PRESSURE PYROLYSED NON-PRECIOUS OXYGEN REDUCTION CATALYSTS FOR
PROTON EXCHANGE MEMBRANE FUEL CELLS
By
Vijayadurga Nallathambi

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Chemical Engineering
2011

ABSTRACT
PRESSURE PYROLYSED NON-PRECIOUS OXYGEN REDUCTION CATALYSTS FOR
PROTON EXCHANGE MEMBRANE FUEL CELLS
By
Vijayadurga Nallathambi
Worldwide energy demand has driven long-term efforts towards developing a clean,
hydrogen-based energy economy. Polymer electrolyte membrane fuel cells (PEMFC) are low
emissions and high efficiency devices that utilize the power of hydrogen and are a key enabling
technology for the hydrogen economy. Carbon supported platinum-black is the state-of the art
catalyst for oxygen reduction in a PEMFC because it can withstand the acidic environment.
However, the high cost and low abundance of this precious metal has limited large-scale
commercialization of PEMFCs. Current efforts focus on developing alternative inexpensive,
non-noble metal-based catalysts for oxygen reduction with performance comparable to
conventional platinum based electrocatalysts.
In this work, inexpensive metal-nitrogen-carbon (MNC) catalysts have been synthesized by
pyrolyzing transition metal and nitrogen precursors together with high surface area carbon
materials in a closed, constant-volume quartz tube. High pressure generated due to nitrogen
precursor evaporation lead to increased surface nitrogen content in the catalysts post-pyrolysis.
Electrochemical oxygen reduction activity of MNC catalysts was analyzed using half-cell
Rotating Ring Disc Electrode (RRDE) studies. The effect of nitrogen precursor morphology on
the generation of active sites has been explored in detail. By increasing the Nitrogen/Carbon
ratio of the nitrogen precursor, the accessible active site density increased by reducing carbon
deposition in the pores of the carbon support during pyrolysis. The most active catalysts were

obtained using melamine, having a N/C ratio of 2. Single PEMFC measurements employing
MNC catalysts as cathodes indicated kinetic current density as high as 15 A cm-3 at 0.8 ViR-free
and over 100 h of stable current at 0.5 V were observed. Effects of carbon free ammonia
generating solid nitrogen precursors such as urea and ammonium carbamate were also studied.
These precursors etched the carbon support and increased the porosity, particularly micro and
mesopores of the catalysts that led to increased active site density and reduced oxygen transport
hindrances respectively.
Collaborative efforts with the University of New Mexico facilitated XPS characterization
of MNC catalysts. XPS analyses indicated that pyridinic nitrogen sites, present in the edge plane
of the catalysts and pyridinic nitrogen coordinated to transition metals correlated to oxygen
reduction activity. Further insight into the role of transition metal and the structure of active site
was gained through EXAFS measurements, carried out in collaboration with Northeastern
University. Electrochemical studies performed in the presence of poisoning anions such as
cyanide in alkaline environment indicated a 25% decrease in oxygen reduction activity,
suggesting that the metal is part of the active sites and participates in oxygen reduction. In-situ
EXAFS analysis of the catalysts indicated the active reaction site for oxygen reduction to be Fe
metal coordinated to 4 nitrogen atoms. These low cost MNC catalysts find direct application in
Proton Exchange Membrane Fuel cells for transportation applications, where there is a huge
drive to improve the economy of the fuel cell by reducing the costs associated with state-of the
art platinum-based catalysts.

Copyright by
VIJAYADURGA NALLATHAMBI
2011

dedicated
to my family

v

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to my advisor Dr. Scott Calabrese
Barton, for giving me this opportunity to work under him in the field of fuel cells. He has
been a constant source of motivation and his valuable critiques have shaped my research
and career. I sincerely appreciate the freedom he gave with regard to the experiments and
his encouragement to participate in various technical symposiums. I should definitely
thank him for the kind of exposure he gave me, from presenting in national level
conferences, working collaboratively with various research groups to working in a
national lab. I admire his patience in correcting manuscripts and I hope to carry forward
the valuable things learned from him. Thank you Dr. Barton, for your esteemed guidance
and valuable support. I would also like to thank you as a Graduate coordinator for
admitting me to the PhD program at MSU and for your guidance with regard to finalizing
my course plan.
I am thankful to all my committee members Dr. Jeff Sakamoto, Dr. Robert Ofoli
and Dr. Greg Swain for agreeing to be part of my committee and their valuable time.
Their feedback and suggestions during comprehensive oral presentation were very
helpful to improve the quality of my thesis further. I appreciate their flexibility and
support for completing this dissertation in a timely fashion.
Also, I would like to extend my thanks to our collaborators Dr. Frank Gibbard and
Dr. Arthur Kaufmann of Gibbard Research Corporation for their guidance when we
worked together on the NSF project, Dr. Sanjeev Mukerjee and his group of Northeastern
University for XAFS experiments carried out at Brookhaven National Lab and Dr.
Kateryna Artushkova of University of New Mexico for her guidance with XPS
vi

experiments and analyses. My thanks to North-Eastern University folks, Matt, Urszula
and my good friend Naggappan for their help with regard to XAFS measurements and
analysis. Dr. Kev Adjemian and his group members Nilesh, Tahee and Chunmei at
Nissan Technical Center were very supportive as well and facilitated fuel cells
characterization of our catalysts at their facility.
I take this opportunity to extend my thanks to all my wonderful group members. I
would like to acknowledge Dr. Raman Kothandaraman for his collaboration in this
project and for helping me to settle down during my initial days at MSU. My special
thanks to all my colleagues Leena, Hao, Hanzi, Nate, Harshal, Dr. Chaminda and Dr.
Piyush for the wonderful time spent in the lab and office. Nate has been very helpful and
assisted me with N2 adsorption measurements of the catalysts. I would like to
acknowledge undergraduates Erik and Raul for their assistance with RDE
characterization. Overall, it was a very pleasant experience to be part of the scb research
group.
I would like the acknowledge the help and support from the CHEMS office staff
Jennifer Sommerville, Jennifer Peterman, Donna Fernandaz, Lauren brown, Nicole
Shook and JoAnn Peterson throughout my Ph.D.
My four years of student life at MSU would not be so memorable without my
friends. I am blessed to have such a wonderful friend Anchita, who has been such a great
support and motivation throughout my Ph.D. and personal life and will be in the years to
come. I cherish the memorable time spent with her during my stay at Lansing. My sincere
thanks go to my other dear friends Leonardo, Nikita, Prachi, Ezhiyl, Rengarajan, Raghav,
Abinand and others for their wonderful company and support. Many thanks to the support
and encouragement extended from my other friends right from my school life, CECRI
vii

and University of South Carolina. I would like to acknowledge all my teachers and
professors from school and college for their part in shaping up my career. Also, I would
like to thank my Masters thesis advisor Dr. Branko Popov of University of South
Carolina for introducing me to the field of catalysis in fuel cells.
My gratitude is endless to my parents Mr. Nallathambi and Mrs. Neelavathy, my
parents-in law, my sisters and all other family members for their constant love, support
and blessings. They stood by me during all the hard times, constantly motivating me. I
thank my father for instigating the values of education right from childhood days and for
assisting me to get the education I wanted. It is a true bliss to have Anbu as my husband,
who with his unswerving love, care, patience and motivation right from my B.Tech years
at CECRI, made this possible without any hardships and his contribution to this thesis is
inexplicable. I am who I am because of you Anbu and thank you for that.
The financial support from National Science Foundation (NSF) and Department
of Energy (DOE) is gratefully acknowledged.

viii

TABLE OF CONTENTS

List of Tables

xi

List of Figures

xii

Chapter 1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8

Introduction
Hydrogen Economy and Fuel Cells Systems
Comparison of Fuel Cell Technologies
Proton Exchange Membrane Fuel Cell
Electrocatalytic Oxygen Reduction Reaction
Non-precious Metal-N4 Macrocycles – Based Catalysts
Metal-Nitrogen-Carbon (MNC) Catalysts
Activity and Durability Evaluation for NPMC
References

1
1
2
4
8
9
11
16
21

Chapter 2

Metal-Nitrogen-Carbon Catalysts Prepared By High-Pressure
Pyrolysis for Low-Temperature Fuel Cells
Abstract
Introduction
Experimental
2.3.1 Catalyst Synthesis
2.3.2 Physical and Chemical Characterization
2.3.3 Electrochemical Characterization
Results and Discussion
Conclusions
References

31
31
32
34
34
36
37
38
58
59

Nitrogen Precursor Effects in Iron – Nitrogen – Carbon Oxygen
Reduction Catalysts
Abstract
Introduction
Experimental
3.3.1 Electrochemical Characterization
3.3.2 Fuel Cell Evaluation
3.3.3 Physical and Chemical Characterization
Results and Discussion
3.4.1 Oxygen Reduction Activity and Selectivity
3.4.2 Composition Studies by XPS
3.4.3 Optimization of Nominal N Content
3.4.4 Pyrolysis Process Optimization
3.4.5 Fuel Cell Measurements
Conclusions
References

65
65
66
67
70
70
72
73
73
77
82
85
89
93
95

2.1
2.2
2.3

2.4
2.5
2.6
Chapter 3
3.1
3.2
3.3

3.4

3.5
3.6

ix

Chapter 4
4.1
4.2
4.3
4.4
4.5

4.6
4.7
Chapter 5
5.1
5.2
5.3
5.4
5.5

5.6

5.7
5.8
Chapter 6

Ammonia Generating Nitrogen Precursors’ Effects in Iron –
Nitrogen – Carbon Oxygen Reduction Catalysts
Abstract
Introduction
Catalyst Synthesis
Experimental
Results and Discussion
4.5.1 Structural Characterization
4.5.2 Electrochemical Characterization
4.5.3 Compositional analysis through XPS
Conclusions
References

100
100
102
103
106
107
107
112
115
127
128

Role of Transition Metal in Oxygen Reduction Catalysis and an
Insight into Catalytic Active Site Structure
Abstract
Introduction
5.2.1 X-Ray Absorption Spectroscopy – A Brief Overview
Catalyst Synthesis
Experimental
5.4.1 X-Ray Absorption Spectroscopy – Experimental
Results and Discussion
5.5.1 Optimization of Nominal Fe Content
5.5.2 Metal-Centered Oxygen Reduction Activity
X-ray Absorption Spectroscopy
5.6.1 EXAFS data analysis
5.6.2 XANES
Conclusions
References

132
132
133
136
138
139
139
142
142
150
152
152
158
162
164

Summary and Future Directions

169

x

LIST OF TABLES

Table 1.1

Synthesis and performance of MNC catalysts.

19

Table 2.1

Catalyst yield with varying bipyridine content.

41

Table 2.2

Elemental quantification by XPS.

43

Table 2.3

Deconvolution results (at%) for N1s high-resolution spectra.

47

Table 3.1

Properties of N precursors used.

69

Table 3.2

Elemental quantification of catalysts by XPS.

80

Table 3.3

XRD Analysis of Melamine based catalysts pyrolyzed at different
temperatures.

88

Table 3.4

Catalytic Activity of MNC catalysts

91

Table 4.1

Structure of N precursors used and their decomposition reactions.

104

Table 4.2

Structural characteristics of carbon support and catalysts.

110

Table 4.3

Elemental quantification (wt%) by XPS.

117

Table 4.4

Deconvolution results (wt %) for N1s high-resolution spectra.

120

Table 5.1

Summary of electrochemical and structural parameters obtained for
Melamine based Fe-N-C catalysts synthesized using varying
nominal Fe loading.

146

Table 5.2

ICP-MS Characterization of MNC catalysts and carbon precursors.

149

Table 5.3

Fe K-edge EXAFS fit results for Melamine based Fe-N-C catalysts
under ex-situ and in-situ (O2 saturated 0.1M HClO4, room
temperature).

157

Table 5.4

Fe K-edge EXAFS fit results for Melamine based Fe-N-C catalysts 160
under in-situ (O2 saturated 0.1M HClO4, room temperature) for
varying Fe-N co-ordination

xi

LIST OF FIGURES

Figure 1.1

Mechanisms and applications of different types of Fuel Cells.

3

Figure 1.2

Proton Exchange Membrane Fuel Cell (PEMFC).

5

Figure 1.3

Current density and over potential plot showing the major
irreversible losses associated with fuel cell operation.

7

Figure 1.4

Proposed catalytic active site structure.

13

Figure 2.1

Nitrogen functional groups present on pyrolyzed metal-nitrogencarbon catalyst surfaces.

35

Figure 2.2

Effect of nitrogen content on surface morphology. (a) Differential
surface area distribution of various catalysts obtained from BJH
desorption. (b) Observed BET surface area for various catalysts.

40

Figure 2.3

Surface and bulk nitrogen content obtained through XPS and CHN
analyses respectively as a function of nominal nitrogen loading.

44

Figure 2.4

High-resolution N1s spectra deconvoluted for a subset of samples (a) 45
As synthesized catalyst (1.5 wt% nominal N, no Fe) before pyrolysis;
(b) As-synthesized catalyst (1.5 wt% nominal N) before pyrolysis; (c)
Pyrolyzed catalyst (1.5 wt% nominal N); (d) Pyrolyzed catalyst (10.3
wt% nominal N).

Figure 2.5

Surface nitrogen content, in pyridinic form (squares) and graphitic
form (circles),as determined by XPS for varying nominal nitrogen
loading. Closed symbols indicate N content of final catalysts, open
symbols indicate observed content of unpyrolyzed precursors.

Figure 2.6

Oxygen reduction at thin-film rotating disk electrodes with Fe-N-C 49
catalysts of varying composition in O2-saturated, 1 N aqueous
sulfuric acid, 40°C. Pseudo-steady state polarization: Scan rate 0.5
-1
mV s , rotation rate 1200 rpm. (a) 6.9 wt % nominal N, (b) 10 .3 wt
% (c) 11.3 wt %, (d) 10.3 wt % w/o Fe, (e) Ketjen carbon black (f) 20
wt % Pt/C [ETEK], (g) Current density at 0.65 V/SHE as a function
of nominal nitrogen content.

xii

48

Figure 2.7

Polarization curves showing the effect of heat-treatment temperature 51
on the catalytic activity towards ORR. Conditions are as in Fig. 2.6.

Figure 2.8

Koutecky-Levich analysis performed on the catalyst having a 54
nominal N content of 10.3 wt% at 40ºC in 1N aqueous sulfuric acid.
(a) Koutecky-Levich plot obtained at three potentials. (b) Tafel plot
for the corresponding catalyst, conditions as in Fig. 2.6.

Figure 2.9

Peroxide generation during ORR at Fe-N-C catalysts via rotating ring 55
disk (RRDE) experiments. Polarization of disk (top) and ring (bottom
left) obtained through RRDE measurements of the catalyst having a
nominal N content of 10.3 wt%. The percentage H2O2 produced is
plotted bottom right. Inset: Percentage H2O2 generated for varying
catalyst loading. All conditions are as in Fig. 2.6.

Figure 2.10

The effect of pyridinic N content on current density at 0.65 V/SHE.

56

Figure 3.1

Structure of pyridinic N precursors with varying N/C ratio.

68

Figure 3.2

Oxygen reduction current density at thin-film rotating disk electrodes 74
with MNC catalysts of varying precursor N/C ratio. (a) Pseudosteady state polarization; (b) iR- and mass-transfer corrected Tafel
curves; Conditions: O2-saturated, 1N aqueous sulfuric acid, 40°C.
-1
Scan rate 0.5 mV s , 1200 rpm, Nominal 6.3 wt% N loading.

Figure 3.3

Kinetic current density at 0.8 V/ RHE and % H2O2 at 0.5 V/RHE as a 75
function of precursor N/C. Conditions: O2-saturated, 1N aqueous
-1
sulfuric acid, 40°C. Scan rate 0.5 mV s , 1200 rpm, Nominal 6.3
wt% N loading.

Figure 3.4

(a) Effect of precursor N/C ratio on differential surface area 76
distribution obtained from BJH analysis of nitrogen desorption. (b)
Observed BET surface area and bulk N obtained through CHN
analysis.

Figure 3.5

Cycle test for Melamine based catalyst in O2-saturated, 0.1M 78
-1
aqueous sulfuric acid, 40°C. Scan rate 50 mV s , 1200 rpm, Nominal
6.3 wt% N loading.

Figure 3.6

High resolution N1s XPS spectra deconvoluted for Melamine based 79
leached catalyst.

xiii

Figure 3.7

Bulk and surface N obtained through CHN and XPS analyses as a 81
function of N/C ratio.

Figure 3.8

a) Summary of XPS results a. Surface N species of catalysts as a 83
function of precursor N/C ratio; b) Correlation of N content to kinetic
current density at 0.8V/RHE.

Figure 3.9

a) Mass transfer corrected Tafel curves observed with increasing 84
nominal N loading: b) Kinetic current density at 0.8 V/ RHE as a
function of nominal N loading. O2-saturated, 1N aqueous sulfuric
-1
acid, 40°C. Scan rate 0.5 mV s , 1200 rpm.

Figure 3.10

a) Main effects and b) interaction plot for 3x3 time-temperature 86
parametric study of pyrolysis time and temperature.

Figure 3.11

Experimental XRD spectrum obtained for Melamine catalysts 87
pyrolysed at four different temperatures.

Figure 3.12

Performance of fuel cell membrane-electrode assemblies employing 90
Fe-N-C oxygen reduction catalysts. (a & b) Polarization curves for
Fe-bipyridine and Fe-melamine based catalysts, in comparison to a
commerical Pt-catalyzed MEA (BASF Corp.); Conditions: H2–O2
°
feeds (pO2 = pH2 = 1.5 bar, 80 C, 100% RH), MNC catalyst loading
-2
1.3 mg cm Solid and dotted lines indicate forward and reverse scans
respectively];

Figure 3.13

(a & b) Durability of Fe-melamine and Fe-bipyridine based MEA 92
fuel cell at 0.5 V/RHE; Conditions: H2–O2 feeds (pO2 = pH2 = 1.5
°
-2
bar, 80 C, 100% RH), MNC catalyst loading 1.3 mg cm ; Solid and
dotted lines indicate forward and reverse scans respectively.

Figure 4.1

Schematic showing the formation of pores due to ammonia etching 105
during high-pressure pyrolysis and the creation of active reaction
sites.

Figure 4.2

(a) N2 adsorption isotherms for Ketjen black and catalysts based on 108
urea and ammonium carbamate precursors. (b) and (c) Pore size
distribution of carbon black and catalysts.

Figure 4.3

SEM images of (a) Ketjen carbon black (b) Ammonium 111
carbamate/Fe/Ketjen black catalysts.

xiv

Figure 4.4

(a) and (b) Polarization curves obtained using RDE and Mass 113
transport corrected tafel representation of polarization plots of Fe–N–
C catalysts synthesized with different ammonia-generating
precursors. (c) Kinetic current densities at 0.8 V/RHE Conditions:
-1
O2-saturated, 1N aqueous sulfuric acid, 40°C. Scan rate 0.5 mV s ,
1200 rpm, Nominal 6.3 wt% N loading.

Figure 4.5

(a) and (b) Polarization curves obtained using RRDE and amount of 114
hydrogen peroxide generated Fe–N–C catalysts synthesized with
different ammonia-generating precursors. Conditions: O2-saturated, 1
-1
N aqueous sulfuric acid, 40°C. Scan rate 0.5 mV s , 1200 rpm,
Nominal 6.3 wt% N loading.

Figure 4.6

Koutecky - Levich analysis performed on Fe–N–C catalysts 116
synthesized with ammonium carbamate precursors. Conditions: O2-1
saturated, 1N aqueous sulfuric acid, 40°C. Scan rate 5 mV s ,
Nominal 6.3 wt% N loading.

Figure 4.7

High resolution N1s spectra deconvoluted for (a) Urea and (b) 119
Ammonium carbamate based leached catalysts.

Figure 4.8

Correlation of RDE kinetic activity at 0.8V /RHE to porosity and 122
nitrogen content.

Figure 4.9

Performance of fuel cell membrane-electrode assemblies employing 123
Fe-N-C oxygen reduction catalysts. (a) and (b) Polarization curves
for Fe-urea and Fe-ammonium carbamate based catalysts;
Conditions: H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80°C, 100% RH),
-2
MNC catalyst loading 1.3 mg cm Solid and dotted lines indicate
forward and reverse scans respectively.

Figure 4.10

(a) and (b) Durability of Fe-urea based MEA fuel cell at 0.5 V/RHE; 125
Conditions: H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80°C, 100% RH),
-2
MNC catalyst loading 1.3 mg cm .

Figure 4.11

Durability of Fe-ammonium carbamate based MEA fuel cell at 0.5 126
V/RHE; Conditions: H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80°C, 100%
-2
RH), MNC catalyst loading 1.3 mg cm .

Figure 5.1

Experimental set up for in-situ XAS measurements at X-3B. 141
Courtesy: NE University.

xv

Figure 5.2

a) Mass transfer corrected Tafel plots for Melamine based Fe-N-C 144
catalysts synthesized using varying nominal Fe loadings. b) Kinetic
current as function of nominal Fe loading c) observed bulk N content
as a function of Fe. Conditions: O2-saturated, 1 N aqueous sulfuric
-1
acid, 40°C. Scan rate 0.5 mV s ,1200 rpm, Nominal 6.3 wt% N
loading.

Figure 5.3

XRD patterns of catalysts synthesized by varying nominal Fe 148
content, N precursor-Melamine.

Figure 5.4

Polarization curves: O2-saturated, 0.1 M aqueous sodium hydroxide, 151
40°C. Scan rate 0.5 mV s-1, 1200 rpm, Nominal 0.75 wt% Fe
loading.

Figure 5.5

a) XAFS data showing pre-edge line. b) Pre-edge subtracted and 154
2
Normalized spectra. c) K weighted Fourier transforms of XAFS
data. Conditions: Ar-saturated 0.1M HClO4, 0.5V vs. RHE insitu Kα
Edge XANES.

Figure 5.6

Fe Kα non-phase corrected Fourier transformed EXAFS spectra of 156
melamine –based Fe-N-C catalysts under (a) ex-situ conditions (b)
in-situ 0.1M HClO4, room temperature, Open circuit potential (0.9 V
vs. RHE).

Figure 5.7

Fe Kα non-phase corrected Fourier transformed EXAFS spectra and 159
data fits of melamine – based Fe-N-C catalysts for Fe-N coordination
of 2,3 and 4. Conditions: 0.1 M HClO4, room temperature, Open
circuit potential (0.9 V vs. RHE).

Figure 5.8

a) Insitu Fe Kα XANES spectra of Melamine based catalysts 161
measured at different potentials under Oxygen together with
standards b) First derivative analysis of the Kα XANES spectra.
Conditions: 0.1M HClO4, room temperature.

xvi

1. INTRODUCTION
1.1. Hydrogen Economy and Fuel Cells Systems
Worldwide demand for energy has been increasing at an alarming rate. The
increasing energy demand has largely been met by fossil fuel reserves that are confined to
4,5

few regions in the world,

and that emit green house gases and other pollutants to the

6

environment. Volatility with regard to fossil fuel supply and prices combined with
stringent environment policies demanding reduction in green house gases and other toxic
emissions has driven long-term efforts towards developing a hydrogen-based energy
7,8

economy.

Hydrogen, the third most abundant element on earth, is a versatile energy

carrier that can be produced from diverse primary energy sources such as coal, natural
gas, biomass and other renewable energy technologies such as solar, wind and
hydroelectric power.
A fuel cell is an energy conversion device that utilizes the power of hydrogen and is
9

a key enabling technology for a hydrogen economy. Fuel cells find applications in
several sectors ranging from portable devices such as mobile phones and laptops,
transportation applications such as cars and buses and stationary applications in domestic
10

and industrial sectors as heat and power generators.

