.
Electroanal. Chem., 1967. 15(2-3): p. 139.
215. Wroblowa, H.S., Yen Chi, P., and Razumney, G., J. Electroanal.
Chem., 1976. 69(2): p. 195-201.
216. Wynn, J.E., Proc., Ann. Power Sources Conf., 1960. 14: p. 52-55.
217. Xie, F., Tian, Z., Meng, H., and Shen, P.K., J. Power Sources, 2005.
141(2): p. 211-215.
218. Yang, L.X., Allen, R.G., Scott, K., Christenson, P., and Roy, S., J.
Power Sources, 2004. 137(2): p. 257-263.
219. Yang, Y.Y. and Sun, S.G., J. Phys. Chem. B, 2002. 106(48): p. 12499-
12507.
220. Yeom, J., Mozsgai, G.Z., Flachsbart, B.R., Choban, E.R., Asthana, A.,
Shannon, M.A., and Kenis, P.J.A., Sens. Actuators, B 2005. 107(2): p.
882-891.
221. Yeom, J., Jayashree, R.S., Rastogi, C., Shannon, M.A., and Kenis,
P.J.A., J. Power Sources, 2006. 160(2): p. 1058-1064.
222. Yu, X.W. and Ye, S.Y., J. Power Sources, 2007. 172(1): p. 133-144.
223. Yu, X.W. and Ye, S.Y., J. Power Sources, 2007. 172(1): p. 145-154.
224. Zhang, L.J., Wang, Z.Y., and Xia, D.G., J. Alloys Compd., 2006. 426(1-
2): p. 268-271.
225. Zhang, L.L., Tang, Y.W., Bao, J.C., Lu, T.H., and Li, C., J. Power
Sources, 2006. 162(1): p. 177-179.
226. Zhang, L.L., Lu, T.H., Bao, J.C., Tang, Y.W., and Li, C., Electrochem.
Commun., 2006. 8(10): p. 1625-1627.
184
227. Zhou, A.H., He, D.L., Xie, N.X., Xie, Q.J., Nie, L.H., and Yao, S.Z.,
Electrochim. Acta, 2000. 45(24): p. 3943-3950.
228. Zhou, X., Xing, W., Liu, C., and Lu, T., Electrochem. Commun., 2007.
9(7): p. 1469-1473.
229. Zhu, Y.M., Ha, S.Y., and Masel, R.I., J. Power Sources, 2004. 130(1-2):
p. 8-14.
230. Zhu, Y.M., Khan, Z., and Masel, R.I., J. Power Sources, 2005. 139(1-2):
p. 15-20.
Abstract (if available)
Abstract
Although fuel cells are based on a concept dated more than 170 years ago, thorough and focused research did not take place in this field until the last five decades. Therefore, considering that fuel cells are the main candidates for portable power for the near future, there are questions that need to be answered and problems that need to be solved if wide applications of fuel cells are to be achieved. The research elaborated in this thesis was carried out to solve some of the problems and overcome some of the challenges. On the subject of Direct Formic Acid Fuel Cells, a search for a better and cheaper catalyst than those already known was conducted, and at the same time conditions for the Membrane Electrode Assembly preparation were explored. Concomitantly, the electrochemical reduction of carbon dioxide was investigated under conditions that primarily yield formic acid. The research was aimed at maximizing the efficiency as well as the rate of the carbon dioxide reduction.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Studies on direct oxidation formic acid fuel cells: advantages, limitations and potential
PDF
Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction
PDF
Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)
PDF
The modification of catalysts and their supports for use in various fuel cells
PDF
PSSA-PVDF semi-IPN blends for direct methanol fuel cells
PDF
Blends of polystyrene sulfonic acid copolymers and polyvinylidene fluoride as polyelectrolyte membranes
PDF
Effect of reductants on sodium chromate-induced cytotoxicity and morphological transformation to C3H/10T1/2 Cl 8 mouse embryo cells
PDF
Electrokinetic transport of Cr(VI) and integration with zero-valent iron nanoparticle and microbial fuel cell technologies for aquifer remediation
Asset Metadata
Creator
Viva, Federico Andres
(author)
Core Title
Studies on direct methanol, formic acid and related fuel cells in conjunction with the electrochemical reduction of carbon dioxide
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Degree Conferral Date
2009-05
Publication Date
10/24/2010
Defense Date
01/28/2009
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
CO₂ electrochemical reduction,formic acid electro-oxidation,formic acid fuel cell,microbial fuel cell,OAI-PMH Harvest,oxygen reduction reaction (ORR)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Prakash, G.K. Surya (
committee chair
), Hoegen-Esch, Thieo (
committee member
), Shing, Kathering S. (
committee member
)
Creator Email
federico_v@hotmail.com,viva@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m2105
Unique identifier
UC1487340
Identifier
etd-Viva-2731 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-231705 (legacy record id),usctheses-m2105 (legacy record id)
Legacy Identifier
etd-Viva-2731.pdf
Dmrecord
231705
Document Type
Dissertation
Rights
Viva, Federico Andres
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
CO₂ electrochemical reduction
formic acid electro-oxidation
formic acid fuel cell
microbial fuel cell
oxygen reduction reaction (ORR)