The key benefit of fuel cell systems

is that they are zero emission devices as they directly convert the chemical energy of
hydrogen to electricity with water and potentially-useful heat as byproducts. Due to their
low noise, light weight and few moving parts, fuel cells are suitable for mobile
applications. Also, when combined with internal combustion engines, fuel cells can be

1

used as auxillary power units and as stationary back-up systems in combination with
reformers for on-board conversion of other fuels, there-by saving energy and reducing
11-13

emissions.

1.2. Comparison of Fuel Cell Technologies
A single fuel cell consists of an electrolyte sandwiched between two electrodes, the
anode and the cathode. Gaseous fuels (Hydrogen or Methanol) are typically fed on the
anode side and an oxidant (Oxygen or air) is supplied on the cathode counterpart. Bipolar
plates on either side of the cell evenly distribute the reactant gases on the electrode
surfaces. Based on the type of electrolyte used, fuel cells are primarily classified as
Proton Exchange Membrane Fuel cells (PEMFC), Alkaline Fuel Cells (AFC), Phosphoric
Acid Fuel cells (PAFC), Molten Carbonate Fuel Cells (MCFC) and Solid Oxide Fuel
cells (SOFC). The electrolyte determines the operational temperature of the fuel cell, the
11,13-15

type of reaction and their applications.

Figure 1.1 shows the mechanism through which the different types of fuel cells
operate, their operational temperature and applications. Among these, PEMFCs use a
solid polymer electrolyte (perfluro-sulfonic acid) that minimizes corrosion and eases
electrolyte management, however their low operational temperature typically necessitates
16,17

precious metal catalysts and are very sensitive to fuel impurities.

AFCs are high

performance fuel cells because oxygen reduction occurs at higher rate in alkaline
electrolyte and hence find applications in military and space. However, electrolyte
management remains a challenge because alkaline fuel cells are very sensitive to
common contaminants such as CO2 in fuel and air.
2

14,18

Fuel Cell
Technology

Anode

Electrolyte

Reaction gases

CO
H2

Molten
Carbonate
Fuel
Cells(MCFC)

Alkaline Fuel
Cells (AFC)
Proton
Exchange
Membrane
Fuel
Cells(PEMFC)

O2

Auxiliary
Power
Distributed
generation

CO2
O2

600 – 700 oC

CO23

Electric utility
Distribute
generation

O2

CO
H2

Phosphoric
Acid Fuel Cells
(PAFC)

600 – 1000 oC
O2-

CO2
H2 O

Applications
Reaction gases

LOAD
CO2
H2 O

Solid Oxide
Fuel Cells
(SOFC)

Cathode

Military
Space

150 – 200 oC
H2

H+

H2 O
H2

90 – 120 oC
H2O

Distributed
generation

H2 O

Back-up Power
Distributed
generation
Transportation

OH50 – 120 oC

H2

H+

Fuel

O2

Oxidant

Figure 1.1 Mechanisms and applications of different types of Fuel Cells.
(For interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation)

3

PAFCs are widely used for distributed generation and they utilize phosphoric acid
stabilized in SiC matrix as the electrolyte. Though their higher operational temperature
enables Combined Heat and Power (CHP) generation, costs associated with platinum
catalysts and their longer start up time have prevented their mass adoption. MCFCs and
SOFCs are high temperature fuel cells primarily used for distributed utility
14,15,19

generation.

MCFCs use a solution of lithium, sodium and potassium carbonates

soaked in a matrix of LiAlO2 and are not susceptible to CO or CO2 contamination. They
operate around 700-800°C and hence eliminate the need for expensive catalysts and are
suitable for CHP. However, higher operational temperature results in corrosion and
20

breakdown of cell components.

SOFCs typically use Yttria Stabilized Zirconia (YSZ)

as the electrolyte and are high power systems (~ 1KW-2KW). Again, their higher
21,22

operational temperature introduces similar challenges as that of MCFCs.

The power produced from a single fuel cell depends on the fuel cell size, type,
operational temperature and the pressure at which gases are supplied. In order to increase
the power, single fuel cells are combined in series to form a stack. Depending on the end
applications that include laptop computers (20-50 W), transportation vehicles (50-120
KW) and distributed generation (1-200 MW), a fuel cell stack may contain a few to
several hundred individual cells.

15

1.3. Proton Exchange Membrane Fuel Cell
The low operational temperature of PEMFCs- typically around 80°C- allows for
rapid startup and they can reach efficiencies up to 60%. Their higher power-densities,
low weight and volume make them highly attractive for transportation and smaller
4

L O A D
Reaction gases

Reaction gases

C AT H O D E

Cathode Gas Diffusion Layer

H+

Electrocatalyst

PFSA Membrane

H2O

Electrocatalyst

Anode Gas Diffusion Layer

ANODE

H+

H2O

Oxidant (O2)

Fuel (H2)

Figure 1.2 Proton Exchange Membrane Fuel Cell (PEMFC).

5

stationary applications. The schematic of a PEMFC is shown in Fig. 1.2. An individual
cell consists of a cathode, an anode and electrolyte membrane. Bipolar plates connect the
individual cells in series to form a fuel cell stack and steel or aluminum end plates
terminate the stack and provide structural integrity. Bipolar plates distribute the reactant
gases across the electrodes, provide electronic conductive paths and remove waste heat.
Adjacent to bipolar plates is a gas diffusion layer (GDL) that distributes the gases to
catalyst layer, conducts electric current and provides a pathway for liquid water removal
to avoid electrode flooding. The five layered structure that includes two GDLs and
electrocatalyst layers on the anode and the cathode side and the membrane separating
them constitute the Membrane Electrode Assembly (MEA). Hydrogen gas flows to the
anode, where the catalyst, typically platinum supported on carbon, splits it into protons
and electrons. The protons are conducted through the polymer membrane to the cathode
side and electrons flow through the external circuit. At the cathode, oxygen combines
with the protons and electrons to form water and heat. The anode, cathode and the net
overall reactions are represented below.
H 2 → 2 H + + 2e −

(1.1)

1
H + + 2e − + O2 → H 2 O
2

(1.2)

1
H 2 + O2 → H 2 O
2

(1.3)

∆G = -237 kJ/mol

Hence the equilibrium cell voltage at the standard conditions of 25°C and 1 atm pressure
can be given as

∆U 0 =

− ∆G
= 1.23V
nF

(1.4)

6

Figure 1.3 Current density and over potential plot showing the major irreversible losses
3

associated with fuel cell operation.

7

However, fuel cells do not reach the theoretical maximum of 1.23 V during practical
operation due to the irreversible voltage losses associated with the flow of current. The
difference between the actual cell voltage measured at a given current density and the
reversible cell voltage is termed as overvoltage or polarization.

13,14,23

Figure 1.3

represents the polarization curve for a hydrogen/air fuel cell. Prominent sources of
overvoltage include 1) Mixed potential losses at open circuit potential due to parasitic
reactions such as fuel crossover; 2) Activation losses at low current densities (1-100
2

mA/cm ), due to sluggish oxygen reduction kinetics at the cathode; 3) Ohmic losses at
2

moderate current densities (100-500 mA/cm ) due to resistive losses in the electrolyte
and electrodes and 4) Mass transport losses arising due to non-reacting diffusion in the
gas diffusion layer and reacting diffusion in the electrodes at high current densities (>500
2 23,24

mA/cm ).

1.4. Electrocatalytic Oxygen Reduction Reaction
Kinetic and mass transport losses at the cathode, where the oxygen reduction reaction
takes place, dominate the major overvoltage losses in PEMFC. Nano-structured platinum
dispersed on a conductive support, either as pure Pt black or as an alloy such as Pt3Co,
25,26

Pt3Ni, is the only viable catalyst currently used in large-scale PEMFCs.

The oxygen

reduction reaction is a complex electrochemical process that involves several
intermediates, depending on the nature of the electrode material, catalyst and electrolyte.
Two possible mechanisms have been proposed following the adsorption of oxygen
molecule onto the catalyst surface. If the O2 molecule adsorbs onto the catalyst surface in
such a way that its axis is parallel to the catalyst surface, it divides into adsorbed oxygen
8

atoms, which later reduces and protonates to form water. If the oxygen molecule adsorbs
onto the catalyst with its axis perpendicular to the catalyst surface, the molecule does not
-

split. Instead the peroxide anion HO2 is formed by partial reduction as an intermediate,
which may eventually get reduced to water.

27,28

The possible reactions in acidic aqueous

solution can be summarized as follows:
o

(1.5)

o

(1.6)

o

(1.7)

O2 + 4 H + + 4e − ⇔ 2 H 2 O

E = 1.23V vs NHE

O 2 + 2 H + + 2e − ⇔ H 2 O 2

E = 0.68V vs NHE

H 2 O2 + 2 H + + 2e − ⇔ 2 H 2 O

E = 1.76 V vs NHE

Though platinum is the only metal that can withstand the acidic PEMFC
environment, the high cost and low abundance of this precious metal has limited largescale commercialization of PEMFCs. For an 80 kW fuel stack, Pt catalysts costs account
for ~ 56% of the total cost, which amounts to $130/kW. However, the U. S. Department
of Energy (DOE) has set a cost target of $30/kW and a platinum utilization target of 0.2 g
of Pt per kW for fuel cells to be commercially viable.

14

In order to reduce the catalyst

layer cost and to achieve the DOE target, the following approaches have been taken: a)
2

Reducing Pt loading to less than 0.2 g/peak kW or 0.05 mg/cm to bring down the
29,30

costs.

b) Maximizing Pt mass activity through alloying with transition metals such
26,31,32

Ni, Cu, Co, Fe or core-shell nanostructuring

and c) Developing non-precious

transition metal based catalysts (NPMCs) for oxygen reduction, which completely
33

eliminates usage of precious metals.

9

1.5. Non-precious Metal-N4 Macrocycles- based catalysts
In 1964, Jasinski reported catalysis of oxygen reduction by cobalt pthalocyanine , (a
macrocycle molecule where the metal is symmetrically surrounded by four nitrogen
34

atoms) in alkaline media.

Subsequently, several other non-noble materials have been

proposed to be active for ORR. Transition metal oxides particularly with pervoskite or
35

pyrochlore structure have been found to be active for the ORR in alkaline media.

36

However Reeve et al. observed that these materials were poorly stable in acidic media.

Transition metal macrocyclic compounds, especially transition metal N4 chelates like
37-40

porphyrins,

pthalocyanines, tetraazannulenes

41-49

were found to be active for ORR

when applied on suitable carbon support materials, however the attack of macrocyclic
49

rings with peroxide intermediates leads to poor stability in acidic media.

Catalytic activity strongly depends on the central metal ion; for N4 chelates, the
50

activity varies as Fe>Co>Ni>Cu>Mn.

Filled dxz orbitals and dyz orbitals and empty dz

orbitals favor partial electron transition from the metal to anti-bonding π orbitals of
oxygen facilitating oxygen reduction in such N4 chelates.

51

52

Beck et al.

elucidated the

role of the central metal ion during oxygen reduction, which includes the reduction of the
central ion as a first step, followed by formation of an chelate-oxygen adsorption
complex and the transfer of first electron. The reaction products are then formed in
53

subsequent reaction steps. However, Durand et al.

claim that the support determines the

oxygen reduction activity through its direct co-ordination to the metal ion, if the complex
is adsorbed parallel to the surface.
10

Heat treatment at temperatures close to 800°C led to improved activity and stability
37,38,40,46,49

of these catalysts.

40

Van Veen et al.

concluded that heat treatment of these

supported N4 chelates leads to formation of isolated metal particles co-ordinated by
thermally modified ligands. Though the macrocyclic structure is destroyed, the metalnitrogen co-ordination remains intact. The amount of metal ion not co-ordinated with
nitrogen form oxides readily in acidic electrolytes and are electrochemically inactive.
56

Van der Putten et al.

54,55

concluded that heat treatment modifies the chemical nature of the

carbon support and increases active surface area through a better distribution of the
chelate on the surface. However, despite being considered non-precious, these metal
57

macrocycles are expensive precursors.

1.6. Metal-Nitrogen-Carbon (MNC) Catalysts
Yeager et al. demonstrated that only a metal-nitrogen center is required, not the
entire macrocycle, by producing an active catalyst for ORR by pyrolysis of a metal
precursor (Cobalt acetate), carbon black and a nitrogen precursor (poly-acrylonitrile) in
58

an inert atmosphere.

Following this approach, many such MNC catalysts have been
57

synthesized using less expensive inorganic metal salts such as Iron acetate,
59

sulfate,
wide

various carbon supports such as Vulcan carbon,

variety

of

nitrogen

precursors

such

as

60

Ketjen carbon black
ammonia,

57

Cobalt

61

and a
62

cyanamide,

63,64

polyacrylonitrile.

The oxygen reduction activity of MNC catalysts is directly related to the number of
catalytically active sites and structure, which in turn depends on the metal precursor, the
11

nitrogen precursor structure, synthesis conditions and heat treatment process.

59,65-67

Amongst the transition metals, Fe- and Co- containing catalysts were more active than
other transition metals similar to N4 chelates in acidic electrolytes.

59

The metal loading in

the catalyst also plays a role in high catalytic activity. Excess metal in the catalyst
transforms to clusters containing metals and metal carbides that are catalytically inactive.
Using FeTMPPCl, Lefevre et al. demonstrated that only 0.2 wt % nominal metal loading
was required to maximize oxygen reduction activity; beyond which inactive graphitized
iron metallic particles were detected.

1,68

Nitrogen constitutes a major component of the catalytic site and the type of N source
and the amount of surface N significantly affects catalytic activity. The structure of the
metal-nitrogen carbon (MNC) catalytic sites and the mechanism of ORR at these sites
still remain uncertain. However, based on time-of-flight secondary ion mass spectrometry
results, Dodelet et al. suggested FeN2/C and FeN4/C or CoN4/C as the catalytic active
1,2

sites(Fig. 1.4).

Using ammonia gasification of defect sites in metal-loaded pristine

carbon, they introduced N-enriched micropores that acted as active centers for ORR.
69,70

Ozkan et al.

71

and Stevenson et al.

identified nitrogen-containing edge planes

without metal as active sites. Popov et al. proposed that active sites containing pyridinic
nitrogen are responsible for the catalytic activity for ORR and reported low levels of
72

H2O2 evolution when reducing oxygen in acidic medium

Ruggeri et al. proposed that
73

surface pyridinic nitrogen in co-ordination with iron forms the catalytically active site.

Following this, Charreteur et al. proposed a possible active site, which contains pyridinic
N in a phenanthroline environment present at graphitic crystallite edges. N present in two
12

1,2

Figure 1.4 Proposed catalytic active site structure.

13

adjacent crystallites then coordinate with Fe to form an active site of the form
74

FeN2+2/C.

The presence of MN4 active sites in catalysts pyrolyzed at temperatures above 800°C
75

was confirmed by Ziegelbauer et al. using ∆µ XANES analysis

which uses a

difference method to isolate changes in XANES due to adsorbates on metal surface and
–

established such sites were proposed to achieve complete, 4 e reduction of oxygen.
76

Recently, Wood et al.

produced highly active catalysts that showed a volumetric
-3

current density of 19 A cm using metal-doped nitroaniline as the nitrogen source. Their
-2

catalysts contained nearly the same amount of iron (6 µg cm ) as Ruggeri et al. but with
two-fold higher surface nitrogen content (4%). This was considered as further evidence
73

that iron may not be a part of the catalytic active site, however N is more significant.

Similarly, Wei et al. reported an increase in catalytic activity of pyrolyzed Co-N/C with
77

increasing surface concentration of nitrogen.

The surface properties of carbon support such as surface area, porosity, electric
conductivity and ordering significantly influence the degree of catalyst dispersion and
78

hence the catalytic activity. Ehrburger et al.

demonstrated that heterogenous carbon

surfaces, consisting basal and edge crystalline planes have higher surface area than
79

homogenous supports and favored higher N retention. Jaouen et al.
2

pyrolyzed a highly-

-1

ordered synthetic graphite powder of low surface area (3.5 m g ) along with an iron
precursor (e.g. iron acetate) in an ammonia environment. As the graphite crystallite size
decreased, the degree of disorder and microporous (< 22Å) specific surface area
14

80

increased, as did the nitrogen content and catalytic activity. Subramanian et al.

oxidized

their carbon support with HNO3 and oxygen rich groups such as quinone on the carbon
surface favored the adsorption of amines and increased the polarity of carbon support.
43

Gouerec et al.

also believed that presence of surface oxygen groups led to slower aging

of the catalyst.
Another important factor determining catalytic activity concerns the synthesis
method involved. Conventional synthesis method for MNC catalysts involves mixing
metal, carbon and nitrogen through wet impregnation or ball-milling and pyrolyzing the
resulting powder under inert flowing gases at atmospheric pressures at temperatures
73-79

between 700-1000°C.

Alternatively, other deposition techniques such as chemical

vapor deposition, vacuum sublimation and sol-gel method were explored in an attempt to
increase the catalytic active sites and dispersion. Atkinson et al.

81

illustrated an ultrasonic

spray pyrolysis technique that allowed for a one-step continuous production of high
surface area, porous carbon spheres impregnated with well-dispersed Fe nanoparticles.
82

Hermann et al.

showed that plasma treatment of their catalysts suppressed particle

agglomeration and formed smaller average particle sizes compared to conventional
pyrolysis methods. Bogadanoff et al.

83

improved their catalytic efficiency by introducing

foaming agents such as iron oxalate to produce carbon insitu during pyrolysis. Lefevre et
84

al.

introduced pore fillers such as perylene-tetracarboxylic dianhydride and 1,10-

phenanthroline and ballmilled them with carbon-support, metal and nitrogen source and
believed that slit pores in carbon support filled up with pore fillers increased site density.

15

85

Wu et al.

introduced polyaniline as a C-N template during their catalysts synthesis step

and showed better distribution of N sites.
1.7. Activity and Durability Evaluation for NPMC
33

Gasteiger et al.

established a criterion for comparing the performances of NPMCs

through two important parameters, the turn-over frequency (TOF) and volumetric activity
(IV). TOF corresponds to the number of electrons transferred per active site per second, at
a given potential and IV is the product of the site density and average single site activity
and is given as

I V = SD × TOF × e −

(1.8)
3

3

Where IV is the volumetric activity (A/cm ), SD is the site density (sites/cm ), TOF is
-

the average turn-over frequency (electrons per site per s ) and e is the charge of a single
electron (1.602x10

-19

86

C).

Though these MNC catalysts are potentially low-cost, enabling high catalyst usage,
the volume and thickness of an MNC cathode cannot be increased excessively without
hindering oxygen transport.

78,87,88

The U.S. Department of Energy (DOE) has established

a figure of merit in terms of volumetric ORR activity. The 2015 target for non-precious
-3

oxygen reduction catalysts has been set at 300 A cm at 0.8 V, as measured in a fuel cell
at 80°C with absolute feed pressures of 1 bar and corrected for extraneous (e.g. ohmic)
cell resistances and over 3000 h of stable intermittent activity is desired. In other words,
the same current density as that of a ~10 µm thick platinum cathode is desired without
33

exceeding an electrode layer thickness of ~100 µm.
16

Activities of NPMC are measured using the Rotating Disc electrode (RDE) close to
room temperature and single fuel cell measurements at 80°C. Mass (A/g) or volumetric
3

(A/cm ) activity obtained at 0.8 V when the electrode is completely under kinetic control
is reported. Mass transfer corrected kinetic current densities are generally obtained from
Tafel plots of oxygen reduction. Corrections for higher gas pressures to 1 bar O2 and H2
pressures used during fuel cell measurements are done using the following equation.

αc
* 0.79  P*  2

O
* =I P 2 
 H2 
IM
M
 PH 
PO 
2
2







(1.9)



where IM* is the corrected mass activity (A/g) at 0.8 V/ NHE, IM is the measured mass
activity (A/g) at 0.8 V/ NHE, PO2 and PH2 are the experimental partial pressures (bar),
PO2* and PH2* are the reference partial pressures (1 bar, 100% RH), αc is the
experimental cathodic exchange transfer co-efficient ~1.
The volumetric activity is estimated from mass activity through
*
*
I v = ρ eff I M

(1.10)

where ρeff is the effective density of NPMC catalysts and is generally assumed to be
3 86

close to that for carbon blacks (~ 0.4 g/cm ).

Table 1.1 gives an overview of the state-of

art MNC catalysts, their synthesis procedure, nitrogen precursors, RDE mass activity,
fuel cell mass and volumetric activities together with the DOE’s target for 2010 and
2015.
Though reasonable activity has been demonstrated with MNC catalysts, the poor
durability of these catalysts in acidic environment prevents large-scale commercialization
17

of these catalysts for practical applications. H2O2 produced during incomplete oxygen
reduction is cited to be the primary cause of catalytic site degradation. H2O2 can directly
-

oxidize active sites or react with Fe to generate more oxidizing HO radicals and attack
89

the active site. Widelov et al.

observed a decrease in mass of their catalyst and through

XPS, a decrease in N1s signal intensity after a fuel cell durability test. Schulenburg et al.
90

also explains that oxidation of N atoms by H2O2 or their self-decomposition in acidic
85

environment leads to the decrease in active sites. Recently, Wu et al.

have

demonstrated over 700 h of stable performance with their PANI-Fe3Co-C catalyst. They
believe that the catalytic active sites are embedded within graphene planes of carbon
support and higher degree of graphitization leads to better durability by preventing active
site flooding. Besides, they believe that the presence of Co oxides in the catalyst
increased the number of peroxide decomposition sites and further improved durability.
The choice of carbon support may influence the durability of the catalysts. The same
group have demonstrated that among Vulcan XC-72, Ketjen black 300J, Black pearl 2000
and Multi walled Carbon Nano-tube (MWNT) catalysts, MWNT – based catalysts
showed almost negligent degradation for over 500, due to higher degree of graphitization
of MWNTs.

91

Further spectroscopic studies to analyze the state of Fe-N sites during fuel cell
operation and eliminating catalyst layer flooding will help to improve the durability of
MNC catalysts.

18

Table 1.1 Synthesis and performance of MNC catalysts.

Catalyst

Synthesis
Method

Pyrolysis
Conditions
(° time (h),
C,
gas comp.)

RDE
Mass
activity
(A/g)

PEMFC
Mass
activity
(A/g)

PEMFC
Volumetric
activity
3

(A/cm )

Reference,
Year
Published

FeSO4·7H2O +
PAN + C P33

Wet
Impregnation
(12 % Fe +
17.5 % N

850, 5, Ar

0.3

-

-

59, 1992

COPPY + Vulcan
xC72

Wet
impregnation
(10 % Co)

Unpyrolyzed

-

303

1.3

92, 2006

FeAC + nonmicroporous
carbon

Wet
impregnation
(0.2 % Fe)

900, 3, NH3

5.0

3.6

1.4

73, 2007

FeCl2 + SiO2 +
Pyrole

Wet
impregnation
(0.5 % Fe)

a) 900, 3, Ar
b) 900, 3,

1.2

14.0

5.6

93, 2009

800-1000, 25

FeCl2 +
Nitroaniline + C

Wet
impregnation
(Fe:N-1:4) +
Ball milling

-

7.3

2.9

76, 2004

FeAC +
Phenanthroline +
C

Ball milling

-

248.0

99

84, 2004

Cyanamide +
FeSO4 + C

Wet
impregnation
(4.9 % Fe)

a) 800, N2

-

124.0

50

86, 2011

Wet
impregnation

a) 900, N2
b) 900, N

17.0

250.0

~100

85, 2011

Target for 2010

-

-

-

325.0

130

14, 2007

Target for 2015

-

-

-

750.0

300

14, 2007

Polyaminiline +
C+ EDA +
+

Co(II) Nitrate +
Fe (III) Chloride

NH3

% NH3 & 75
% N2
a) 900, Ar
b) 900, NH3

b) 800, N2

19

This thesis concerns the design and engineering of a novel synthetic process for MNC
catalysts that increases the accessible population of catalytically vital pyridinic nitrogen
species on the carbon support. Our approach for synthesizing MNC catalysts includes
pyrolysing a mixture of pyridinic nitrogen rich precursors, transition metal Iron and high
surface area carbon support, Ketjen black in a closed constant volume reactor. The
individual steps toward realizing this goal include:
1) Demonstration of catalyst synthesis by high- pressure pyrolysis in a closed constant
volume vessel (quartz tube), wherein the partial pressure of the nitrogen precursor
29

increases during pyrolysis to maintain high nitrogen activity.

2) Evaluation of N precursors with high nitrogen-to carbon (N/C) ratio to increase the
accessible active site density by reducing carbon deposition in the pores of carbon
30

support.

3) Evaluation of ammonia based, carbon-free N precursors to increase micro and
mesopores during pyrolysis to host more active sites and to improve oxygen transport
respectively.
4) Evaluation of the role of transition metal and gaining an insight into Metal-Nitrogen
active reaction sites for oxygen reduction.
The outcome of this research will find direct application in automotive industry,
where there is a huge drive for clean, efficient, sustainable energy technologies and
where the cost of the platinum based catalysts is a major barrier for fuel cell
commercialization.

20

REFERENCES

21

1.8. References
1.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "Molecular oxygen reduction in PEM
fuel cells: Evidence for the simultaneous presence of two active sites in Fe-based
catalysts", Journal of Physical Chemistry B 106, 8705 (2002).

2.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "Molecular oxygen reduction in PEM
fuel cell conditions: ToF-SIMS analysis of Co-based electrocatalysts", Journal of
Physical Chemistry B 109, 16718 (2005).

3.

L. Carrette, K. A. Friedrich, and U. Stimming, "Fuel Cells - Fundamentals and
Applications", Fuel Cells 1, 5 (2001).

4.

M. Asif and T. Muneer, "Energy supply, its demand and security issues for
developed and emerging economies", Renewable & Sustainable Energy Reviews
11, 1388 (2007).

5.

http://www.wikinvest.com/concept/Rising_worldwide_demand_for_energy.

6.

T. R. Karl and K. E. Trenberth, "Modern global climate change", Science 302, 1719
(2003).

7.

L. Barreto, A. Makihira, and K. Riahi, "The hydrogen economy in the 21st century:
a sustainable development scenario", International Journal of Hydrogen Energy 28,
267 (2003).

8.

M. Conte, A. Iacobazzi, M. Ronchetti, and R. Vellone, "Hydrogen economy for a
sustainable development: state-of-the-art and technological perspectives", Journal
of Power Sources 100, 171 (2001).

9.

B. E. Logan, "Extracting hydrogen electricity from renewable resources",
Environmental Science & Technology 38, 160A (2004).

10.

R. Shinnar, "The hydrogen economy, fuel cells, and electric cars", Technology in
Society 25, 455 (2003).

11.

EG & G Services (Firm), Ralph M. Parsons Company., Science Applications
International Corporation., and United States. Office of Fossil Energy., Fuel cell
handbook, 5th ed. (U.S. Dept. of Energy Office of Fossil Energy National Energy
Technology Laboratory, Morgantown, WV, 2000).

12.

K. V. Kordesch and G. R. Simader, in Fuel Cells and Their Applications (WileyVCH Verlag GmbH & Co. KGaA, 2006), p. I.

13.

J. Larminie and A. Dicks, Fuel cell systems explained (Wiley, Chichester England ;
New York, 2000).
22

14.

http://www1.eere.energy.gov/hydrogenandfuelcells/mypp/pdfs/fuel_cells.pdf.

15.

B. C. H. Steele and A. Heinzel, "Materials for fuel-cell technologies", Nature 414,
345 (2001).

16.

T. R. Ralph, "Proton exchange membrane fuel cells: Progress in cost reduction of
the key components", Platinum Metals Review 41, 102 (1997).

17.

E. A. Ticianelli, C. R. Derouin, A. Redondo, and S. Srinivasan, "Methods to
advance technology of proton exchange membrane fuel cells", Journal of the
Electrochemical Society 135, 2209 (1988).

18.

G. F. McLean, T. Niet, S. Prince-Richard, and N. Djilali, "An assessment of
alkaline fuel cell technology", International Journal of Hydrogen Energy 27, 507
(2002).

19.

M. Watanabe, K. Tsurumi, T. Mizukami, T. Nakamura, and P. Stonehart, "Activity
and Stability of Ordered and Disordered Co-Pt Alloys for Phosphoric Acid Fuel
Cells", Journal of The Electrochemical Society 141, 2659 (1994).

20.

J. P. P. Huijsmans, G. J. Kraaij, R. C. Makkus, G. Rietveld, E. F. Sitters, and H. T.
J. Reijers, "An analysis of endurance issues for MCFC", Journal of Power Sources
86, 117 (2000).

21.

E. F. Sverdrup, C. J. Warde, and R. L. Eback, "Design of high-temperature solidelectrolyte fuel-cell batteries for maximum power output per unit volume", Energy
Conversion 13, 129 (1973).

22.

B. Zhu, "Advantages of intermediate temperature solid oxide fuel cells for
tractionary applications", Journal of Power Sources 93, 82 (2001).

23.

H. A. Gasteiger, Handbook of Fuel Cell Technology and Applications (Wiley,
Chichester, UK, 2003).

24.

V. Ramani, "Fuel cells", Electrochemical Society Interface 15, 41 (2006).

25.

N. M. Marcovic, " Fuel Cells and Fundamentals Systems 1, 105 (2001).

26.

V. Stamenkovic, T. J. Schmidt, P. N. Ross, and N. M. Markovic, "Surface
composition effects in electrocatalysis: Kinetics of oxygen reduction on welldefined Pt3Ni and Pt3Co alloy surfaces", Journal of Physical Chemistry B 106,
11970 (2002).

23

27.

Damjanov.A, M. A. Genshaw, and J. O. Bockris, "Mechanism of oxygen reduction
at platinum in alkaline solutions with special reference to H2O2", Journal of the
Electrochemical Society 114, 1107 (1967).

28.

D. B. Sepa, M. V. Vojnovic, and A. Damjanovic, "Reaction intermediates as a
controlling factor in the kinetics and mechanism of oxygen reduction at platinumelectrodes", Electrochimica Acta 26, 781 (1981).

29.

R. Giorgi, L. Giorgi, S. Gagliardi, E. Salernitano, M. Alvisi, T. Dikonimos, N. Lisi,
D. Valerini, M. F. De Riccardis, and E. Serra, "Nanomaterials-based PEM
electrodes by combining chemical and physical depositions", Journal of Fuel Cell
Science and Technology 8, 041004 (2011).

30.

B. Liu, S. Liao, and Z. Liang, "Core-shell structure: The best way to achieve low-pt
fuel cell electrocatalysts", Progress in Chemistry 23, 852 (2011).

31.

N. M. Markovic, C. A. Lucas, A. Rodes, V. Stamenkovic, and P. N. Ross, " Surf.
Sci. 499, L149 (2002).

32.

S. Mukerjee, S. Srinivasan, M. P. Soriaga, and J. McBreen, "Role of Structural and
Electronic-Properties of Pt and Pt Alloys on Electrocatalysis of Oxygen Reduction an in-Situ Xanes and Exafs Investigation", Journal of the Electrochemical Society
142, 1409 (1995).

33.

H. A. Gasteiger, S. S. Kocha, B. Sompalli, and F. T. Wagner, "Activity benchmarks
and requirements for Pt, Pt-alloy, and non-Pt oxygen reduction catalysts for
PEMFCs", Applied Catalysis B-Environmental 56, 9 (2005).

34.

R. Jasinski, "New Fuel Cell Cathode Catalyst", Nature 201, 1212 (1964).

35.

A. M. Kannan and A. K. Shukla, "Rechargeable iron/air cells employing
bifunctional oxygen electrodes of oxide pyrochlores", Journal of Power Sources 35,
113 (1991).

36.

R. W. Reeve, P. A. Christensen, A. J. Dickinson, A. Hamnett, and K. Scott,
"Methanol-tolerant oxygen reduction catalysts based on transition metal sulfides
and their application to the study of methanol permeation", Electrochimica Acta 45,
4237 (2000).

37.

V. S. Bagotzky, M. R. Tarasevich, K. A. Radyushkina, O. A. Levina, and S. I.
Andrusyova, "Electrocatalysis of the oxygen reduction process on metal chelates in
acid electrolyte", Journal of Power Sources 2, 233 (1977).

38.

V. S. Bagotzky, M. R. Tarasevich, K. A. Radyushkina, O. A. Levina, and S. I.
Andrusyova, "Electrocatalysis of oxygen reduction process on metal-chelates in
acid electrolyte", Journal of Power Sources 2, 233 (1978).
24

39.

A. Biloul, P. Gouerec, M. Savy, G. Scarbeck, S. Besse, and J. Riga, "Oxygen
electrocatalysis under fuel cell conditions: Behaviour of cobalt porphyrins and
tetraazaannulene analogues", Journal of Applied Electrochemistry 26, 1139 (1996).

40.

J. A. R. van Veen and H. A. Colijn, "Oxygen reduction on transition-metal
porphyrins
in
acid
electrolyte
II.
Stability",
Berichte
der
Bunsengesellschaft/Physical Chemistry Chemical Physics 85, 700 (1981).

41.

A. L. Bouwkamp-Wijnoltz, W. Visscher, and J. A. R. van Veen, "The selectivity of
oxygen reduction by pyrolysed iron porphyrin supported on carbon",
Electrochimica Acta 43, 3141 (1998).

42.

S. L. Gojkovic, S. Gupta, and R. F. Savinell, "Heat-treated iron(III)
tetramethoxyphenyl porphyrin chloride supported on high-area carbon as an
electrocatalyst for oxygen reduction - Part II. Kinetics of oxygen reduction",
Journal of Electroanalytical Chemistry 462, 63 (1999).

43.

P. Gouerec and M. Savy, "Oxygen reduction electrocatalysis: ageing of pyrolyzed
cobalt macrocycles dispersed on an active carbon", Electrochimica Acta 44, 2653
(1999).

44.

R. Z. Jiang and D. Chu, "Remarkably active catalysts for the electroreduction of O2 to H2O for use in an acidic electrolyte containing concentrated methanol",
Journal of the Electrochemical Society 147, 4605 (2000).

45.

H. Kalvelage, A. Mecklenburg, U. Kunz, and U. Hoffmann, "Electrochemical
reduction of oxygen at pyrolyzed iron and cobalt N4-chelates on carbon black
supports", Chemical Engineering & Technology 23, 803 (2000).

46.

M. Ladouceur, G. Lalande, D. Guay, J. P. Dodelet, L. Dignard-Bailey, M. L.
Trudeau, and R. Schulz, "Pyrolyzed Cobalt Phthalocyanine as Electrocatalyst for
Oxygen Reduction", Journal of The Electrochemical Society 140, 1974 (1993).

47.

G. Lalande, G. Tamizhmani, R. Cote, L. Dignardbailey, M. L. Trudeau, R. Schulz,
D. Guay, and J. P. Dodelet, "Influence of Loading on the Activity and Stability of
Heat-Treated Carbon-Supported Cobalt Phthalocyanine Electrocatalysts in Solid
Polymer Electrolyte Fuel-Cells", Journal of the Electrochemical Society 142, 1162
(1995).

48.

G. Q. Sun, J. T. Wang, and R. F. Savinell, "Iron(III) tetramethoxyphenylporphyrin
(FeTMPP) as methanol tolerant electrocatalyst for oxygen reduction in direct
methanol fuel cells", Journal of Applied Electrochemistry 28, 1087 (1998).

25

49.

K. Wiesener, D. Ohms, V. Neumann, and R. Franke, "N-4 Macrocycles as
Electrocatalysts for the Cathodic Reduction of Oxygen", Materials Chemistry and
Physics 22, 457 (1989).

50.

H. Alt, H. Binder, and Sandsted.G, "Mechanism of electrocatalytic reduction of
oxygen on metal-chelates", Journal of Catalysis 28, 8 (1973).

51.

M. Savy, P. Andro, C. Bernard, and G. Magner, "Study on oxygen reduction on
monomeric and polymeric phthalocyanines.1. Basic principles, selection of central
ion", Electrochimica Acta 18, 191 (1973).

52.

F. Beck, "Voltametric investigations of electrocatalytically acting phthalocyanines
and tetraazaannulenes in concentrated sulfuric-acid", Berichte Der BunsenGesellschaft-Physical Chemistry Chemical Physics 77, 353 (1973).

53.

R. R. Durand and F. C. Anson, "Catalysis of dioxygen reduction at graphiteelectrodes by an adsorbed Cobalt (II) Porphyrin", Journal of Electroanalytical
Chemistry 134, 273 (1982).

54.

R. W. Joyner, J. A. R. Vanveen, and W. M. H. Sachtler, "Extended X-ray
absorption fine-structure (EXAFS) study of cobalt-porphyrin catalysts supported on
active-carbon", Journal of the Chemical Society-Faraday Transactions I 78, 1021
(1982).

55.

B. Van Wingerden, J. A. R. Van Veen, and C. T. J. Mensch, "An extended X-ray
absorption fine structure study of heat-treated cobalt porphyrin catalysts supported
on active carbon", Journal of the Chemical Society, Faraday Transactions 1:
Physical Chemistry in Condensed Phases 84, 65 (1988).

56.

A. Van Der Putten, A. Elzing, W. Visscher, and E. Barendrecht, "Oxygen reduction
on pyrolysed carbon-supported transition metal chelates", Journal of
Electroanalytical Chemistry 205, 233 (1986).

57.

C. Medard, M. Lefevre, J. P. Dodelet, F. Jaouen, and G. Lindbergh, "Oxygen
reduction by Fe-based catalysts in PEM fuel cell conditions: Activity and selectivity
of the catalysts obtained with two Fe precursors and various carbon supports",
Electrochimica Acta 51, 3202 (2006).

58.

I. T. Bae and D. A. Scherson, "In situ X-ray absorption of a carbon monoxide-iron
porphyrin adduct adsorbed on high-area carbon in an aqueous electrolyte", Journal
of Physical Chemistry B 102, 2519 (1998).

59.

D. Ohms, S. Herzog, R. Franke, V. Neumann, K. Wiesener, S. Gamburcev, A.
Kaisheva, and I. Iliev, "Influence of metal-ions on the electrocatalytic oxygen
reduction of carbon materials prepared from pyrolyzed polyacrylonitrile", Journal
of Power Sources 38, 327 (1992).
26

60.

U. A. Paulus, T. J. Schmidt, H. A. Gasteiger, and R. J. Behm, "Oxygen reduction on
a high-surface area Pt/Vulcan carbon catalyst: a thin-film rotating ring-disk
electrode study", Journal of Electroanalytical Chemistry 495, 134 (2001).

61.

V. Nallathambi, J. W. Lee, S. P. Kumaraguru, G. Wu, and B. N. Popov,
"Development of high performance carbon composite catalyst for oxygen reduction
reaction in PEM Proton Exchange Membrane fuel cells", Journal of Power Sources
183, 34 (2008).

62.

H. T. Chung, C. M. Johnston, and P. Zelenay, in Synthesis and evaluation of heattreated, cyanamide-derived non-precious catalysts for oxygen reduction, 2009, p.
485.

63.

M. C. M. Alves, J. P. Dodelet, D. Guay, M. Ladouceur, and G. Tourillon, "Origin of
the electrocatalytic properties for O2 reduction of some heat-treated
polyacrylonitrile and phthalocyanine cobalt compounds adsorbed on carbon-black
as probed by electrochemistry and x-ray absorption-spectroscopy", Journal of
Physical Chemistry 96, 10898 (1992).

64.

S. Gupta, D. Tryk, I. Bae, W. Aldred, and E. Yeager, "Heat-Treated
Polyacrylonitrile-Based Catalysts for Oxygen Electroreduction", Journal of Applied
Electrochemistry 19, 19 (1989).

65.

F. Jaouen, F. Charreteur, and J. P. Dodelet, "Fe-based catalysts for oxygen
reduction in PEMFCs - Importance of the disordered phase of the carbon support",
Journal of the Electrochemical Society 153, A689 (2006).

66.

T. Okada, M. Gokita, M. Yuasa, and I. Sekine, "Oxygen reduction characteristics of
heat-treated catalysts based on cobalt-porphyrin ion complexes", Journal of the
Electrochemical Society 145, 815 (1998).

67.

B. Wang, "Recent development of non-platinum catalysts for oxygen reduction
reaction", Journal of Power Sources 152, 1 (2005).

68.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "O-2 reduction in PEM fuel cells:
Activity and active site structural information for catalysts obtained by the pyrolysis
at high temperature of Fe precursors", Journal of Physical Chemistry B 104, 11238
(2000).

69.

P. H. Matter, E. Wang, and U. S. Ozkan, "Preparation of nanostructured nitrogencontaining carbon catalysts for the oxygen reduction reaction from SiO2- and MgOsupported metal particles", Journal of Catalysis 243, 395 (2006).

27

70.

P. H. Matter, L. Zhang, and U. S. Ozkan, "The role of nanostructure in nitrogencontaining carbon catalysts for the oxygen reduction reaction", Journal of Catalysis
239, 83 (2006).

71.

S. Maldonado and K. J. Stevenson, "Influence of Nitrogen Doping on Oxygen
Reduction Electrocatalysis at Carbon Nanofiber Electrodes", J. Phys. Chem. B 109,
4707 (2005).

72.

Y. Nabae, S. Moriya, K. Matsubayashi, S. M. Lyth, M. Malon, L. B. Wu, N. M.
Islam, Y. Koshigoe, S. Kuroki, M. A. Kakimoto, S. Miyata, and J. Ozaki, "The role
of Fe species in the pyrolysis of Fe phthalocyanine and phenolic resin for
preparation of carbon-based cathode catalysts", Carbon 48, 2613 (2010).

73.

S. Ruggeri and J. P. Dodelet, "Influence of structural properties of pristine carbon
blacks on activity of Fe/N/C cathode catalysts for PEFCs", Journal of the
Electrochemical Society 154, B761 (2007).

74.

F. Charreteur, F. Jaouen, S. Ruggeri, and J. P. Dodelet, "Fe/N/C non-precious
catalysts for PEM fuel cells: Influence of the structural parameters of pristine
commercial carbon blacks on their activity for oxygen reduction", Electrochimica
Acta 53, 2925 (2008).

75.

J. M. Ziegelbauer, T. S. Olson, S. Pylypenko, F. Alamgir, C. Jaye, P. Atanassov,
and S. Mukerjee, "Direct spectroscopic observation of the structural origin of
peroxide generation from co-based pyrolyzed porphyrins for ORR applications",
Journal of Physical Chemistry C 112, 8839 (2008).

76.

T. E. Wood, Z. Tan, A. K. Schmoeckel, D. O'Neill, and R. Atanasoski, "Nonprecious metal oxygen reduction catalyst for PEM fuel cells based on nitroaniline
precursor", Journal of Power Sources 178, 510 (2008).

77.

G. Wei, J. S. Wainright, and R. F. Savinell, "Catalytic activity for oxygen reduction
reaction of catalysts consisting of carbon, nitrogen and cobalt", Journal of New
Materials for Electrochemical Systems 3, 121 (2000).

78.

P. Ehrburger, A. Mongilardi, and J. Lahaye, "Dispersion of iron phthalocyanine on
carbon surfaces", Journal of Colloid and Interface Science 91, 151 (1983).

79.

F. Jaouen, S. Marcotte, J. P. Dodelet, and G. Lindbergh, "Oxygen reduction
catalysts for polymer electrolyte fuel cells from the pyrolysis of iron acetate
adsorbed on various carbon supports", Journal of Physical Chemistry B 107, 1376
(2003).

80.

N. P. Subramanian, S. P. Kumaraguru, H. Colon-Mercado, H. Kim, B. N. Popov, T.
Black, and D. A. Chen, "Studies on Co-based catalysts supported on modified
carbon substrates for PEMFC cathodes", Journal of Power Sources 157, 56 (2006).
28

81.

J. D. Atkinson, M. E. Fortunato, S. A. Dastgheib, M. Rostam-Abadi, M. J. Rood,
and K. S. Suslick, "Synthesis and characterization of iron-impregnated porous
carbon spheres prepared by ultrasonic spray pyrolysis", Carbon 49, 587 (2011).

82.

I. Herrmann, V. Bruser, S. Fiechter, H. Kersten, and P. Bogdanoff, "Electrocatalysts
for oxygen reduction prepared by plasma treatment of carbon-supported cobalt
tetramethoxyphenylporphyrin", Journal of the Electrochemical Society 152, A2179
(2005).

83.

P. Bogdanoff, I. Herrmann, M. Hilgendorff, I. Dorbandt, S. Fiechter, and H.
Tributsch, "Probing structural effects of pyrolysed CoTMPP-based electrocatalysts
for oxygen reduction via new preparation strategies", Journal of New Materials for
Electrochemical Systems 7, 85 (2004).

84.

M. Lefevre, E. Proietti, F. Jaouen, and J.-P. Dodelet, "Iron-Based Catalysts with
Improved Oxygen Reduction Activity in Polymer Electrolyte Fuel Cells", Science
324, 71 (2009).

85.

G. Wu, K. L. More, C. M. Johnston, and P. Zelenay, "High-Performance
Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and Cobalt",
Science 332, 443 (2011).

86.

F. Jaouen, E. Proietti, M. Lefevre, R. Chenitz, J.-P. Dodelet, G. Wu, H. T. Chung,
C. M. Johnston, and P. Zelenay, "Recent advances in non-precious metal catalysis
for oxygen-reduction reaction in polymer electrolyte fuel cells", Energy &
Environmental Science 4, 114 (2011).

87.

“Hydrogen, Fuel Cells & Infrastructure Technologies Program: Multi-Year
Research, Development and Demonstration Plan,” Report No. DOE/GO-1020031741 (2005).

88.

“Hydrogen, Fuel Cells & Infrastructure Technologies Program Multi-Year
Research, Development, and Demonstration Plan,” (2007).

89.

A. Widelov and R. Larsson, "ESCA and electrochemical studies on pyrolyzed iron
and cobalt tetraphenylporphyrins", Electrochimica Acta 37, 187 (1992).

90.

H. Schulenburg, S. Stankov, V. Schunemann, J. Radnik, I. Dorbandt, S. Fiechter, P.
Bogdanoff, and H. Tributsch, "Catalysts for the oxygen reduction from heat-treated
iron(III) tetramethoxyphenylporphyrin chloride: Structure and stability of active
sites", Journal of Physical Chemistry B 107, 9034 (2003).

91.

G. Wu, K. Artyushkova, M. Ferrandon, J. Kropf, D. Myers, and P. Zelenay, in
Performance durability of polyaniline-derived non-precious cathode catalysts,
2009, p. 1299.
29

92.

R. Bashyam and P. Zelenay, "A class of non-precious metal composite catalysts for
fuel cells", Nature 443, 63 (2006).

93.

F. Jaouen, J. Herranz, M. Lefevre, J. P. Dodelet, U. I. Kramm, I. Herrmann, P.
Bogdanoff, J. Maruyama, T. Nagaoka, A. Garsuch, J. R. Dahn, T. Olson, S.
Pylypenko, P. Atanassov, and E. A. Ustinov, "Cross-Laboratory Experimental
Study of Non-Noble-Metal Electrocatalysts for the Oxygen Reduction Reaction",
ACS Applied Materials & Interfaces 1, 1623 (2009).

30

2. METAL – NITROGEN – CARBON CATALYSTS
PREPARED BY HIGH-PRESSURE PYROLYSIS FOR
LOW-TEMPERATURE FUEL CELLS∗
2.1 Abstract
Iron-based MNC catalysts for the oxygen reduction reaction (ORR) were produced by
®

pyrolysis of iron-acetate and Ketjenblack

carbon with varying amounts of 2,2´-

bipyridine in a closed, constant-volume quartz vessel between 700-900°C in inert
atmosphere. Electrochemical evaluation for oxygen reduction conducted using a rotating
-2

disc electrode (RDE) indicated a maximum catalytic activity (1.9 mA cm

at 0.65

V/RHE) for catalysts pyrolyzed at 800°C. The total surface nitrogen content detected by
XPS varied from 0.37% to 2.2% when the nitrogen precursor loading was varied from 1.5
to 11.3 wt%. The proportion of nitrogen in the pyridinic-Fe form is estimated to vary
between 4.5 to 19 %, correlating to catalytic activity (current density) variation from 0.07
-2

-2

to 1.9 mA cm at a catalyst loading of 0.5 mg cm . These results support the theory that
pyridinic-Fe nitrogen is part of the catalytic active site responsible for oxygen reduction.

*I acknowledge the collaboration of my former colleague Dr. Kothandaraman
Ramanujam in this work and Dr. Kateryna Artushkova of University of New Mexico for
XPS experiments and analysis.
31

2.2 Introduction
The increasing cost of precious metals has accelerated efforts to replace Pt-group
metals with cost-effective non-precious metal catalysts for the oxygen reduction reaction
(ORR) in low-temperature fuel cells, such as Polymer Electrolyte Fuel Cells (PEFCs) and
Direct Methanol Fuel Cells (DMFCs). Non-precious metal catalysts (NPMCs) based on
transition metals such as Fe and Co have demonstrated ORR activity, albeit lower than
1

that of platinum. Such catalysts also exhibit selectivity towards ORR in the presence of a
methanol, enabling concentrated methanol feed and increased volumetric energy density
in DMFCs.
catalysts,

2-5

Many methods have been developed to prepare non-precious metal

such

as

heat-treating

carbon-supported organometallic

complexes,

pyrolyzing metal and carbon precursors in ammonia or acetonitrile atmosphere,

8-10

6-7

co-

sputtering Co or Fe and carbon in a dinitrogen atmosphere with or without subsequent
heat-treatment,

11

and mixing N-containing ligands with a cobalt oxide sol, followed by

entrapment in a carbon-supported polypyrrole matrix.

12

As explained in Chapter 1, the structure of the metal-nitrogen carbon (MNC) catalytic
sites and the mechanism of ORR at these sites remain uncertain. The most widely cited
active site structures include FeN2/C and FeN4/C or CoN4/C as suggested by Dodelet et
al.,

13-14

nitrogen-containing edge planes without metal as identified by Ozkan et al.
17

and Stevenson et al.

15-16

and surface pyridinic nitrogen in co-ordination with iron as

suggested by Popov et al.

18

and Ruggeri et al.

19

Following this, Charreteur et al.

proposed a possible active site, which contains pyridinic N in a phenanthroline

32

environment present at graphitic crystallite edges. N present in two adjacent crystallites
then coordinate with Fe to form an active site of the form FeN2+2/C.

20

Figure 2.1 indicates various nitrogen functionalities that may be present in the carbon
support. Most literature studies report two types of catalytic sites: a low temperature site
wherein a metal atom is coordinated with four nitrogen atoms and a high temperature site
wherein metal atoms coordinate with two nitrogens.

14,21-23

Iron oxide has also been

9

proposed as a site of lower activity. Among the surface nitrogen groups, two types of N
are widely cited.

18,21,24-25

One is quaternary nitrogen present in the plane of graphene

sheets (Fig. 2.1), often described as graphitic nitrogen, and not catalytically active. The
second nitrogen species is pyridinic, wherein nitrogen atoms present at the edge of
graphene layers bind to two carbon atoms and retain two free electrons that can
+

-

coordinate with metal ions. In the literature, formation of pyridinic N -O species was
observed during exposure to oxygen,

24,26-27

which suggests a possible mechanism of

oxygen reduction.
In addition to the transition-metal and nitrogen precursors, the carbon support and the
method by which these catalysts are synthesized play a critical role in determining the
activity of the catalyst towards ORR. In our approach, a nitrogen source (2,2´-bipyridine),
a high surface area carbon source (Ketjenblack

®

600JD) and a metal precursor (iron

acetate) were pyrolyzed together in a closed container (quartz ampule), generating
autogenic pressure due to bipyridine evaporation and subsequent decomposition. The
synthesis was designed to increase the activity of the nitrogen precursor around the

33

carbon support while maintaining nitrogen mobility, thereby increasing the density of
nitrogen-based catalytic sites. As nitrogen is believed to be a component of the catalytic
site, increasing the surface nitrogen density could potentially increase active site density.
The morphology of the pyrolyzed catalysts was analyzed by nitrogen physisorption, the
composition was studied by XPS and CHN, and the activity of the catalysts was assessed
electrochemically using the thin-film rotating disk and rotating ring-disk electrode
techniques. Deconvolution of the XPS spectra allowed correlation of electrochemical
activity with the concentration of pyridinic nitrogen sites.
2.3 Experimental
®

Iron (II) acetate, Nafion solution (1100 EW, 5 wt. %) and sulfuric acid (ACS grade)
were obtained from Alfa Aesar (Ward Hill, MA). 2-2´-bipyridine was obtained from
®

Fluka (St. Louis, MO). Ketjenblack 600JD carbon black was obtained from Akzo Nobel
(Chicago, IL). Carbon-supported platinum (20 wt% Pt/C) was purchased from BASF
Fuel Cell (Somerset, NJ). Potassium ferrocyanide, (K3Fe(CN)6, ACS grade) was obtained
from Spectrum Chemical (Gardena, CA), and sodium hydroxide (NaOH, ACS grade) was
obtained from J.T. Baker (Phillipsburg, NJ). Pressurized oxygen cylinders were obtained
from Airgas (Lansing, MI). All materials were used as received.
2.3.1 Catalyst Synthesis
®

Ketjenblack

600JD, used as carbon support, was dispersed in a 95% ethanol

solution, to which iron (II) acetate corresponding to 1.6 wt.% Fe was added. This slurry
was stirred for about 6 hr followed by solvent evaporation to yield a dry powder. Powder

34

Figure 2.1 Nitrogen functional groups present on pyrolyzed metal-nitrogen-carbon catalyst
surfaces.

35

samples of 55 mg were ground with 2,2´-bipyridine in amounts ranging from 5 to 125 mg,
and this powder was subsequently charged into a 1.7 ml quartz ampule. The ampule was
flame-sealed under vacuum, and subjected to heat treatment at 700–900°C in a tube
furnace. At around 272.5°C, bipyridine vaporizes, increasing the pressure inside the
vessel, and decomposes at elevated temperature, enabling nitrogen fixation within the
carbon support. Some catalysts thus obtained were subjected to acid leaching in aqueous1
N sulfuric acid at 50°C for about 5 hr, followed by thorough rinsing with Millipore water.
Leaching in acidic solution removed excess iron and soluble impurities.
2.3.2 Physical and Chemical Characterization
Surface analyses of the catalysts were performed by X-ray photoelectron
spectroscopy (XPS) using a Kratos Axis Ultra spectrometer with an Mg Kα X-ray source
operating at 130 W. The base pressure was 2×10
-9

2×10

-10

torr and the operating pressure was

torr. Charge neutralization was used for all samples to prevent charge

accumulation. Three 700 µm × 300 µm areas per sample were analyzed. The survey (low
resolution wide scan) was acquired at 80 eV pass energy for 4 minutes. High-resolution
spectra were acquired at 20 eV pass energy. Surface nitrogen content of the pyrolyzed
samples was also evaluated by CHN (CE440 CHN analyzer, Microanalysis Laboratory,
University of Illinois, Urbana, IL) to quantify bulk amounts of C, H and N. Surface area
and pore size analysis were performed using nitrogen adsorption in a Micromeritics
Gemini V BET analyzer that implements the Barrett-Joyner-Halenda (BJH) method.

36

28

2.3.3 Electrochemical Characterization
Electrochemical characterization was conducted using a glassy carbon rotating disk
2

electrode (RDE, 0.2 cm area) and a rotating ring-disk electrode (RRDE, Pine Research
2

Instrumentation, Raleigh, NC) having a glassy carbon disk (0.25 cm disk area) and a
platinum ring (6.25 mm inner diameter, 7.92 outer diameter). All experiments were
conducted in aqueous 1 N sulfuric acid at 40°C. Potential was measured relative to an
Hg|Hg2SO4 reference electrode (0.68 V/SHE) and all potentials were corrected to the
SHE scale. A platinum wire served as the counter electrode. The catalyst ink was
prepared by dispersing 4 mg of the catalyst powder ultrasonically in a solution mixture
containing 150 µL isopropyl alcohol and 50 µL Nafion (5 wt% solution). A catalyst
-2

loading of 0.5 mg cm was achieved by depositing appropriate amount of the suspension
on the glassy carbon electrode followed by drying for 10 min at 60°C. The electrode thus
obtained was immersed in a solution of 1 N H2SO4 saturated with O2. Linear-sweep
voltammograms were recorded in the potential range 0.85–0.2 V/SHE at a scan rate of
-1

0.5 mV s and 1200 rpm rotation rate.

The collection efficiency, N= Iring/Idisk, for the RRDE was verified using the disk
-2

electrode loaded with carbon-supported platinum at 14 µgPt cm , which catalyzed the
reduction of ferricyanide from 10 mM K3[Fe(CN)6] in deaerated aqueous 0.1 M NaOH,
followed by re-oxidation at the ring electrode poised at 1.5 V/RHE.

29

A collection

efficiency N = 0.39 ± 0.01 was observed, independent of disk potential and rotation rate,
and consistent with manufacturer specifications. The fraction of H2O2 generated during

37

oxygen reduction was measured with the platinum ring poised at 1.2 V while the
potential of the disk electrode was scanned.
2.4 Results and Discussion
Oxygen reduction catalysts were prepared from iron acetate, bipyridine, and
®

Ketjenblack

precursor materials with varying bipyridine proportions such that nominal

nitrogen content varied from 0 to 11.3 wt%. Figure 2.2a shows the differential surface
area, determined by BJH analysis of nitrogen desorption, of Ketjenblack

®

600JD and

catalysts of selected nitrogen content. Most of the observed pore area occurs at pore sizes
ranging between 20 to 60 Å. A peak in pore area at 35 Å pore width decreases in
amplitude (and slightly in peak pore width) with increasing nitrogen content such that it
reaches 64% of its initial value at 10.3 wt% nominal nitrogen. Pore area in the
®9

microporous region at 20 Å and below is small, which is expected for Ketjenblack .

The total BET surface area of all catalysts is presented in Fig. 2b. A 51% decrease in
2

BET surface area from 1440 m /g can be observed when the nominal N content increases
from 1.5 wt% to 6.9 wt%, likely due to the pore closure by bipyridine decomposition
products within micropores of less than 4 nm diameter. Above 6.9 wt% (nominal) N, the
2

surface area stabilizes in the range of 700–900 m /g. However beyond 10.9 wt%, there is
a large decrease in the surface area, likely due to incomplete decomposition of deposited
bipyridine from supersaturated bipyridine vapor, leading to increased pore blocking. As
evident from Table 2.1, a large increase in post-pyrolysis mass and a steep decrease in
surface area can be observed confirming that the bipyridine is not decomposed

38

completely. Even at 10.9 wt% N, the large increase in surface area at 20 Å pore size (Fig.
2.2) may be explained by the deposition of amorphous carbon material from bipyridine.
Similar partially decomposed organic moieties have been reported in the literature. From
20

the results of Raman spectra, Charreterur et al.

proved the presence of amorphous

organic molecules in developmental carbon black due to the injection of CH4 during
carbon black production. Pylypenko et al.

30

reported the presence of pyrolysis products

comprising carbon, nitrogen and metal atoms with intermediate stoichiometry. Sun et
al.

31

found that the surface area of their catalyst decreased linearly with increased loading

of iron in the form of an iron porphyrin. They hypothesized that either metal
nanoparticles block pores in the support or that metal chelates or fragments chemically
bind to the support surface.
High-resolution spectra obtained from X-ray photoelectron spectroscopy (XPS) were
processed to obtain elemental quantification. Seven nitrogen compositions were
considered, with and without leaching by sulfuric acid to remove excess metallic iron. An
unpyrolyzed sample of 1.5 wt% N was also analyzed. The effect of pyrolysis temperature
on composition was determined using catalysts of 10.3 wt% N pyrolyzed at 700°C,
800°C and 900°C, and an iron-free sample of 10.3 wt% N pyrolyzed at 800°C was also
studied. As detailed in Table 2.2, all samples consist of 96–98% C, 0.6 to 2 % O, 0.6–2%
N and negligible Fe (<0.3%). Insignificant changes in composition due to pyrolysis are
observed for the 1.5 wt% N sample: Slightly increased O and N are observed, but Fe
content decreases during pyrolysis.

39

Figure 2.2 Effect of nitrogen content on surface morphology. (a) Differential surface area
distribution of various catalysts obtained from BJH desorption. (b) Observed BET surface
area for various catalysts.

40

Table 2.1 Catalyst yield with varying bipyridine content.
a

b

Nominal nitrogen
content
(wt%)

Product mass
(mg)

1.5

60

57.5

6.9

90

69.0

8.9

110

66.5

9.7

120

65.0

10.3

130

66.5

10.9

140

68.0

11.3

160

123.0

12.5
a

Initial mass
(mg)

180

138.0

Includes 53 mg Ketjenblack

®

1.7 mg Fe acetate and sufficient
b

bipyridine to achieve the indicated mass. Mean product mass
of two batches.

41

Increasing the pyrolysis temperature results in an increase in C content and decrease
in both N and Fe content. It can be observed from Fig. 2.3 that surface nitrogen
composition, as determined by XPS, and bulk nitrogen determined by CHN increase with
increasing nominal nitrogen content for all samples. Surface and bulk N content increases
rapidly beyond 10.3 wt% N, again attributable to deposits of partially decomposed
bipyridine at high nominal nitrogen.
High-resolution N 1s spectra of all samples containing were fitted using four
components: pyridinic (~398.8 eV) and amine or iron-associated pyridinic (~399.3 eV)
quaternary (~401.0 eV) and graphitic (~403.3 eV) nitrogen species. Figures 2.4a and b
show nitrogen spectra and curve fits for as synthesized catalysts of 1.5 wt% N prior to
pyrolysis, without (Fig. 2.4a) and with (Fig. 2.4b) added iron from iron-acetate.
Thpyridinic peak dominates the spectra and little difference is observed except a small
increase in the peak at 399.3 eV that may be attributable to iron-associated moieties. As
shown in Figs. 2.4c and d, pyrolysis appears to increase the prominence of the quaternary
and graphitic peaks.
Table 2.3 shows deconvoluted N1s results for all samples. Unpyrolyzed samples
without Fe contain 0.18 at% N in 399.3 eV attributable to amine species. Addition of Feacetate decreases this content to 0.10 at% ascribed to the overall decrease in the nitrogen
content discussed above. Relative to other components, this component actually increases
2+

upon addition of Fe-acetate and can thus be interpreted as Fe

co-ordinated with

pyridinic nitrogen (Fe-pyr). Pyrolysis decreases pyridinic N and increases quaternary N
content. The lowest levels of pyridinic nitrogen content were observed for the sample

42

Table 2.2 Elemental quantification by XPS.
N content
(wt%)

Pyrolysis
Temp.
(°
C)

1.5 (no Fe)
1.5

Unleached
C
1s
%

O
1s
%

N
1s
%

Fe
2p
%

—
—

98.3
97.0

0.7
2.0

0.97
0.37

—
0.15

1.5

800

97.7

1.8

0.60

0.06

6.9

800

97.0

1.6

1.32

0.11

8.9

800

96.8

1.4

1.68

0.09

9.7

800

97.3

1.4

1.24

0.10

10.3

700

97.4

1.1

1.26

0.27

10.3

800

97.4

1.2

1.14

0.13

10.3

900

98.6

0.6

0.64

0.11

10.9

800

97.3

1.3

1.23

0.16

11.3

800

96.1

1.6

2.18

0.15

43

Figure 2.3 Surface and bulk nitrogen content obtained through XPS and CHN analyses
respectively as a function of nominal nitrogen loading.

44

Figure 2.4 High-resolution N 1s spectra deconvoluted for a subset of samples (a) As
synthesized catalyst (1.5 wt% nominal N, no Fe) before pyrolysis; (b) As-synthesized catalyst
(1.5 wt% nominal N) before pyrolysis; (c) Pyrolyzed catalyst (1.5 wt% nominal N); (d) Pyrolyzed
catalyst (10.3 wt% nominal N).

45

pyrolyzed at 900°C, suggesting that higher pyrolysis temperature leads to increased
conversion of pyridinic nitrogen to quaternary and graphitic nitrogen.
Figure 2.5 shows the trends for pyridinic Fe-N and graphitic nitrogen as a function of
nominal N content. The absolute content of all species increases as nominal N increases
up to 9 wt% N, beyond which Fe-pyr content shows a maximum and pyridinic and
quaternary N decreases to a minimum. This volatility occurs close to the point at which
surface area drops precipitously (Fig. 2.2), suggesting partially decomposed bipyridine
mass may cover or screen out some pyridinic N existing on the high surface area support.
Gloaguen et al. proposed the use of RDE measurements as a fast screening tool for
the characterization of supported catalysts with respect to their ORR activity.

32

Figure 2.6

shows the polarization curves recorded at 1200 rpm and 40°C for a series of catalysts
synthesized at 800°C with precursor nitrogen content varying from 1.5 wt% to 11.3 wt%.
An anodic shift in the onset potential is observed with increasing N content up to 10.3
wt%. Beyond this value, further increase in nominal N content results in a very high
overpotential towards ORR, exemplified by data for 11.3 wt% N. This overpotential is
possibly due to excess partially decomposed bipyridine, as discussed above. In the
absence of any precursor metal, a very high over-potential is observed, confirming the
significance of the metal precursor in the generation of active sites.

1,21

Figure 2.6b shows

the dependence of catalyst activity on N content at 0.65 V/SHE, where the current is
controlled mainly by ORR kinetics and where the 10.3 wt% N catalyst showed the
highest current density. Beyond 10.3 wt% N, the activity of the catalysts decreased, again
attributable to excess partially-decomposed bipyridine that imparts transport limitations

46

Table 2.3 Deconvolution results (at%) for N 1s high-resolution spectra.
Pyrolysis
Temp. (°
C)

Pyridinic
(398.4eV)

Fe-Pyr
(399.3eV)

Quaternary
(401.0eV)

Graphitic
(403.3eV)

1.5, no Fe
1.5
1.5

800

0.67
0.22
0.20

0.18
0.10
0.14

0.12
0.05
0.24

0.03

6.9
8.9
9.7
10.3

800
800
800
700

0.36
0.49
0.41
0.41

0.17
0.17
0.14
0.28

0.69
0.94
0.55
0.50

0.09
0.08
0.15
0.07

10.3
10.3
10.9
11.3

800
900
800
800

0.27
0.17
0.34
0.48

0.24
0.05
0.19
0.12

0.50
0.35
0.63
1.23

0.13
0.08
0.07
0.35

N%

47

Figure 2.5 Surface nitrogen content, in pyridinic form (squares) and graphitic form (circles),
as determined by XPS for varying nominal nitrogen loading. Closed symbols indicate N
content of final catalysts, open symbols indicate observed content of unpyrolyzed
precursors.

48

Figure 2.6 Oxygen reduction at thin-film rotating disk electrodes with Fe-N-C catalysts of
varying composition in O2-saturated, 1 N aqueous sulfuric acid, 40°C. Pseudo-steady state
-1

polarization: Scan rate 0.5 mV s , rotation rate 1200 rpm. (a) 6.9 wt % nominal N, (b) 10 .3
wt % (c) 11.3 wt %, (d) 10.3 wt % w/o Fe, (e) Ketjen carbon black, (f) 20 wt % Pt/C [ETEK],
(g) Current density at 0.65 V/SHE as a function of nominal nitrogen content.

49

in the catalysts due to pore closure. Higher mesoporosity in the carbon support may
ameliorate such effects, and hence a material displaying a combination of microporosity
and mesoporosity may be ideal.
Pyrolysis temperature effects on catalyst activity are shown in Fig. 2.7. Catalysts
pyrolyzed at 700°C and 800°C shows similar onset potential, however the catalyst
pyrolyzed at 800°C showed improved kinetics. Loss of activity at 900°C could be due to
poor fixation of nitrogen in the carbon support, as suggested by XPS analysis. KouteckyLevich analysis was conducted for the ORR using catalysts of 10.3 wt% nominal nitrogen
at potentials in the range of 0.65–0.5 V/SHE. The charge-transfer limited current density
at the electrode surface, ik, was obtained by

1 1
1
=
+
i i k ilim

(2.1)

where i is the current density observed at the RDE and ilim = B ω is the mass-transfer
limited current density at the RDE, with ω the electrode rotation rate in radians per
second, and B the Levich slope given as

B = 0.62nFCO

2

33

2 −1
D 3ν 6

(2.2)

where n is the electron stoichiometry for the ORR, F the Faraday constant, D the
-5

diffusion coefficient for oxygen in 1N H2SO4 solution (1.93×10
-6

-3 35

concentration of dissolved oxygen (1.26×10 mol cm ),
2 -1 36

of sulfuric acid (0.01 cm s ).

-1

-1/2

Plots of i versus ω

50

2

-1 34

cm s ),

CO2 the

and ν the kinematic viscosity

are shown in Fig. 2.8a for

Figure 2.7 Polarization curves showing the effect of heat-treatment temperature on the
catalytic activity towards ORR. Conditions are as in Fig. 2.6.

51

potentials of 0.65–0.5 V/SHE. Linearity in the fit can be observed at all potentials with
less than 2% standard deviation. The nearly parallel slopes observed in Fig. 2.8a confirms
that the ORR follows first-order kinetics and electron stoichiometry does not vary
substantially in this potential range.

33

From the slope, the electron stoichiometry was estimated to be n = 3.57 ± 0.07,
indicating a nearly complete reduction of oxygen to water. Figure 8b shows masstransfer-corrected Tafel plots for the same catalyst. The mass transfer correction was
done using the relation,

ik =

ilim i

(2.3)

(ilim − i)

using ilim obtained from Fig. 8a. The calculated Tafel slope at potentials lower than 0.7
-1

-1

V/SHE is 130 mV dec , close to theoretical value of 120 mV dec

for a one electron
-1

rate-determining step. The Tafel slope obtained above 0.7 V/SHE is 80 mV dec and a
-7

corresponding exchange current density of 4×10

-2

mA cm

was observed at the

reversible potential of 0.93 V/SHE.
Rotating ring disc electrode (RRDE) studies was performed to quantify hydrogen
peroxide production and selectivity of the ORR toward water production. The disc
-1

potential was scanned at 0.5 mV s , while the ring electrode was held at 1.2 V/SHE, a
potential sufficiently high to oxidize any peroxide formed during the ORR. Figure 2.9
shows the disk and ring currents obtained using the optimal catalyst (10.3 wt% N). The

52

electron stoichiometry, n, and fractional H2O2 formation, χH2O2 were determined from
37

the collection efficiency (N=0.39), ring current, Iring, and disk current, Idisk, by

n=

4 I disk
I ring
I disk +
N

χ H 2 O2 =

(2.4)

I
2 ring N

(2.5)

I
I disk + ring N

Poor stability of similar metal-nitrogen-carbon catalysts has been attributed to the attack
of active sites by H2O2 generated during oxygen reduction.

38-40

As shown in Fig. 2.9, the

fractional peroxide generation remained near 6% in the range 0.4–0.9 V/SHE, and the
electron stoichiometric number remained near 3.88 [plot not shown], demonstrating high
–

selectivity toward 4e reduction of oxygen to water. It has been demonstrated that the
amount of H2O2 generated depends strongly on catalyst loading on the RRDE, associated
with residence time of the peroxide intermediate within the porous catalyst.

41

The inset in

Fig. 2.9 shows the effect of catalyst loading on peroxide generation at 0.6 V/SHE. The
-2

observed maximum in χH2O2 is about 6.5% at 200 µg cm , lower than values reported
for similar catalysts.

41

However, under real fuel cell conditions, the thickness of the
-2

cathode catalyst layer is about 100 µm, corresponding to a loading of ca. 2–3 mg cm .
Hence the amount of hydrogen peroxide generated would be less significant due to the
higher residence time and therefore more complete reduction.

53

Figure 2.8 Koutecky-Levich analysis performed on the catalyst having a nominal N content
of 10.3 wt% at 40ºC in 1N aqueous sulfuric acid. (a) Koutecky-Levich plot obtained at three
potentials. (b) Tafel plot for the corresponding catalyst, conditions as in Fig. 2.6.

54

Figure 2.9 Peroxide generation during ORR at Fe-N-C catalysts via rotating ring disk
(RRDE) experiments. Polarization of disk (top) and ring (bottom left) obtained through
RRDE measurements of the catalyst having a nominal N content of 10.3 wt%. The
percentage H2O2 produced is plotted bottom right. Inset: Percentage H2O2 generated for
varying catalyst loading. All conditions are as in Fig. 2.6.

55

Figure 2.10 The effect of pyridinic N content on current density at 0.65 V/SHE.

56

The correlation between ORR current density at 0.65 V/SHE and relative pyridinic Fe-N
content obtained by XPS is shown in Fig. 2.10. Samples containing 10.3 wt% nominal N,
for which the largest current density is observed, also displayed the largest Fe-pyr
nitrogen content (0.24 wt%). The overall trend indicates that increased Fe-pyr content
leads to increased catalytic activity. Current density correlated poorly with variations in
the content of other nitrogen species. These results suggest that the pyridinic form of
nitrogen co-ordinated with Fe is associated with the active site of the catalyst.
Compared to previous reports of metal-nitrogen-carbon ORR catalysts using
®

Ketjenblack

as the carbon support, we demonstrate more than twofold increase in the
9

nitrogen surface content. The optimized catalyst also exhibits an open circuit potential of
0.83 V/NHE, 0.1 V higher than previous results.

42

In RDE-based experiments, the

present observed current density is about 3.5 fold higher at 0.55 V/SHE and 15 fold
higher at 0.7 V/SHE. Charreteur et al. compare the ORR activity of their catalysts based
on Vpr, the potential of peak ORR current in a cyclic voltammetric scan, which is directly
proportional to the logarithm of exchange current density.

20

Vpr measured with our

optimized catalysts is 0.73 V/SHE, equivalent to the highest value reported by Charreteur
et al. for catalysts supported on developmental carbon. Although surface N content in the
optimal catalyst is about 1.5%, smaller than the value reported by Charreteur et al., the
catalytic activity is comparable.

57

2.5 Conclusions
Active Fe-N based electrocatalysts have been prepared using Ketjenblack

®

as a

carbon support and bipyridine as a nitrogen-bearing precursor, by pyrolysis in a quartz
ampule at high pressure. Improved catalytic activity is obtained by tuning the amount of
precursor nitrogen, introduced in the form of bipyridine. In comparison to the use of
flowing nitrogen gas at atmospheric pressure, this approach allows amplification and
control of nitrogen activity during pyrolysis. A two-fold increase in surface nitrogen
content is obtained which correlates with increased open-circuit potential and improved
ORR kinetics in the potential region of interest for operating fuel cells. The current
density obtained at 0.65 V/SHE increases with increasing pyridinic-Fe nitrogen content,
indicating the significance of this species as part of the catalytic site. An optimized
catalyst obtained in this work demonstrates nearly complete reduction of oxygen into
water with high selectivity in the potential range of interest. A key path to improve on
these results may be the use of carbon support having a combination of microporosity and
mesoporosity that simultaneously maximizes active site density and oxygen transport.
Moreover, stability of the present catalysts in acid media is an ongoing concern that is
best addressed by studies of membrane electrode assemblies prepared with these catalysts.

58

REFERENCES

59

2.6 References
1.

F. Jaouen and J. P. Dodelet, "Average turn-over frequency of O-2 electro-reduction
for Fe/N/C and Co/N/C catalysts in PEFCs", Electrochimica Acta 52, 5975 (2007).

2.

A. K. Shukla and R. K. Raman, "Methanol-rusistant oxygen-reduction catalysts for
direct methanol fuel celms", Annual Review of Materials Research 33, 155 (2003).

3.

Y. K. Sun, "Synthesis and electrochemical characterization of a new Se-doped
spinel material for lithium secondary batteries", Journal of Applied
Electrochemistry 31, 1149 (2001).

4.

G. Q. Sun, J. T. Wang, and R. F. Savinell, "Iron(III) tetramethoxyphenylporphyrin
(FeTMPP) as methanol tolerant electrocatalyst for oxygen reduction in direct
methanol fuel cells", Journal of Applied Electrochemistry 28, 1087 (1998).

5.

C. W. B. Bezerra, L. Zhang, K. C. Lee, H. S. Liu, A. L. B. Marques, E. P. Marques,
H. J. Wang, and J. J. Zhang, "A review of Fe-N/C and Co-N/C catalysts for the
oxygen reduction reaction", Electrochimica Acta 53, 4937 (2008).

6.

S. L. Gojkovic, S. Gupta, and R. F. Savinell, "Heat-treated iron(III)
tetramethoxyphenyl porphyrin supported on high-area carbon as an electrocatalyst
for oxygen reduction - I. Characterization of the electrocatalyst", Journal of the
Electrochemical Society 145, 3493 (1998).

7.

P. Gouerec, A. Biloul, O. Contamin, G. Scarbeck, M. Savy, J. Riga, L. T. Weng,
and P. Bertrand, "Oxygen reduction in acid media catalysed by heat treated cobalt
tetraazaannulene supported on an active charcoal: Correlations between the
performances after longevity tests and the active site configuration as seen by XPS
and ToF-SIMS", Journal of Electroanalytical Chemistry 422, 61 (1997).

8.

K. Sawai and N. Suzuki, "Heat-treated transition metal hexacyanometallates as
electrocatalysts for oxygen reduction insensitive to methanol", Journal of the
Electrochemical Society 151, A682 (2004).

9.

F. Jaouen, S. Marcotte, J. P. Dodelet, and G. Lindbergh, "Oxygen reduction
catalysts for polymer electrolyte fuel cells from the pyrolysis of iron acetate
adsorbed on various carbon supports", Journal of Physical Chemistry B 107, 1376
(2003).
60

10.

M. Bron, J. Radnik, M. Fieber-Erdmann, P. Bogdanoff, and S. Fiechter, "EXAFS,
XPS and electrochemical studies on oxygen reduction catalysts obtained by heat
treatment of iron phenanthroline complexes supported on high surface area carbon
black", Journal of Electroanalytical Chemistry 535, 113 (2002).

11.

C. Z. Deng and M. J. Dignam, "Sputtered cobalt-carbon-nitrogen thin films as
oxygen reduction electrocatalysts - I. Physical and electrochemical
characterization", Journal of the Electrochemical Society 145, 3507 (1998).

12.

R. Bashyam and P. Zelenay, "A class of non-precious metal composite catalysts for
fuel cells", Nature 443, 63 (2006).

13.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "Molecular oxygen reduction in PEM
fuel cells: Evidence for the simultaneous presence of two active sites in Fe-based
catalysts", Journal of Physical Chemistry B 106, 8705 (2002).

14.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "Molecular oxygen reduction in PEM
fuel cell conditions: ToF-SIMS analysis of Co-based electrocatalysts", Journal of
Physical Chemistry B 109, 16718 (2005).

15.

P. H. Matter, E. Wang, and U. S. Ozkan, "Preparation of nanostructured nitrogencontaining carbon catalysts for the oxygen reduction reaction from SiO2- and MgOsupported metal particles", Journal of Catalysis 243, 395 (2006).

16.

P. H. Matter, L. Zhang, and U. S. Ozkan, "The role of nanostructure in nitrogencontaining carbon catalysts for the oxygen reduction reaction", Journal of Catalysis
239, 83 (2006).

17.

S. Maldonado and K. J. Stevenson, "Influence of nitrogen doping on oxygen
reduction electrocatalysis at carbon nanofiber electrodes", Journal of Physical
Chemistry B 109, 4707 (2005).

18.

V. Nallathambi, J. W. Lee, S. P. Kumaraguru, G. Wu, and B. N. Popov,
"Development of high performance carbon composite catalyst for oxygen reduction
reaction in PEM Proton Exchange Membrane fuel cells", Journal of Power Sources
183, 34 (2008).

61

19.

S. Ruggeri and J. P. Dodelet, "Influence of structural properties of pristine carbon
blacks on activity of Fe/N/C cathode catalysts for PEFCs", Journal of the
Electrochemical Society 154, B761 (2007).

20.

F. Charreteur, F. Jaouen, S. Ruggeri, and J. P. Dodelet, "Fe/N/C non-precious
catalysts for PEM fuel cells: Influence of the structural parameters of pristine
commercial carbon blacks on their activity for oxygen reduction", Electrochimica
Acta 53, 2925 (2008).

21.

J. M. Ziegelbauer, T. S. Olson, S. Pylypenko, F. Alamgir, C. Jaye, P. Atanassov,
and S. Mukerjee, "Direct spectroscopic observation of the structural origin of
peroxide generation from co-based pyrolyzed porphyrins for ORR applications",
Journal of Physical Chemistry C 112, 8839 (2008).

22.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "O-2 reduction in PEM fuel cells:
Activity and active site structural information for catalysts obtained by the pyrolysis
at high temperature of Fe precursors", Journal of Physical Chemistry B 104, 11238
(2000).

23.

H. Schulenburg, S. Stankov, V. Schunemann, J. Radnik, I. Dorbandt, S. Fiechter, P.
Bogdanoff, and H. Tributsch, "Catalysts for the oxygen reduction from heat-treated
iron(III) tetramethoxyphenylporphyrin chloride: Structure and stability of active
sites", Journal of Physical Chemistry B 107, 9034 (2003).

24.

J. R. Pels, F. Kapteijn, J. A. Moulijn, Q. Zhu, and K. M. Thomas, "Evolution of
Nitrogen Functionalities in Carbonaceous Materials during Pyrolysis", Carbon 33,
1641 (1995).

25.

D. Villers, X. Jacques-Bedard, and J. P. Dodelet, "Fe-based catalysts for oxygen
reduction in PEM fuel cells - Pretreatment of the carbon support", Journal of the
Electrochemical Society 151, A1507 (2004).

26.

K. Stanczyk, R. Dziembaj, Z. Piwowarska, and S. Witkowski, "Transformation of
Nitrogen Structures in Carbonization of Model Compounds Determined by Xps",
Carbon 33, 1383 (1995).

27.

A. N. Buckley, "Nitrogen Functionality in Coals and Coal-Tar Pitch Determined by
X-Ray Photoelectron-Spectroscopy", Fuel Processing Technology 38, 165 (1994).

62

28.

E. P. Barrett, L. G. Joyner, and P. P. Halenda, "The Determination of Pore Volume
and Area Distributions in Porous Substances .1. Computations from Nitrogen
Isotherms", Journal of the American Chemical Society 73, 373 (1951).

29.

U. A. Paulus, T. J. Schmidt, H. A. Gasteiger, and R. J. Behm, "Oxygen reduction on
a high-surface area Pt/Vulcan carbon catalyst: a thin-film rotating ring-disk
electrode study", Journal of Electroanalytical Chemistry 495, 134 (2001).

30.

S. Pylypenko, S. Mukherjee, T. Olson, and P. Atanassov, "Non-platinum oxygen
reduction electrocatalysts based on pyrolyzed transition metal macrocyles",
Electrochimica Acta 53, 7875 (2008).

31.

G. Q. Sun, J. T. Wang, S. Gupta, and R. F. Savinell, "Iron (III)
tetramethoxyphenylporphyrin (FeTMPP-Cl) as electrocatalyst for oxygen reduction
in direct methanol fuel cells", Journal of Applied Electrochemistry 31, 1025 (2001).

32.

F. Gloaguen, F. Andolfatto, R. Durand, and P. Ozil, "Kinetic-Study of
Electrochemical Reactions at Catalyst-Recast Ionomer Interfaces from Thin Active
Layer Modeling", Journal of Applied Electrochemistry 24, 863 (1994).

33.

V. Stamenkovic, T. J. Schmidt, P. N. Ross, and N. M. Markovic, "Surface
composition effects in electrocatalysis: Kinetics of oxygen reduction on welldefined Pt3Ni and Pt3Co alloy surfaces", Journal of Physical Chemistry B 106,
11970 (2002).

34.

N. A. Anastasijevic, Z. M. Dimitrijevic, and R. R. Adzic, "Oxygen Reduction on a
Ruthenium Electrode in Acid Electrolytes", Electrochimica Acta 31, 1125 (1986).

35.

A. Schumpe, I. Adler, and W. D. Deckwer, "Solubility of Oxygen in ElectrolyteSolutions", Biotechnology and Bioengineering 20, 145 (1978).

36. M. Pattabi, R. H. Castellanos, R. Castillo, A. L. Ocampo, J. Moreira, P. J. Sebastian,
J. C. McClure, and X. Mathew, "Electrochemical characterization of tungsten
carbonyl compound for oxygen reduction reaction", International Journal of
Hydrogen Energy 26, 171 (2001).
37.

T. J. Schmidt, U. A. Paulus, H. A. Gasteiger, N. Alonso-Vante, and R. J. Behm,
"Oxygen reduction on Ru1.92Mo0.08SeO4, Ru/carbon, and Pt/carbon in pure and
63

methanol-containing electrolytes", Journal of the Electrochemical Society 147,
2620 (2000).
38.

B. Wang, "Recent development of non-platinum catalysts for oxygen reduction
reaction", Journal of Power Sources 152, 1 (2005).

39.

H. S. Liu, C. J. Song, Y. H. Tang, J. L. Zhang, and H. J. Zhang, "High-surface-area
CoTMPP/C synthesized by ultrasonic spray pyrolysis for PEM fuel cell
electrocatalysts", Electrochimica Acta 52, 4532 (2007).

40.

C. Fierro, A. B. Anderson, and D. A. Scherson, "Electron-Donor Acceptor
Properties of Porphyrins, Phthalocyanines, and Related Ring Chelates - a
Molecular-Orbital Approach", Journal of Physical Chemistry 92, 6902 (1988).

41.

A. Bonakdarpour, M. Lefevre, R. Z. Yang, F. Jaouen, T. Dahn, J. P. Dodelet, and J.
R. Dahn, "Impact of loading in RRDE experiments on Fe-N-C catalysts: Two- or
four-electron oxygen reduction?", Electrochemical and Solid State Letters 11, B105
(2008).

42.

S. Marcotte, D. Villers, N. Guillet, L. Roue, and J. P. Dodelet, "Electroreduction of
oxygen on Co-based catalysts: determination of the parameters affecting the twoelectron transfer reaction in an acid medium", Electrochimica Acta 50, 179 (2004).

64

3. NITROGEN PRECURSOR EFFECTS IN IRON –
NITROGEN – CARBON OXYGEN REDUCTION
CATALYSTS∗

3.1 Abstract
Metal-nitrogen-carbon (MNC) cathode catalysts were synthesized using nitrogen
precursors of varying Nitrogen / Carbon (N/C) ratio by pyrolysis in a constant volume
reaction vessel. Here, we demonstrate that increasing a key property, the N/C ratio of the
nitrogen precursor increased the accessible active site density by reducing carbon
deposition in the pores of the carbon support during pyrolysis. The most active catalysts
were obtained in this work using melamine, having a N/C ratio of 2. Kinetic current
-3

density as high as 15 A cm at 0.8 ViR-free and over 100 h of stable current at 0.5 V were
observed with melamine based MNC catalysts. Composition analysis through XPS
indicated the pyridinic N sites and Pyridinic Fe-N sites contributed to improved activity.

*I acknowledge the help from my colleague Nate Leonard for N2 physisorption
experiments and analysis and Dr. Kateryna Artushkova of University of New Mexico for
XPS experiments and analysis.
65

3.2 Introduction
In Chapter 1 and 2, the significance of nitrogen towards generating active catalysts
has been emphasized. A wide variety of nitrogen precursors have been reported for
synthesizing active MNC catalysts such as pyrrole,
9

6,10-11

ethylene-di-amine,
14

12

purine, pyrimidine,
4,15

nitroaniline,

polyaniline,

20

21-22

acetonitrile,

1-2

16

23

3-4

5

6-

glycine, pyridine,

melamine, urea, thiourea, selenourea,

trimethyltetramine,

phenanthroline,

polypyrrole,

17

tripyridyl triazine,

and cyanamide.

24-25

11,13

18-

ammonia gas,

The properties of the nitrogen

precursor, the synthesis approach and the carbon support determine the structure of MNC
catalytic active sites.
In Chapter 2, a novel means to increase the accessible population of catalytically-vital
pyridinic nitrogen species in the carbon support through high pressure pyrolysis of a
metal precursor, pyridinic nitrogen precursor and a high surface area carbon support was
®

described. Briefly, 2,2´-bipyridine, Ketjenblack 600JD and iron acetate were pyrolyzed
together in a closed container (quartz ampule). At high temperatures, bipyridine
evaporates, decomposes, and precipitates on the carbon support due to autogenic pressure
and upon cooling. The synthesis was designed to increase the activity of the nitrogen
precursor near the carbon support while maintaining nitrogen mobility.

26

A strong

increase in catalytic activity was observed with increasing pyridinic nitrogen content,
confirming N participation in the catalytic active site. However, surface pyridinic N could
not be increased beyond 2 wt%, as measured by XPS. This limit was attributed to pore
blocking by excess carbon from N precursor decomposition that lead to lower observed
surface area and electrochemical activity. Here, we demonstrate a key property of
66

nitrogen

precursors

for

metal-nitrogen-carbon

(MNC)

catalysts,

namely

the

Nitrogen/Carbon (N/C) ratio. Increasing the N/C ratio of the nitrogen precursor increased
the accessible active site density by reducing carbon deposition in the pores of the carbon
support during pyrolysis.
High surface area MNC catalysts were synthesized from pyridinic nitrogen precursors
(Fig. 3.1) that include bipyridine (N/C = 0.2), pyrazine (N/C = 0.5), purine (N/C = 0.8),
and melamine (N/C = 2.0) and the physical properties of the precursors are given in Table
3.1. Catalysts based on melamine, which had the highest N/C ratio, showed improved
activity by a factor of 10 compared to a catalyst made using bipyridine. We surmise that
this is mainly due to increased surface nitrogen and hence increased density of active
catalytic sites, without significant pore blockage via carbon deposition. Stable activity for
over 100 hours was demonstrated using the melamine based catalysts in a single fuel cell
and hence these catalysts can find potential applications in PEMFC, direct methanol fuel
cells (DMFC) and alkaline fuel cells (AFC).
3.3 Experimental
®

Iron (II) acetate, Nafion solution (1100 EW, 5 wt. %), Melamine and sulfuric acid
(ACS grade) were obtained from Alfa Aesar (Ward Hill, MA). 2-2´-bipyridine, pyrazine,
®

purine were obtained from Fluka (St. Louis, MO). Ketjenblack 600JD carbon black was
obtained from Akzo Nobel (Chicago, IL). Pressurized oxygen cylinders were obtained
from Airgas (Lansing, MI). All materials were used as received. Carbon-supported
platinum (20 wt% Pt/C), gas diffusion layer for cathode (ELAT LT 1400W) and gas

67

Bipyridine (C10H8N2)

Pyrazine (C4H4N2)

Purine (C5H4N4)

Melamine (C3H6N6)

Figure 3.1 Structure of pyridinic N precursors with varying N/C ratio.

68

Table 3.1 Properties of N precursors used.
Molar mass Boiling point
(g)
(°
C)

Vapor pressure
(mmHg @ 25°
C)

N Precursor Formula

N/C

Bipyridine

C10H8N2

0.2

158

273

0.0101

Pyrazine

C4H4N2

0.5

80

115

19.7

Purine

C5H4N4

0.8

120

N/A

5.3×10

Melamine

C3H6N6

2.0

126

Sublimes

1.8×10

69

-7

-12

diffusion electrode for anode (ELAT LT 140EW) were purchased from BASF Fuel Cell
(Somerset, NJ).
Catalysts were synthesized in a similar fashion as explained in Chapter 2. Briefly,
®

Ketjenblack 600JD, used as carbon support, was dispersed in a 95% ethanol solution, to
which iron (II) acetate corresponding to 0.75 wt.% Fe was added. This slurry was stirred
for 6 h followed by solvent evaporation to yield a dry powder. Powder samples of 55 mg
were ground with varying amounts of 2,2’ bipyridine, purine, pyrazine and melamine
separately such that the nominal N loading is 6.3 wt% in all cases. The powder was
subsequently charged into a 1.7 ml quartz ampule. The ampule was flame-sealed under
vacuum, and subjected to heat treatment at 800°C in a tube furnace. The catalysts thus
obtained were subjected to acid leaching in aqueous 1N sulfuric acid at 80°C for about 5
hr, followed by thorough rinsing with Millipore water. Leaching in acidic solution
removed excess iron and soluble impurities.
3.3.1

Electrochemical Characterization

Electrochemical characterization was conducted using a glassy carbon rotating disk
2

electrode (RDE, 0.2 cm area) and a rotating ring-disk electrode (RRDE, Pine Research
2

Instrumentation, Raleigh, NC) having a glassy carbon disk (0.25 cm disk area) and a
platinum ring (6.25 mm inner diameter, 7.92 outer diameter). All experiments were
conducted in aqueous 1 N sulfuric acid at 40°C as explained in Chapter 2.
3.3.2

Fuel cell evaluation

Single fuel cell tests were carried out in a fuel cell test stand (Fuel Cell Technologies
Inc.) The cathode catalyst ink was prepared by ultrasonically blending the MNC catalyst
70

®

powder with 5 wt% Nafion in a solution of ethanol and water. The weight ratio of the
®

catalyst powder and dry Nafion was maintained to be 2 to 1. The catalyst ink was then
sprayed onto a gas diffusion layer (ELAT LT 1200W) obtained from ETEK (BASF
-2

Chemical company) using a Paasche airbrush to a desired loading of 2 mg cm . For the
-2

anode, a catalyzed gas diffusion layer (ETEK) consisting of 0.4 mg cm of 20% Pt/C
®

was used. A thin layer of Nafion

-2

(0.5 mg cm ) was applied to both the anode and
®

cathode catalyst layers in order to improve adhesion to the Nafion

membrane. The

®

anode and the cathode were then hot-pressed onto a Nafion 112 membrane (Ion power
o

2

Inc.) at 140 C for 3 min. The MEA geometric area was 5 cm .

Fuel cells were performed using a fuel cell test stand purchased from Fuel cell
Technologies Inc. The cell temperature was set at 80°C and the temperatures of O2 and
H2 humidifiers were set at 90 and 110°C respectively. The anode and cathode gas flow
rates were set at 0.1 slpm and the backpressures were set at 30 psig. Initially, the
27

hydrogen crossover current was measured using linear sweep voltammetry.

The anode

served as the reference and counter electrode and the cathode served as the working
electrode, while equal amount of H2 and N2 gases were purged on the anode and cathode
side respectively. The cell potential was scanned between 0.01 V to 0.5 V. The hydrogen
that permeates from the anode side electrochemically oxidizes at the cathode side and is
measured as a limiting current in the voltage range of 0.35 V to 0.5 V. Limiting currents
2

of 1 mA/cm or less we obtained for all MEAs, indicating low levels of hydrogen
71

crossover. In order to evaluate the ionic membrane resistance, impedance measurements
were carried out at open circuit potential. The resistance was measured by reading the
real part of the impedance when the imaginary part of the impedance reached zero at high
2

frequency. Typically, values of 50–120 mΩ cm were obtained near 10 kHz. All MEAs
were subjected to break-in at 0.5 V cell potential for at least 10h, after which polarization
curves were recorded. Long term durability of the catalyst layer was then measured by
holding the cell potential at 0.5 V for 100 h while recording steady-state current, at the
same conditions as mentioned for the polarization data acquisition.
3.3.3

Physical and chemical characterization

The details of surface analysis of the catalysts through XPS and surface area and pore
size analysis through a BET analyzer are provided in Chapter 2. The bulk Fe content of
the catalysts was evaluated through a Micromass Platform Inductively Coupled PlasmaMass Spectrometer (ICP-MS). X-ray diffraction (XRD) analysis was performed with a
Bruker Davinci Diffractometer using Cu Kα radiation. The lattice spacing between the
graphene layers d002 and the mean stacking height of the graphitic crystallites Lc was
determined using the following equations.

d002 =

λ
(3.1)

2sin θ

where λ is the X-ray wavelength of the Cu Kα radiation and θ the angular position of the
002 diffraction line and Lc is obtained from

72

Lc =

Kλ
β cos θ

(3.2)

where the form factor K is 0.91, β is the full width at half maximum in radians of the 002
28

diffraction peak.

3.4 Results and Discussions
3.4.1 Oxygen reduction activity and selectivity
Figure 3.2a shows a series of polarization curves measured at 1200 rpm and 40°C in
1N H2SO4 for catalysts synthesized using the four nitrogen precursors. A nominal
precursor N loading of 6.3 wt% was maintained in all these catalysts. An anodic shift in
the potential is observed from left to right as a function of the N/C ratio of the N
precursors used in synthesis. The polarization curves were corrected for mass transfer
resistance using equation 2.3 (Chapter 2) and the resulting kinetic current density (ik) is
shown in Fig. 3.2b. All these catalysts exhibited a very high onset potential ~ 0.9 V vs.
2

RHE (I at 0.1 mA/cm ) and the kinetic current at 0.8 V vs. RHE can be read directly from
29

the Tafel plot.

As the precursor N/C ratio increased from 0.2 to 2.0, the kinetic current
-3

-3

density measured at 0.8 V increased by a factor of 4, from 0.6 A cm to 2.4 A cm as
shown in Fig. 3.3. Selectivity of these catalysts towards four-electron oxygen reduction
impacts chemical stability in acidic media during the PEMFC operation. Hydrogen
peroxide radicals generated through incomplete oxygen reduction attack active materials
30

and reduce stability.

RRDE measurements were also carried out under the same

conditions mentioned for RDE measurements and the amount of hydrogen peroxide
generated was calculated using equation 2.5. As seen in Fig. 3.3, low levels of peroxide
73

Figure 3.2. Oxygen reduction current density at thin-film rotating disk electrodes with MNC
catalysts of varying precursor N/C ratio. (a) Pseudo-steady state polarization; (b) iR- and
mass-transfer corrected Tafel curves; Conditions: O2-saturated, 1N aqueous sulfuric acid,
-1

40° Scan rate 0.5 mV s , 1200 rpm, Nominal 6.3 wt% N loading.
C.

74

Figure 3.3 Kinetic current density at 0.8 V/ RHE and % H2O2 at 0.5 V/RHE as a function of
precursor N/C. Conditions: O2-saturated, 1N aqueous sulfuric acid, 40° Scan rate 0.5 mV
C.
-1

s , 1200 rpm, Nominal 6.3 wt% N loading.

75

Figure 3.4 (a) Effect of precursor N/C ratio on differential surface area distribution obtained
from BJH analysis of nitrogen desorption. (b) Observed BET surface area and bulk N
obtained through CHN analysis.

76

generation close to 3-5% was observed for all four nitrogen precursors, indicating the
electron stoichiometry of the catalysts was close to 4.
Barret-Joyner-Halenda (BJH) analyses of nitrogen desorption of the four catalysts
(Fig. 3.4a) indicated that mesoporous pore volume appears in the 20–40 Å range.

31

A

33% increase in BET surface area was observed for the melamine catalysts as compared
to bipyridine (Fig. 3.4b), indicating that increased surface accessibility contributes to
improved activity. Similarly, nitrogen retention of these catalysts, as measured by CHN
combustion analysis, increased with increasing N/C ratio as shown in Fig. 3.4b. A 50 %
increase in N retention was observed for melamine catalysts as compared to bipyridine.
Durability measurements for melamine based catalysts, that indicated the highest oxygen
reduction activity in aqueous electrolyte were obtained by cyclic voltammetry at 50 mV/s
in the 0.6–1.0 V/RHE range, as shown in Fig. 3.5. After 3000 cycles, melamine-based
catalysts showed 17% loss in open circuit potential and ~20% loss in half wave potential
were observed. However, after 3500 cycles a steep decrease in half wave potential and
limiting current was observed due to the loss of active catalyst material from the
electrode surface. Optimization of catalyst loading and ionomer fraction might reduce
material loss.
3.4.2 Composition studies by XPS
High-resolution spectra obtained from X-ray photoelectron spectroscopy (XPS) were
processed to obtain elemental quantification. Catalysts synthesized using the four N
precursors, with and without leaching by sulfuric acid to remove excess metallic iron
were analyzed. High-resolution N 1s spectra were fitted using seven components: the

77

Figure 3.5 Cycle test for Melamine based catalyst in O2-saturated, 0.1 M aqueous sulfuric
-1
acid, 40° Scan rate 50 mV s , 1200 rpm, Nominal 6.3 wt% N loading.
C.

78

Figure 3.6 High resolution N1s XPS spectra deconvoluted for Melamine based leached
catalyst.

79

Table 3.2 Elemental quantification of catalysts by XPS.
N Precursor

C 1s
%

O 1s
%

N 1s
%

Fe 2p
%

Bulk N
Wt%
(CHN)

Bipyridine

90.0

8.5

1.39

0.18

1.3

Pyrazine

85.5

11.8

2.66

0.00

1.0

Purine

84.7

14.7

1.11

0.02

1.5

Melamine

86.7

9.5

3.85

0.01

4.1

80

Figure 3.7 Bulk and surface N obtained through CHN and XPS analyses as a function of
N/C ratio. Nominal N – 6 wt%.

81

predominant peaks observed were due to due to nitrile (398 eV), pyridinic (398.8 eV),
+

pyridinic Fe-N (399.5 eV), quaternary N(CH3)3 (403 eV). Figure 3.6 shows the nitrogen
spectra and curve fits for melamine based leached catalysts. Elemental composition of the
catalysts was obtained by deconvolution of the individual XPS peaks as shown in Table
3.2. Figure 3.7 shows the bulk and surface N composition of the catalysts obtained
through CHN and XPS analyses as a function of precursor N/C ratio. An increase in
nitrogen retention was clearly observed for Melamine based catalysts (N/C = 2) and this
correlated directly to increased kinetic current densities for oxygen reduction.
Figure 3.8a shows the trends of all surface nitrogen species as a function of
precursor N/C ratio. A good correlation between oxygen reduction kinetic current density
and pyridinic and pyridinic Fe-N species was established as shown in Fig 3.8b, indicating
that these two species contribute mainly to catalytically active sites.
3.4.3 Optimization of nominal N content
Nitrogen content of the post-pyrolysis catalyst has been has been established as a
contributing factor to oxygen reduction activity through XPS and CHN measurements.
In an effort to further increase the N retention and site density of the catalysts, the
nominal N loading, using Melamine as the precursor, was increased up to 24% before
pyrolysis. To reduce pressure due to N precursor evaporation during pyrolysis, the tube
volume was increased from 1.7 mL to 2.5 mL. Figure 3.9a shows the mass-transfer
corrected polarization curves obtained through RDE measurements in O2 saturated 1N
-2

H2SO4 at 40°C. High onset potentials (measured at 0.1 mA cm ), close to 0.92~0.95
V/RHE was observed for all catalysts. Increasing the nominal N loading from 6 to 24 %
82

Figure 3.8 a) Summary of XPS results a. Surface N species of catalysts as a function of
precursor N/C ratio; b) Correlation of N content to kinetic current density at 0.8V/RHE.

83

Figure 3.9 a) Mass transfer corrected Tafel curves observed with increasing nominal N
loading: b) Kinetic current density at 0.8 V/ RHE as a function of nominal N loading. O2
-1
saturated, 1 N aqueous sulfuric acid, 40° Scan rate 0.5 mV s , 1200 rpm.
C,

84

increased the kinetic current density at 0.8 V/RHE by a factor of 3, as shown in Fig. 3.9b.
However, the effect of the pyrolysis tube volume on the autogenic pressure generated due
to evaporation of N precursor during pyrolysis need to be established.
3.4.4 Pyrolysis process optimization
Parametric studies were conducted to identify optimal pyrolysis time and temperature
for the melamine-based catalyst using RDE-based current density measurements at 0.8
V/SHE as a figure of merit. Measurements of catalysts pyrolyzed for 5 hours at various
pyrolysis temperatures indicated an optimal pyrolysis temperature of 800°C [Figure not
shown] similar to bipyridine-based catalyst.
It is desired to reduce the pyrolysis time to well below 5 hours in order to increase
pyrolysis throughput and to pave the way to a continuous pyrolysis process with
residence times on the order of seconds. Because optimal pyrolysis time might depend on
pyrolysis temperature and vice versa, a two-parameter parametric study was conducted
incorporating pyrolysis times of 0.5, 1 and 3 hours, combined with pyrolysis
temperatures of 700, 800, and 900°C to form a 9-experiment matrix. As shown in Fig.
3.10a, from the main temperature effects the performance appears to increase
monotonically with pyrolysis time.

However, the time-temperature interaction

parameters of Fig. 3.10b indicate slightly higher levels for a pyrolysis time of 1 hour,
suggesting that this is the optimal time for the current process.
Figure 3.11 shows the experimental XRD spectra of melamine based MNC catalysts
pyrolyzed at different temperatures. A broad carbon peak (002) appears at approximately
2θ = 26°. As the pyrolysis temperature increases beyond 800°C, a sharp diffraction peak

85

Figure 3.10 a) Main effects and b) interaction plot for 3x3 time-temperature parametric
study of pyrolysis time and temperature.

86

Figure 3.11 Experimental XRD spectrum obtained for Melamine-based catalysts pyrolyzed
at varying temperatures.

87

Table 3.3 XRD Analysis of Melamine based catalysts pyrolyzed at varying temperatures.
Pyrolysis
Temperature °
C

2θ
degrees

FWHM(β),
degrees

700

25.3831

5.5

800

24.7158

900
1000

La
Å

Lc
Å

3.5061

14.48

29.79

4.5

3.6022

17.83

36.92

24.3802

4.7

3.6510

17.91

37.18

24.6262

0.3

3.6151

20.21

42.43

88

d002
Å

between 2θ = 40 to 50° appears, which has been attributed to catalytically inactive Fe3N
phase

32-33

and hence a decrease in ORR activity is observed for catalysts pyrolyzed

beyond 800°C. The dimensions of the catalyst crystallites have also been determined
from the analysis of X-ray diffraction line broadening with the help of X-pert High score
software and the details are tabulated in Table 3.3. The parameter d002, which is the
distance between the 002 planes of carbon increases with pyrolysis temperature so does
Lc the average height of graphitic crystallites. Increase in Lc signifies a higher degree of
34

graphitization and this may be linked to improved stability of catalysts,

however

catalytic activity declines beyond 800°C.
3.4.5 Fuel cell measurements
Polarization curves for a single fuel cell membrane-electrode assemblies (MEAs),
fabricated using these MNC cathode catalysts and commercial Pt/C anodes were recorded
in a fuel cell test stand. Polarization curves obtained using catalysts based on bipyridine
and melamine precursors are shown in Figs. 3.12(a-b). Current densities around 210
2

mA/cm has been obtained using Melamine based catalysts in comparison with 50-430
2

mA/cm at 0.6 ViR-free /RHE reported for similar Metal-Nitrogen-Carbon catalysts.

34

The

-3

volumetric current density (A cm , corrected for ohmic losses) is given as a Tafel plot in
-2

Fig. 3.12b. The cathode catalyst loading was maintained at less than 1.3 mg cm to avoid
mass transport limitations. To obtain volumetric current density from mass current
-3

density, an electrode density of 0.4 g cm
35

carbon materials.

was assumed, a typical value for porous

The volumetric current density of the melamine-based catalysts is

89

Figure 3.12 Performance of fuel cell membrane-electrode assemblies employing Fe-N-C
oxygen reduction catalysts. (a & b) Polarization curves for Fe-bipyridine and Fe-melamine
based catalysts, in comparison to a commerical Pt-catalyzed MEA (BASF Corp.);
Conditions: H2–O2 feeds (pO2 = pH2 = 1.5 bar, 80° 100% RH), MNC catalyst loading 1.3
C,
-2

mg cm

Solid and dotted lines indicate forward and reverse scans respectively];

90

Table 3.4 Catalytic Activity of MNC catalysts.
†

RDE
Nitrogen
Precursor

-1

ik / mA mg
(RDE)

MEA
-3

-2

†
-3

ik ( A cm )
(RDE)

ik (A cm )
(MEA)

ik (A cm )
(MEA)

% H2O2
@ 0.6 V

Bulk N
(wt%)

Bipyridine

1.6

0.64

0.0035

2.0

4.8

1.3

Pyrazine

2.2

0.88

—

—

5.9

1.0

Purine

3.9

1.7

—

—

5.7

1.5

Melamine
†

5.9

2.4

0.021

12.4

5.2

4.1

at 0.8 V/RHE, iR-corrected

91

Figure 3.13 (a & b) Durability of Fe-melamine and Fe-bipyridine based MEA fuel cell at 0.5
V/RHE; Conditions: H2–O2 feeds (pO2 = pH2 = 1.5 bar, 80° 100% RH), MNC catalyst
C,
-2

loading 1.3 mg cm ; Solid and dotted lines indicate forward and reverse scans respectively.

92

-3

close to 12.4 A cm , comparable to other literature reports.

14,23

Though this value is

tenfold lower than the DOE target, optimization and further engineering may yield
additional improvements. For the four precursors considered, Table 3.4 summarizes
kinetic current density measured via rotating disk electrode (RDE) experiments, and bulk
nitrogen content measured via CHN combustion experiments. For the baseline material,
bipyridine, and the more promising melamine, fuel cell membrane electrode assembly
(MEA) current density measurements and hydrogen peroxide levels measured via
rotating ring disk electrode (RRDE) experiments are included as well.
Figure 3.13a presents the current density decay for melamine and bipyridine-based
MEAs poised at 0.5 V over a period of 100 h under pure H2 and O2. The performance
loss for both the catalysts was less than 10% over this period. The degradation rate was
similar for both precursors, though the current density of the bipyridine was much lower,
suggesting that the carbon support, comprising 95% of the catalyst material, may play a
crucial role in determining catalyst layer durability.

36

Other issues, such as electrode

structure and water management may also contribute to performance losses. Figure 3.13b
shows polarization curves obtained from melamine-based MEAs before and after
durability measurements, showing a similar degree of degradation.
3.5 Conclusions
We have synthesized a class of MNC catalysts for oxygen reduction in fuel cells,
using a nitrogen precursor with a high N/C ratio that resulted in high N content, surface
area and catalytic kinetic activity as measured using a single fuel cell. In moving from a
low N/C precursor (bipyridine) to one of high N/C (melamine), a four-fold (RDE) and

93

six-fold (MEA) increase in kinetic current density was observed. This may be attributed
to the three-fold increase in bulk nitrogen content for the improved catalyst and hence
increased density of active catalytic sites, without significant pore blockage via carbon
deposition. Composition analysis through XPS indicated the pyridinic N sites and
Pyridinic Fe-N sites contributed to improved activity. Stable activity for over 100 hours
was demonstrated using the melamine-based catalysts in a single fuel cell. With further
improvements in activity and stability, these catalysts can find potential applications in
PEMFC, direct methanol fuel cells (DMFC) and alkaline fuel cells (AFC).

94

REFERENCES

95

3.6 References
1.

K. Lee, L. Zhang, H. Lui, R. Hui, Z. Shi, and J. Zhang, “Oxygen reduction reaction
(ORR) catalyzed by carbon-supported cobalt polypyrrole (Co-PPy/C)
electrocatalysts”, Electrochimica Acta 54, 4704 (2009).

2.

X. Yuan, X. Zeng, H.-J. Zhang, Z.-F. Ma, and C.-Y. Wang, “Improved Performance
of Proton Exchange Membrane Fuel Cells with p-Toluenesulfonic Acid-Doped CoPPy/C as Cathode Electrocatalyst”, Journal of the American Chemical Society 132,
1754 (2010).

3.

A. Garsuch, K. MacIntyre, X. Michaud, D. A. Stevens, and J. R. Dahn, “Fuel cell
studies on a non-noble metal catalyst prepared by a template-assisted synthesis
route”, Journal of the Electrochemical Society 155, B953 (2008).

4.

G. Wu, Z. Chen, K. Artyushkova, F. H. Garzon, and P. Zelenay, in Polyanilinederived non-precious catalyst for the polymer electrolyte fuel cell cathode,
Honolulu, HI, 2008, p. 159.

5.

J. Maruyama and I. Abe, “Fuel cell cathode catalyst with heme-like structure
formed from nitrogen of glycine and iron”, Journal of the Electrochemical Society
154, B297 (2007).

6.

Z. Chen, D. Higgins, H. Tao, R. S. Hsu, and Z. Chen, “Highly Active NitrogenDoped Carbon Nanotubes for Oxygen Reduction Reaction in Fuel Cell
Applications”, Journal of Physical Chemistry C 113, 21008 (2009).

7.

J.-Y. Choi, R. S. Hsu, and Z. Chen, “Highly Active Porous Carbon-Supported
Nonprecious Metal-N Electrocatalyst for Oxygen Reduction Reaction in PEM Fuel
Cells”, Journal of Physical Chemistry C 114, 8048 (2010).

8.

G. Liu, X. Li, P. Ganesan, and B. N. Popov, “Studies of oxygen reduction reaction
active sites and stability of nitrogen-modified carbon composite catalysts for PEM
fuel cells”, Electrochimica Acta 55, 2853 (2010).

9.

J. D. Wiggins-Camacho and K. J. Stevenson, “Effect of Nitrogen Concentration on
Capacitance, Density of States, Electronic Conductivity, and Morphology of NDoped Carbon Nanotube Electrodes”, Journal of Physical Chemistry C 113, 19082
(2009).
96

10.

G. Liu, X. Li, and B. N. Popov, in Stability study of nitrogen-modified carbon
composite catalysts for oxygen reduction reaction in polymer electrolyte membrane
fuel cells, Vienna, 2009, p. 1251.

11.

V. Nallathambi, J. W. Lee, S. P. Kumaraguru, G. Wu, and B. N. Popov,
“Development of high performance carbon composite catalyst for oxygen reduction
reaction in PEM Proton Exchange Membrane fuel cells”, Journal of Power Sources
183, 34 (2008).

12.

J. Maruyama, N. Fukui, M. Kawaguchi, and I. Abe, “Use of purine and pyrimidine
bases as nitrogen sources of active site in oxygen reduction catalyst”, Journal of
Power Sources 194, 655 (2009).

13.

N. P. Subramanian, X. G. Li, V. Nallathambi, S. P. Kumaraguru, H. ColonMercado, G. Wu, J. W. Lee, and B. N. Popov, “Nitrogen-modified carbon-based
catalysts for oxygen reduction reaction in polymer electrolyte membrane fuel cells”,
Journal of Power Sources 188, 38 (2009).

14.

T. E. Wood, Z. Tan, A. K. Schmoeckel, D. O'Neill, and R. Atanasoski, “Nonprecious metal oxygen reduction catalyst for PEM fuel cells based on nitroaniline
precursor”, Journal of Power Sources 178, 510 (2008).

15.

G. Wu, K. Artyushkova, M. Ferrandon, J. Kropf, D. Myers, and P. Zelenay, in
Performance durability of polyaniline-derived non-precious cathode catalysts, 2009,
p. 1299.

16.

H.-J. Zhang, Q.-Z. Jiang, L. Sun, X. Yuan, and Z.-F. Ma, “Influence of heat
treatment on the activity and structure of CoTETA/C catalysts for oxygen reduction
reaction”, Electrochimica Acta 55, 1107 (2010).

17.

S. Li, L. Zhang, H. Liu, M. Pan, L. Zan, and J. Zhang, “Heat-treated cobalttripyridyl triazine (Co-TPTZ) electrocatalysts for oxygen reduction reaction in
acidic medium”, Electrochimica Acta 55, 4403 (2010).

18.

F. Charreteur, F. Jaouen, and J.-P. Dodelet, “Iron porphyrin-based cathode catalysts
for PEM fuel cells: Influence of pyrolysis gas on activity and stability”,
Electrochimica Acta 54, 6622 (2009).

97

19.

M. Lefèvre and J.-P. Dodelet, “Fe-based electrocatalysts made with microporous
pristine carbon black supports for the reduction of oxygen in PEM fuel cells”,
Electrochimica Acta 53, 8269 (2008).

20.

M. Lefèvre, E. Proietti, F. Jaouen, and J. P. Dodelet, in Iron-based catalysts for
oxygen reduction in PEM fuel cells: Expanded study using the pore-filling method,
Vienna, 2009, p. 105.

21.

K. Prehn, A. Warburg, T. Schilling, M. Bron, and K. Schulte, “Towards nitrogencontaining CNTs for fuel cell electrodes”, Composites Science and Technology 69,
1570 (2009).

22.

Y. Tang, B. L. Allen, D. R. Kauffman, and A. Star, “Electrocatalytic Activity of
Nitrogen-Doped Carbon Nanotube Cups”, Journal of the American Chemical
Society 131, 13200 (2009).

23.

M. Lefevre, E. Proietti, F. Jaouen, and J.-P. Dodelet, “Iron-Based Catalysts with
Improved Oxygen Reduction Activity in Polymer Electrolyte Fuel Cells”, Science
324, 71 (2009).

24.

H. T. Chung, C. M. Johnston, F. H. Garzon, and P. Zelenay, in A non-precious
electrocatalyst for oxygen reduction based on simple heat-treated precursors,
Honolulu, HI, 2008, p. 385.

25.

J. Tian, L. Birry, F. Jaouen, and J. P. Dodelet, “Fe-based catalysts for oxygen
reduction in proton exchange membrane fuel cells with cyanamide as nitrogen
precursor and/or pore-filler”, Electrochimica Acta 56, 3276 (2011).

26.

R. Kothandaraman, V. Nallathambi, K. Artyushkova, and S. Calabrese Barton,
“Non-precious oxygen reduction catalysts prepared by high-pressure pyrolysis for
low-temperature fuel cells”, Applied Catalysis B-Environmental 92, 209 (2009).

27.

Y. Song, J. M. Fenton, H. R. Kunz, L. J. Bonville, and M. V. Williams, “Highperformance PEMFCs at elevated temperatures using Nafion 112 membranes”,
Journal of the Electrochemical Society 152, A539 (2005).

28.

A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner, and U. Poschl, “Raman
micro spectroscopy of soot and related carbonaceous materials: Spectral analysis
and structural information”, Carbon 43, 1731 (2005).
98

29.

M. S. Thorum, J. Yadav, and A. A. Gewirth, “Oxygen Reduction Activity of a
Copper Complex of 3,5-Diamino-1,2,4-triazole Supported on Carbon Black”,
Angewandte Chemie-International Edition 48, 165 (2009).

30.

A. Bosnjakovic and S. Schlick, “Naflon perfluorinated membranes treated in Fenton
media: radical species detected by ESR spectroscopy”, Journal of Physical
Chemistry B 108, 4332 (2004).

31.

F. Jaouen, S. Marcotte, J. P. Dodelet, and G. Lindbergh, “Oxygen reduction
catalysts for polymer electrolyte fuel cells from the pyrolysis of iron acetate
adsorbed on various carbon supports”, Journal of Physical Chemistry B 107, 1376
(2003).

32.

E. B. Easton, A. Bonakdarpour, R. Yang, D. A. Stevens, and J. R. Dahn,
“Magnetron sputtered Fe-C-N, Fe-C, and C-N based oxygen reduction
electrocatalysts”, Journal of the Electrochemical Society 155, B547 (2008).

33.

G. Lalande, R. Cote, D. Guay, J. P. Dodelet, L. T. Weng, and P. Bertrand, “Is
nitrogen important in the formulation of Fe-based catalysts for oxygen reduction in
solid polymer fuel cells?”, Electrochimica Acta 42, 1379 (1997).

34.

F. Jaouen, J. Herranz, M. Lefevre, J. P. Dodelet, U. I. Kramm, I. Herrmann, P.
Bogdanoff, J. Maruyama, T. Nagaoka, A. Garsuch, J. R. Dahn, T. Olson, S.
Pylypenko, P. Atanassov, and E. A. Ustinov, “Cross-Laboratory Experimental
Study of Non-Noble-Metal Electrocatalysts for the Oxygen Reduction Reaction”,
Acs Applied Materials & Interfaces 1, 1623 (2009).

35.

H. A. Gasteiger and N. M. Markovic, “Just a Dream-or Future Reality?”, Science
324, 48 (2009).

36.

P. Zelenay, “Advanced Cathode Catalysts,” in United States DOE Hydrogen
Program Annual Progress Report, (2010).

99

4. AMMONIA GENERATING NITROGEN PRECURSORS’
EFFECTS IN IRON – NITROGEN – CARBON OXYGEN
REDUCTION CATALYSTS*

4.1 Abstract
Non-precious Metal-nitrogen-carbon (MNC) cathode catalysts were synthesized
using ammonia generating nitrogen precursors such as urea and ammonium carabamate
via high-temperature pyrolysis in a constant volume reaction vessel. These precursors
etched and increased the surface area of the catalysts during pyrolysis by creating
porosity in the support, while generating various nitrogen bearing functionalities on the
carbon support. N2 adsorption isotherms indicated a 33- % increase in total surface area,
18- % increase in microporosity and 30- % increase in mesoporosity for ammonium
carbamate based catalysts in comparison with bare ketjen black carbon support. Active
catalysts obtained using ammonium carbamate as the nitrogen source indicated an onset
-2

potential close to 0.95 V/RHE and current densities close to 12 mA cm at 0.8 V/RHE in
RDE measurements, which is twice as high in comparison with melamine-based catalysts
addressed in Chapter 3. Through XPS analysis, a correlation between oxygen reduction
activity to pyridinic and pyridinic Fe-sites has been established. However, fuel cell
measurements indicated comparable activity to that of melamine, with volumetric current
3

density close to 15 A/cm at 0.8 V/RHE achieved using ammonium carbamate based
catalysts.

*I acknowledge the help from my colleague Nate Leonard for N2 physisorption
experiments and analysis and Dr. Kateryna Artushkova of University of New Mexico for
XPS experiments and analysis.
100

4.2 Introduction
In Chapter 3, it has been demonstrated that increasing a key property of nitrogen
precursors, the N/C ratio, increased the accessible active site density by increasing the N
content of the catalysts and reducing carbon deposition from the nitrogen precursor in the
pores of the carbon support during pyrolysis. Melamine (N/C = 2.0) based MNC catalysts
-3

exhibited the highest ORR activity (13.2 A cm
2

1,2

(1120 m /g).

at 0.8 V) and very high surface area

However, to completely eliminate carbon deposition effects from N

precursors, ammonia generating, carbon-free precursors have been considered as N
precursors in this work.
Various groups have used gaseous ammonia as a nitrogen precursor for synthesizing
oxygen reduction catalysts.

3-8

The reaction of ammonia with the carbon support in the

presence of transition metal may have two major consequences: 1) ammonia may etch
and create porosity in the carbon support, and 2) ammonia may generate nitrogen bearing
4-6

functionalities on the carbon surface.

High surface area and porosity of carbon supports are critical for achieving a high
dispersion of active sites in heterogeneous catalysts. In most conventional carbon
supports, a large part of this surface area may be contained in narrow micropores that are
inaccessible to gaseous reactants. Dodelet et al. claimed that micropores created in their
non-porous carbon support due to ammonia etching during pyrolysis hosted a large
number of Fe-N4 sites that are believed to be catalytically active. However, the same
group recently reported that pyrolysis in ammonia resulted in highly active but unstable
9

catalysts. They explain that micropores formed under ammonia are either prone to
101

flooding during fuel cell operation or active N bearing sites undergo undesirable chemical
9

changes in the acidic medium leading to their instability.

Dai et al. recently developed metal-free nitrogen doped ordered mesoporous carbons
through NH3 activation at high temperatures. The reactions between carbon and NH3
resulted in replacement of oxygen bearing species by nitrogen containing groups and
etching of carbon fragments by the radicals generated by decomposition of NH3 at
10

elevated temperatures.

Similarly, Mangun et al. describe the treatment of activated

carbon fibers with ammonia at high temperature. They report that there was an increase
in the surface area and pore size of the activated carbon with increasing treatment
11

temperature demonstrating that the ammonia acted as an etchant.

In addition, they

reported enhanced adsorption of an acidic gas, which confirmed the presence of a basic
nitrogen functional group introduced by the ammonia treatment.
The mechanism of incorporating nitrogen functionality into the carbon involves
decomposition of ammonia at high temperature into radicals such as NH2, NH, and H.

12

These radicals can etch carbon fragments leading to increased porosity and they can
replace oxygen-containing species on the carbon to form surface groups such as –NH2, –
13

CN, pyrrolic and quaternary nitrogen thus producing surface functionalization.

In spite

of the controversies surrounding active site structure and oxygen reduction mechanism on
these sites, it has been repeatedly established that nitrogen surface concentration on
4-6

porous carbon support is a key to improving activity.

102

In this work, highly porous Ketjen black 600JD was used as the carbon support and
solid compounds such as ammonium hydroxide, urea and ammonium carbamate (Table
4.1) have been employed as nitrogen precursors. The effect of these precursors on the
generation of oxygen reduction sites during pyrolysis has been addressed. Carbon
deposition from urea and ammonium carbamate have been prevented by adding
equimolar amounts of water and the decomposition of these salts proceed according to
the reactions listed in Table 4.1. A schematic representation of the catalytic site formation
in the pores of the catalyst is illustrated in Fig. 4.1.
Through this approach, high pressure and high temperature, combined with increased
residence time of the N precursor, will result in an increase in the number of reaction
sites by effective decomposition of ammonia in the presence of Fe. The difference in
oxygen reduction activity of catalysts made with the different precursors is assessed, and
oxygen reduction activity is correlated to nitrogen content and microporous and
mesoporous surface area of the catalyst.
4.3 Catalyst Synthesis
®

Iron (II) acetate, Nafion solution (1100 EW, 5 wt. %) and sulfuric acid (ACS grade)
were obtained from Alfa Aesar (Ward Hill, MA). Ammonium hydroxide, urea,
®

ammonium carbamate was obtained from Fluka (St. Louis, MO). Ketjenblack

600JD

carbon black was obtained from Akzo Nobel (Chicago, IL). Carbon-supported platinum
(20 wt% Pt/C) and ELAT gas diffusion layers were purchased from BASF Fuel Cell
(Somerset, NJ). Pressurized oxygen cylinders were obtained from Airgas (Lansing, MI).
All materials were used as received.
103

Table 4.1 Structure of N precursors used and their decomposition reactions.
Ammonia generating
precursor

Formula

Decomposition reactions

NH4+
Ammonium hydroxide

HO-

NH4OH

NH2CONH2

Urea + water

Ammonium carbamate

HCNO + H2O

NH4+
-O

NH3 + H2O

NH3 + HCNO
NH3 + CO2

NH2
NH4COONH2

O

104

2NH3 + CO2

∆ 800 C , 3 h

NH3

Mesopores

Fe
N
C
H

Figure 4.1 Schematic showing the formation of pores due to ammonia etching during highpressure pyrolysis and the creation of active reaction sites.

105

Catalysts based on ammonia generating precursors were synthesized in a similar
®

fashion described in Chapter 2 and 3. Briefly, Ketjenblack

600JD was dispersed in a

95% ethanol solution, to which iron (II) acetate corresponding to 0.75 wt% Fe was added.
This slurry was stirred for 6 hr followed by solvent evaporation to yield a dry powder.
Powder samples of 60 mg were ground with varying amounts of ammonia based nitrogen
rich precursors such as ammonium hydroxide, urea and ammonium carbamate, to achieve
6.3 wt% nominal N loading. Equimolars amounts of water was added to urea and
ammonium carbamate based samples to eliminate carbon deposition from precursor. The
powder was subsequently charged into a 3.2 mL quartz ampule. The ampule was flamesealed under vacuum, and subjected to heat treatment at 800°C for 3h in a tube furnace.
To remove excess iron, the resulting catalysts were exposed to aqueous 1 N sulfuric acid
at 80°C for about 5 hr and filtered and washed several times with Millipore water. The
sample was then oven-dried overnight at 80°C before use.
4.4 Experimental
Surface area and pore size analysis were performed using nitrogen adsorption in a
Micromeritics ASAP 2020 analyzer. Weighed catalysts were degassed at 200°C for 6 h
under vacuum and were then transferred to the analysis port. Adsorption and desorption
isotherms were collected at 77 K using N2 as the adsorbate. The adsorption data was
analyzed using a non-local density functional theory model (NLDFT) assuming slit pore
geometry to determine the pore size distribution. The sample morphology was
characterized by scanning electron microscopy (SEM) on a JEOL 6400V with LaB6
emitter. Transmission electron microscopy images were obtained using a JEOL 2200FS
equipment operating at 200 kV.
106

Surface elemental analyses of the catalysts were performed by X-ray photoelectron
spectroscopy (XPS) using a Kratos Axis Ultra spectrometer using an Mg Kα X-ray
source operating at 130 W and the details of the experimental parameters are provided in
Chapter 3. High-resolution spectra were acquired at 20 eV pass energy for C 1s, N 1s, O
1s and Fe 2p. Also, the surface nitrogen and Iron content of the pyrolyzed samples was
obtained through XPS.
Electrochemical characterization was conducted using a glassy carbon rotating disk
2

electrode (RDE, 0.2 cm area) and a rotating ring-disk electrode (RRDE, Pine Research
2

Instrumentation, Raleigh, NC) having a glassy carbon disk (0.25 cm disk area) and a
platinum ring (6.25 mm inner diameter, 7.92 outer diameter in aqueous 1 N sulfuric acid
at 40°C, as described previously in Chapter 2 and 3. Single fuel cell tests were carried out
in a fuel cell test stand (Fuel Cell Technologies Inc.). Membrane electrode assemblies
(MEAs) were fabricated as described in Chapter 2.
4.5 Results and Discussions
4.5.1

Structural characterization

According to IUPAC notation, pores of size < 2 nm are classified as micropores,
between 2 and 50 nm are classified as mesopores and pores of size > 50 nm are classified
14,15

as macropores.

Figure 4.2a shows the adsorption and desorption isotherms for Ketjen

carbon black 600JD and catalysts based on urea and ammonium carbamate precursors.
All these isotherms are of type II according to IUPAC specification, which are
characteristic of systems containing micropores and they exhibit a steep rise in the
isotherm at low partial pressures.

14,15

At partial pressures close to unity, no saturation is
107

Figure 4.2 (a) N2 adsorption isotherms for Ketjen black and catalysts based on urea and
ammonium carbamate precursors. (b) Pore size distribution of carbon black and catalysts.

108

observed indicating the presence of macropores or large aggregates of particles. Both the
Ketjen black carbon support and the catalysts exhibited hysteresis of type H3 according
to IUPAC notation in the adsorption isotherms reaching a limiting hysteresis close to 0.45
P/P0, which is close to the critical partial pressure of 0.4 for N2.

16

The well pronounced

hysteresis indicates the presence of an extensively developed pore network, and may
incorporate larger pores trapped in between a network of smaller pores.
Pore size distributions of ammonia- based catalysts were obtained from the
adsorption branch of the isotherms is presented in Figs. 4.2-b. An increase in
microporous area (<2nm) was observed for urea based catalysts compared to melamine
based catalysts, however for ammonium carbamate based materials, an increase in both
microporous and mesoporous area was observed, signifying that ammonia generated
during pyrolysis etched the carbon support.
Table 4.2 summarizes the microstructural characteristics of Ketjen black and urea
ammonium carbamate and melamine based catalysts. A slight decrease in porosity was
observed for urea and melamine based catalysts in comparison to Ketjen carbon support,
possibly due to pore filling from carbon and nitrogen from the nitrogen precursor.
However in ammonium carbamate based catalysts, no pore blocking was witnessed,
2

which is evident from the fact that the surface area of the catalyst increased (~2100 m /g)
2

in comparison with the Ketjen carbon support (~1400 m /g).

SEM images of the Ketjen carbon support and catalysts synthesized with ammonium
carbamate shown in Figs. 4.3a and b respectively. Ketjen black carbon support is a high
surface area carbon support with uniform particle size distribution and less agglomeration
109

Table 4.2 Structural characteristics of carbon support and catalysts.
Microporosity
<2 nm

Mesoporosity
2-10nm

BET surface
area total

Macroporosity
> 10 nm

m /gmaterial

m /gmaterial

m /gmaterial

m /gmaterial

Ketjen Black (KB)

413

241

1439

785

Urea/Fe/KB

293

208

1140

639

507.4

411

2015

1095

192

188

1120

740

Catalyst/Support

Ammonium
Carbamate/Fe/KB
Melamine/Fe/KB

2

2

110

2

2

(a)

1 µm

(b)

1 µm

Figure 4.3 SEM images of (a)
carbamate/Fe/Ketjen black catalysts

Ketjen

111

black

carbon

support

(b)

Ammonium

as can be seen in Figure 4.3a. Introduction of Fe-N from ammonium carbamate and Iron
acetate precursors during pyrolysis on the support have induced few particle
agglomeration into spherical particles of ~ 100nm (brighter spots in the image), however
there is no evidence of significant pore blockage in the carbon support.
4.5.2

Electrochemical characterization

Thin-film rotating disc electrode (RDE) and rotating ring disc electrode (RRDE)
studies were used to assess oxygen reduction activity and stability of the catalysts. Figure
4.4-a shows a series of polarization curves measured at 1200 rpm and 40°C in 1 N H2SO4
for catalysts synthesized using the ammonia based nitrogen precursors. The onset
2

potential was measured as the potential at 0.1 mA/cm . Very high onset potentials close
to 0.93 ~ 0.95 V/RHE was observed for all catalysts, which is roughly 50 mV less than
that of commercial platinum based electrocatalysts. Ammonium carbamate based
catalysts exhibited the lowest overpotential for oxygen reduction, followed by urea and
ammonium hydroxide. Catalysts synthesized using urea without adding equimolar
amounts of water is also shown in Fig. 4.4 for comparison and exhibited comparatively
lesser activity than the catalysts synthesized by adding equimolar amounts of water,
probably due to carbon interference from urea precursor.
The polarization curves were corrected for mass transfer resistance and the resulting
kinetic current density (ik) is shown in Fig. 4.5b; the kinetic current at 0.8 V vs. RHE can
2

be read directly from the Tafel plot. Kinetic current density as high as 23 mA/cm , was
estimated using ammonium carbamate as the N precursor. The increase in oxygen
reduction activity can be attributed to the huge increase in surface area of the catalyst
112

Figure 4.4 (a) and (b) Polarization curves obtained using RDE and Mass transport corrected Tafel
representation of polarization plots of Fe–N–C catalysts synthesized with different ammoniagenerating precursors. (c) Kinetic current densities at 0.8 V/RHE Conditions: O2-saturated, 1N
-1
aqueous sulfuric acid, 40° Scan rate 0.5 mV s , 1200 rpm, Nominal 6.3 wt% N loading.
C.

113

Figure 4.5 (a) and (b) Polarization curves obtained using RRDE and amount of hydrogen
peroxide generated Fe–N–C catalysts synthesized with different ammonia-generating
-1

precursors. Conditions: O2-saturated, 1 N aqueous sulfuric acid, 40° Scan rate 0.5 mV s ,
C.
1200 rpm, Nominal 6.3 wt% N loading.

114

post pyrolysis due to ammonia etching for ammonium carbamate based catalysts. RRDE
voltammograms and the amount of hydrogen peroxide generated due to incomplete twoelectron oxygen reduction are shown in Figs. 4.5a and b.
The amount of hydrogen peroxide radicals generated during oxygen reduction is
reported to be directly dependant on the loading of the catalyst on the electrode, due to
the fact these radicals might find another active site in a thicker electrode and gets
reduced to water through another two-electron reduction. Hence, the loading of the
-2

catalyst for RRDE has been maintained at 0.2 mg cm , which was the lowest possible
loading achieved for the catalysts using the RDE ink composition described in Chapter 2.
Ammonium carbamate based catalysts indicated low levels of hydrogen peroxide, close
to 4.5 % at 0.5 V and 7% at 0.8 V, closer to an electron number of ~3.4-3.7. These results
clearly indicated that an increased pore volume/active surface area of the catalyst led to
-1

improved activity and selectivity for oxygen reduction. Figure 4.6 shows a plot of i vs
-1/2

ω

for ammonium carbamate based catalyst at 0.8V and 0.7 V vs. RHE. The fits were

linear and parallel slopes indicate that oxygen reduction follows first-order stoichiometry.
The electron stoichiometry number calculated from the slope was close to ~3.5, which
agrees with the values obtained through RRDE analysis, the details of the analysis can be
found in Chapter 2, section 2.5.
4.5.3

Compositional analysis through XPS

Besides the surface area, we have demonstrated earlier in our reports that the amount
of nitrogen in the post-pyrolysis catalysts plays a vital role in determining the oxygen
reduction activity. The elemental composition of the catalysts and the different nitrogen
115

Figure 4.6. Koutecky-Levich analysis performed on Fe–N–C catalysts synthesized with
ammonium carbamate precursors. Conditions: O2-saturated, 1N aqueous sulfuric acid, 40°
C.
-1
Scan rate 5 mV s , Nominal 6.3 wt% N loading.

116

Table 4.3 Elemental quantification (wt %) by XPS.
Catalyst

C 1s

O 1s

N 1s

Fe 2p

Urea

90.6

8.9

0.5

0.01

Urea Leached

86.2

12.4

1.37

0.03

Ammonium Carbamate

87.1

2.50

0.15

0.01

Ammonium Carbamate
Leached

94.3

12.8

3.17

0.05

117

species present in the catalysts were obtained from high-resolution spectra obtained from
X-ray photoelectron spectroscopy (XPS). Catalysts synthesized using urea and
ammonium carbamate N precursors, with and without leaching by sulfuric acid to remove
excess metallic iron were analyzed. Table 4.3 presents the elemental quantification of the
catalysts obtained through deconvolution of the individual peaks. Ammonium carbamate
indicated higher surface N content (3 wt %) compared to that of urea (1 wt %). Higher N
content combined with higher pore volume in the catalysts resulted in higher oxygen
reduction activity. Trace amounts of transition metal Fe was detected in both urea and
ammonium carbamate based catalysts indicating the possibility of metal-centered activity
as opposed to other reports, where metal-free active sites have been cited, because XPS
18,19

could not detect any transition metal on the surface of their catalysts.

Acid leaching

in sulfuric acid also introduced large amounts of oxygen in the catalysts.
High-resolution N 1s spectra were fitted using seven components as shown in Fig.
4.7: the predominant peaks observed were due to nitrile (398 eV), pyridinic (398.8 eV),
+

pyridinic Fe-N (399.5 eV), quaternary N(CH3)3 (403 eV). Figure 4.7 shows the nitrogen
spectra and curve fits and Table 4.4 represents the deconvolution results for urea and
ammonium carbamate based leached catalysts.
Both urea and ammonium carbamate based catalysts displayed large amounts of
pyridinic and pyridinic Fe-N types of nitrogen, that are mainly believed to be part of
1,2,20

active sites.

The lone pair of electrons in Pyridinic N groups present in graphitic

edges has been frequently reported to be active for oxygen reduction.

118

7-10

However, lower

Figure 4.7 High resolution N1s spectra deconvoluted for (a) Urea and (b) Ammonium
carbamate based leached catalysts.

119

Table 4.4 Deconvolution results (wt %) for N1s high-resolution spectra.
N
Precursors

Nitrogen functionalities in unleached
samples (B.E. /eV)

Nitrogen functionalities in leached
samples (B.E. /eV)

Pyridinic
(398.8)

Graphitic
(400.9)

Pyridinic Fe-N
(399.5)

Pyridinic
(398.8)

Graphitic
(400.9)

Pyridinic Fe-N
(399.5)

Urea

16.8

5.3

25.0

20.0

3.8

29.7

Ammonium
Carbamate

21.3

7.5

38.0

23.2

7.7

23.8

120

amounts of graphitic N that are believed to contribute to stability of the catalysts was
21

observed.

The correlation between oxygen reduction activity at 0.8 V/RHE measured through
RDE experiments and Fe-Pyridinic content, pyridinic content obtained through XPS is
shown in Fig. 4.8. Also shown is the correlation between microporosity and mesoporosity
of the catalysts to current density. The trend indicates that both nitrogen content and
porosity of the catalyst influences the oxygen reduction activity. Improved porosity and
higher concentration of pyridinic and pyridinic-Fe content of ammonium carbamate
based catalysts resulted in high oxygen reduction activity. Though, the structure of the
active site of the catalyst post pyrolysis could not be clearly identified because of the
complexities related to high temperature pyrolysis, some insight on the active centers for
oxygen reduction can be gained through Extended X-ray Absorption Fine Structure
(EXAFS) analysis, which has been addressed in the next chapter.
Polarization curves for a single fuel cell membrane-electrode assemblies (MEAs),
fabricated using these MNC cathode catalysts and commercial Pt/C anodes were recorded
in a fuel cell test stand. Fuel cell polarization curves obtained using catalysts based on
urea and ammonium carbamate precursors are shown in Figs. 4.9(a-b). At 0.6 ViR-free
2

/RHE, current densities around 235 mA/cm has been obtained using urea based catalysts
2

and 215 mA/cm has been obtained for ammonium carbamate based catalysts.
-2

Similar values of 50 – 430 mA cm at 0.6 ViR-free /RHE has been reported for state22

-3

of-art Metal-Nitrogen-Carbon catalysts. The volumetric current density (A cm ,

121

Figure 4.8 Correlation of RDE kinetic activity at 0.8V /RHE to porosity and nitrogen content.

122

Figure 4.9. Performance of fuel cell membrane-electrode assemblies employing Fe-N-C
oxygen reduction catalysts. (a) and (b) Polarization curves for Fe-urea and Fe-ammonium
carbamate based catalysts in comparison with pyridine and melamine based Fe-N-C
catalysts; Conditions: H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80° 100% RH), Fe-N-C catalyst
C,
-2

loading 1.3 mg cm

.

123

corrected for ohmic losses) is given as a Tafel plot in Fig. 4.9b in comparison with
melamine and bipyridine based catalysts. Catalysts based on urea demonstrated
-3

reasonable activity (15 A cm ) at 0.8V/RHE. However, catalysts synthesized using
ammonium carbamate precursors displayed a lower onset potential ~ 0.8 V compared to
0.85 V for urea based catalysts. In the intermediate mixed potential regions, (0.75 – 0.5
V), their activities matched. Lack of a clear Tafel slope at high potentials (0.95 – 0.8 V)
prevents extrapolation of the true kinetic current density at 0.8 V. Other reports have
demonstrated a well-defined Tafel regime using thicker Nafion 117 membranes to reduce
[21,22]

hydrogen gas cross-over.

Besides, at high current densities, significant water

flooding was witnessed leading to a decrease in performance. This may attributed to very
high surface area of ammonium carbamate based catalysts, particularly their high micro
and mesoporosity, as liquid water formed during oxygen reduction may be retained in the
narrow pores

[21,22]

. Optimization of catalyst loading to nafion content in these electrodes

may maximize the performance of these catalysts without introducing flooding.
Figure 4.10a presents the current density decay for urea-based MEAs poised at 0.5 V
over a period of 50 h under pure H2 and O2. The performance loss for the catalyst was
less than 8% over this period. Figure 4.11 represents the current density decay for
ammonium carbamate based catalyst. The degradation rate was similar for both
precursors suggesting again that the carbon support, comprising 95% of the catalyst
23

material, may play a crucial role in determining catalyst layer durability. Other issues,
such as electrode structure and water management may also contribute to performance

124

Figure 4.10 (a) and (b) Durability of Fe-urea based MEA fuel cell at 0.5 V/RHE; Conditions:
-2

H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80° 100% RH), MNC catalyst loading 1.3 mg cm .
C,

125

Figure 4.11 Durability of Fe-ammonium carbamate based MEA fuel cell at 0.5 V/RHE;
Conditions: H2–O2 feeds (PO2 = PH2 = 1.5 bar, 80° 100% RH), MNC catalyst loading 1.3
C,
-2
mg cm .

126

losses. Figure 4.9b shows polarization curves obtained from urea-based MEAs before and
after durability measurements, showing a similar degree of degradation.
4.6 Conclusions
Non-precious Metal Nitrogen Carbon catalysts were synthesized using Iron acetate,
ammonia generating nitrogen precursors and Ketjen black carbon support through a hightemperature, high pressure pyrolysis process. Ammonia gas generated during pyrolysis
etched the carbon support and increased the pore-volume of the catalyst. Ammonium
carbamate based catalysts exhibited over ~10% increase in microporosity and
mesoporosity and surface N content as high as 3.8 wt % was detected using XPS analysis.
Oxygen reduction activity evaluated using RDE and RRDE techniques indicated over
two-fold increase in kinetic current densities for urea and ammonium carbamate based
catalysts in comparison with melamine- based catalysts and less than 7% H2O2 was
observed for both catalysts. Deconvolution of N1s spectra obtained through XPS to
individual peaks indicated a higher concentration of pyridinic and pyridinic-Fe sites,
which are believed to be active for oxygen reduction. Single fuel cell measurements
measured using MEAS fabricated using urea and ammonium carbamate catalysts as
3

cathodes showed reasonable activity and volumetric current density close to 15 A/cm at
0.8 V/RHE. However, the role of transition metal in creating active sites and the exact
site structure in terms of metal-nitrogen coordination needs to be addressed, which will
be the primary objective of Chapter 5.

127

REFERENCES

128

4.7 References
1.

R. Kothandaraman, V. Nallathambi, K. Artyushkova, and S. Calabrese Barton,
"Non-precious oxygen reduction catalysts prepared by high-pressure pyrolysis for
low-temperature fuel cells", Applied Catalysis B-Environmental 92, 209 (2009).

2.

V. Nallathambi, N. Leonard, R. Kothandaraman, and S. C. Barton, "Nitrogen
Precursor Effects in Iron-Nitrogen-Carbon Oxygen Reduction Catalysts",
Electrochemical and Solid State Letters 14, B55 (2011).

3.

D. S. Geng, Y. Chen, Y. G. Chen, Y. L. Li, R. Y. Li, X. L. Sun, S. Y. Ye, and S.
Knights, "High oxygen-reduction activity and durability of nitrogen-doped
graphene", Energy & Environmental Science 4, 760 (2011).

4.

F. Jaouen, S. Marcotte, J. P. Dodelet, and G. Lindbergh, "Oxygen reduction
catalysts for polymer electrolyte fuel cells from the pyrolysis of iron acetate
adsorbed on various carbon supports", Journal of Physical Chemistry B 107, 1376
(2003).

5.

M. Lefèvre and J.-P. Dodelet, "Fe-based electrocatalysts made with microporous
pristine carbon black supports for the reduction of oxygen in PEM fuel cells",
Electrochimica Acta 53, 8269 (2008).

6.

M. Lefevre, E. Proietti, F. Jaouen, and J.-P. Dodelet, "Iron-Based Catalysts with
Improved Oxygen Reduction Activity in Polymer Electrolyte Fuel Cells", Science
324, 71 (2009).

7.

G. C. K. Liu and J. R. Dahn, "Fe-N-C oxygen reduction catalysts supported on
vertically aligned carbon nanotubes", Applied Catalysis a-General 347, 43 (2008).

8.

L. T. Qu, Y. Liu, J. B. Baek, and L. M. Dai, "Nitrogen-Doped Graphene as Efficient
Metal-Free Electrocatalyst for Oxygen Reduction in Fuel Cells", Acs Nano 4, 1321
(2010).

9.

F. Jaouen, E. Proietti, M. Lefevre, R. Chenitz, J.-P. Dodelet, G. Wu, H. T. Chung, C.
M. Johnston, and P. Zelenay, "Recent advances in non-precious metal catalysis for
oxygen-reduction reaction in polymer electrolyte fuel cells", Energy &
Environmental Science 4, 114 (2011).

10.

X. Q. Wang, J. S. Lee, Q. Zhu, J. Liu, Y. Wang, and S. Dai, "Ammonia-Treated
Ordered Mesoporous Carbons as Catalytic Materials for Oxygen Reduction
Reaction", Chemistry of Materials 22, 2178 (2010).

11.

C. L. Mangun, K. R. Benak, J. Economy, and K. L. Foster, "Surface chemistry, pore
sizes and adsorption properties of activated carbon fibers and precursors treated
with ammonia", Carbon 39, 1809 (2001).
129

12.

B. Stohr, H. P. Boehm, and R. Schlogl, "Enhancement of the catalytic activity of
activated carbons in oxidation reactions by thermal-treatment with ammonia or
hydrogen-cyanide and observation of a superoxide species as a possible
intermediate", Carbon 29, 707 (1991).

13.

H. P. Boehm, G. Mair, T. Stoehr, A. R. Derincon, and B. Tereczki, "Carbon as a
catalyst in oxidation reactions and hydrogen halide elimination-reactions", Fuel 63,
1061 (1984).

14. J. Rouquerol, D. Avnir, C. W. Fairbridge, D. H. Everett, J. H. Haynes, N. Pernicone,
J. D. F. Ramsay, K. S. W. Sing, and K. K. Unger, "Recommendations for the
characterization of porous solids", Pure and Applied Chemistry 66, 1739 (1994).
15.

K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J.
Rouquerol, and T. Siemieniewska, "Reporting physisorption data for gas solid
systems with special reference to the determination of surface-area and porosity",
Pure and Applied Chemistry 57, 603 (1985).

16.

S. J. Gregg and K. S. W. Sing, Adsorption, Surface Area, & Porosity, 2 ed.
(Academic Press, New York, 1982).

17.

T. Soboleva, X. S. Zhao, K. Mallek, Z. Xie, T. Navessin, and S. Holdcroft, "On the
Micro-, Meso- and Macroporous Structures of Polymer Electrolyte Membrane Fuel
Cell Catalyst Layers", Acs Applied Materials & Interfaces 2, 375 (2010).

18.

T. Iwazaki, R. Obinata, W. Sugimoto, and Y. Takasu, "High oxygen-reduction
activity of silk-derived activated carbon", Electrochemistry Communications 11,
376 (2009).

19.

N. P. Subramanian, X. G. Li, V. Nallathambi, S. P. Kumaraguru, H. ColonMercado, G. Wu, J. W. Lee, and B. N. Popov, "Nitrogen-modified carbon-based
catalysts for oxygen reduction reaction in polymer electrolyte membrane fuel cells",
Journal of Power Sources 188, 38 (2009).

20.

V. Nallathambi, J. W. Lee, S. P. Kumaraguru, G. Wu, and B. N. Popov,
"Development of high performance carbon composite catalyst for oxygen reduction
reaction in PEM Proton Exchange Membrane fuel cells", Journal of Power Sources
183, 34 (2008).

21.

G. Wu, C. M. Johnston, N. H. Mack, K. Artyushkova, M. Ferrandon, M. Nelson, J.
S. Lezama-Pacheco, S. D. Conradson, K. L. More, D. J. Myers, and P. Zelenay,
"Synthesis-structure-performance correlation for polyaniline-Me-C non-precious
metal cathode catalysts for oxygen reduction in fuel cells", Journal of Materials
Chemistry 21, 11392 (2011).

130

22.

F. Jaouen, J. Herranz, M. Lefevre, J. P. Dodelet, U. I. Kramm, I. Herrmann, P.
Bogdanoff, J. Maruyama, T. Nagaoka, A. Garsuch, J. R. Dahn, T. Olson, S.
Pylypenko, P. Atanassov, and E. A. Ustinov, "Cross-Laboratory Experimental
Study of Non-Noble-Metal Electrocatalysts for the Oxygen Reduction Reaction",
Acs Applied Materials & Interfaces 1, 1623 (2009).

23.

P. Zelenay, “Advanced Cathode Catalysts,” in United States DOE Hydrogen
Program Annual Progress Report, (2010).

131

5. ROLE OF TRANSITION METAL IN OXYGEN
REDUCTION CATALYSIS AND AN INSIGHT INTO
CATALYTIC ACIVE SITE STRUCTURE*

5.1 Abstract
The effect of transition metal towards generating active sites in Metal-NitrogenCarbon catalysts has been studied and an optimal nominal Iron content close to 0.75 %
was established for melamine based Fe-N-C catalysts. Electrochemical studies performed
in the presence of poisoning anions such as cyanide in alkaline environment indicated a
25% decrease in oxygen reduction activity, suggesting that Fe based sites are part of
active sites and participate in catalyzing oxygen reduction reaction. In-situ Fe
Kα XANES spectra of melamine-based electrocatalysts measured under oxygen saturated
0.1 M HClO4 indicated the absence of metallic Fe particles on the surface. Comparison of
the adsorption edge and white line features reveal an oxidation state of +2 for Fe in the
catalysts for potentials up to 0.7 V (region of mixed electrode potential control due to
kinetics and mass transport) and an oxidation state of +3 for potentials over 0.7 V (purely
kinetically controlled). In-situ EXAFS analysis of the catalysts suggested the active site
structure to be predominantly Fe metal co-ordinated to 4 nitrogen atoms with the
possibility of presence of Fe-N2 sites.

*In collaboration with North-Eastern University, XAFS Experiments were done at NSLS
Facility at Brookhaven National Lab. The cells were assembled and tested by Matt
Trahan of Northeastern University.
132

5.2 Introduction
In Chapter 2, a high-pressure pyrolysis approach for synthesizing active MNC
catalysts was addressed. In chapters 3 and 4, the significance of nitrogen precursor
properties towards improving oxygen reduction activity was analyzed. It has been
demonstrated in chapter 3 that by increasing the N/C ratio of the nitrogen precursor, the
accessible active site density increased due to reduced carbon deposition in the pores of
the carbon support during pyrolysis. Melamine (N/C = 2.0) based MNC catalysts
-3

exhibited the highest ORR activity (~16 A cm
2

1,2

(~1100 m /g).

at 0.8 V) and very high surface area

In chapter 4, to completely eliminate carbon deposition effects from N

precursors, NH3 generating precursors such as urea and ammonium carbamate were used
-3

and high ORR activity (~10-14 A cm

at 0.8 V) and increased surface area (~1400 –

2

2000 m /g) have been achieved. Through XPS characterization, the amount of various
nitrogen groups present the catalyst was obtained and a correlation between pyridinic and
pyridinic-Fe sites and oxygen reduction activity has been established, signifying the role
of nitrogen in catalytic active sites. In this chapter, the role of transition metal towards
generating active sites will be explored; through in-situ X-ray Absorption Spectroscopy
(XAS) analysis, insight into Fe-nitrogen co-ordination will be gained.
The activity of pyrolyzed non-precious MNC catalysts has seen tremendous
1-4

improvement in the recent years,

however the catalytic active site structure of such

catalysts and the mechanism of oxygen reduction on these sites still remains
undetermined. As described in the previous chapters, metal cations co-ordinated to two
pyridinic nitrogens at the edges of graphene sheets have been widely cited as the active
133

5

center. Recently, Ozaki et al. have elucidated the role of iron in the pyrolysis of iron
pthalocyanine and phenolic resin to form carbon based cathode catalysts. Through XPS
and FTIR characterization, their model confirms that during carbonization, iron species
promote the edge exposure of carbon and increase the nitrogen content through ironnitrile co-ordination. But, if the pyrolysis temperature was too high, the nitrogen content
of the resulting catalysts reduces due to the formation of iron nano-particles that enhance
6,7

the elimination of nitrogen at elevated temperatures.

However, there have been

conflicting reports claiming that Metal-N coordination is no longer detected in catalysts
8-10

pyrolyzed over 700°C.

Carbon-Nitrogen catalysts synthesized without any metal

additive have also been reported; their oxygen reduction activity was significantly lower
11-15

compared to MNC catalysts.

Carbon materials are widely known for their inactivity

towards oxygen reduction due to the absence of d-orbital electrons. However, edge
defects in graphene have been shown to possess density of states near the Fermi level;
therefore these edge defects may play a similar role as that of d-bands of transition metal
12,16

for oxygen reduction.

More over, the presence of edge defects on carbon promotes

the density of N atoms on these sites during pyrolysis, suggesting that C-N co-ordination
may be responsible for oxygen reduction activity.
X-ray absorption spectroscopy (XAS) has been used as a powerful tool in recent
years to study fuel cell electro-catalysts to gain information regarding the oxidation state
of metal in the catalyst, their local co-ordination number and identity of the adsorbing
atom. The ability to perform XAS measurements in-situ under environments that closely
mimic a working fuel cell makes them more attractive over other characterization

134

techniques such as XPS, SEM and EDAX. Using Extended X-ray Absorption
17

Fluorescence Spectroscopy (EXAFS) analysis, Titov et al.

have established pyridinic

Fe-N4 sites to be the active site for oxygen reduction in their aligned carbon nanotube
18

catalysts doped with iron and nitrogen. Choi et al.

19

and Bron et al.

also individually

confirmed the presence of Fe-N4 sites in their carbon supported iron-phtalocyanine and
19

iron-phenanthroline catalysts pyrolyzed at 800°C. Bron et al.

observed the formation of

Fe2O3 structures for catalysts pyrolyzed beyond 900°C and reported a decline in oxygen
20

reduction activity. Similarly for cobalt based non-precious metal catalysts, Joyner et al.
21

and Wingerden et al.

detected the presence of Co-N4 structures in Co-porphyrin
8

catalysts pyrolyzed at 850°C. Through XANES analysis, Alves et a. showed that Co-N4
structure has a square planar D4h symmetry of Co and for catalysts pyrolyzed above
700°C, a loss the square planar configuration was observed and the XANES spectra
resembled that of a pure Co metal. Fourier transforms of the EXAFS region also
indicated the formation of Co particles after pyrolysis, as a new peak appeared at 2.2 Å,
corresponding to Co-Co bond.
22

Niwa et al.

established through the pre-edge of N1s XAS spectra that higher

amounts of graphite-like nitrogen contributed to increased oxygen reduction activity.
23

Ziegelbauer et al.

applied ∆µ analysis, an X-ray absorption near-edge structure

difference technique to Co K edge EXAFS data and concluded that the specific
geometrical adsorption of molecular oxygen with respect to the plane of Co-N4 moieties
influence the oxygen reduction pathway. They observed that majority of the Co-N4 active
135

sites in the catalysts broke down to Co-N2 for pyrolysis temperatures 800°C and beyond.
The objective of the present study is to address whether Fe species are part of the
catalytic center or merely promote the formation of catalytic actives site without
participation in oxygen reduction and to determine the structure of active reaction sites.
5.2.1

X-ray Absorption Spectroscopy – a brief overview
24,25

XAS is a local probe of the electronic and geometric structure of materials.

XAS

measures the change in absorbance or fluorescence of the sample as the energy of
incident X-rays is scanned through the absorption region of a specific element.
Availability of high intensity synchrotron sources X-ray radiation facilitates XAS to be
performed on electro-catalysts in-situ. Monochromatic X-rays of energy hν, where h is
-34

the Planck’s constant (6.63 x 10

2

m kg /s) and ν is the frequency (Hz), irradiate the

sample and an electron is excited to an unoccupied state of the atom. When the energy of
the incident x-ray is sufficient to excite a core- level electron of the absorbing atom to an
unoccupied atomic or molecular orbital, an absorption edge results.
XAFS comprises of two distinct regions: The region closer to the absorption edge is
called X-ray absorption near edge spectra (XANES) and typically extends up to 30 eV
above the absorption edge. XANES is characteristic of the local symmetry and electronic
structure of the absorbing atoms. The oxidation state of the absorber can also be obtained
from the position of the absorption edge.
At energies well above the absorption edge, the energy of the incident photon is
sufficient to excite a core electron of the absorber into the continuum, producing a
photoelectron. The ejected photo-electron may be approximated by a spherical wave,
136

which is backscattered by neighboring atoms. The interference between the outgoing
forward scattered or ejected photoelectron wave gives rise to an oscillation in the
absorbance as a function of incident photon. These oscillations may extend up to 1000 eV
above the absorption edge and are called the extended X-ray absorption fine structure
(EXAFS); Analysis of the EXAFS provides information regarding the identity, number of
neighboring atoms and their bond-length. The EXAFS fine structure function χ(E) is
defined as

χ (E) =

µ ( E ) − µ0 ( E )
∆µ 0 ( E )

(5.1)

where µ(E) represents the measured absorption co-efficient, µ0(E) is a smooth
background function representing the absorption of an individual atom and ∆µ0(E)
represents the jump in absorption µ(E) at the adsorption edge or binding energy E0.

Data analysis of EXAFS becomes convenient when considering the wave behavior of
the emitted photoelectron rather than the energy of the incident X-ray. This is
accomplished by converting the X-ray energy ‘E’ into ‘k’ space, where k is the wave
number of the photo-electron and is defined as

k=

2m( E − E o )

(5.2)

h2

where E0 is the absorption edge energy and m is the electron mass.

EXAFS is then defined as χ(k), the oscillations as a function of photo-electron wave
number. Peak frequencies in χ(k) correspond to different near-neighbor co-ordination
137

shells and is modeled according the following equation

χ (k ) = ∑

N j f j ( k )e

−2k 2σ 2
j

2
kR j

j

sin  2kR j + δ j ( k ) 



(5.3)

where f(k) and δ(k) are associated with the scattering properties of the neighboring atoms,
N is the number of neighboring atoms, R is the distance of the neighboring atom from the
2

2

excited atom, and σ represents the disorder in the neighboring atom. N, R and σ are
determined from the EXAFS equation, knowing the scattering amplitude f(k) and phase
24-26

shift δ(k).

5.3 Catalyst Synthesis
®

Iron (II) acetate, melamine, Nafion solution (1100 EW, 5 wt. %), Sodium hydroxide,
potassium cyanide and sulfuric acid (ACS grade) were obtained from Alfa Aesar (Ward
®

Hill, MA). Ketjenblack 600JD carbon black was obtained from Akzo Nobel (Chicago,
IL). Pressurized oxygen cylinders were obtained from Airgas (Lansing, MI). All
materials were used as received.
Catalysts based on melamine- based precursors were synthesized in a similar fashion
described in Chapter 2 and 3. To remove excess iron in the catalysts post-pyrolysis, the
catalysts were exposed to aqueous 1 N sulfuric acid at 80°C for 5 hr, filtered and washed
several times with Millipore water. The sample was then oven-dried overnight at 80°C
before use.

138

5.4 Experimental
Electrochemical characterization was conducted using a glassy carbon rotating disk
2

electrode (RDE, 0.2 cm area) in aqueous 1 N sulfuric acid at 40°C, as described
previously in Chapter 2 and 3. Electrochemical poisoning experiments were performed in
O2 saturated 0.1M NaOH containing 10 and 20 mM KCN. Data were collected at 1200
rpm and at a scan rate of 0.5 mV s

-1

. In

order to determine the bulk metal content of the

catalysts after pyrolysis and acid-leaching, a range of catalyst and precursor carbon
samples was submitted for ICP-MS analysis for the determination of iron. Samples were
digested in closed vessels in aqua regia for 24 h, evaporated to near dryness, dissolved in
2% nitric acid filtered through a 0.45 µm filter before analysis.
5.4.1

X-ray Absorption Spectroscopy

All XAS measurements were carried out in a flow-through electrochemical cell,
designed by Northeastern University as shown in Fig. 5.1. Unleached melamine based
catalysts was chosen as model MNC catalysts and the catalyst ink was prepared by
sonicating calculated amounts of the catalyst with 30% nafion ionomer in a solution of
ethanol to obtain a 1:2 nafion - catalyst weight ratio. The resulting catalyst ink was dropcasted onto a carbon cloth, (PANEX 30 obtained from ZOLTEK) until a catalyst loading
2

of 1 mg/cm was achieved. This serves as the working electrode (sample size - 3cm x
1cm), a bare carbon cloth (PANEX 30) served as the counter electrode and RHE was
used as the reference electrode. The working electrode was soaked in a solution of 0.1 M
HClO4 for ca. 1h prior to assembly in the sample holder. The samples were assembled
onto a Teflon sample holder and gold current collectors were used. 0.1 M HClO4 was
139

-1

used as the electrolyte and a flow rate of 3 ml min

of de-aerated or O2 saturated

electrolyte was purged through the cell through a peristaltic pump. The cell potential was
controlled using an AUTOLAB potentiostat (Metroohm USA, Inc.).
All measurements were carried out at beamline X3-B of the National Synchrotron
Light Source (NSLS, Brookhaven National Laboratory, Upton NY) under near-ultra high
vacuum conditions. Light source from synchrotron consists of light of varying
wavelengths across the spectrum and is typically called “white light”. By tuning the
incident angle (θ) of the white light on the monochromator, which operates on Bragg’s
law (2d sinθ = nλ), energy tunable source of high intensity X-rays are achieved. Fe Kα
edge (E0 = 7112 eV) spectra were collected (-25 eV to 200 eV with respect to E0) using a
double crystal Si (111) monochromator with a Bragg angle range of 10° – 35° and
located 16.5 m from the source (bending magnet). Due to the low concentration of Fe,
XAS data were collected in fluorescence mode using a Canberra 13-element solid state
Ge detector. Experiments were carried out in Ar and O2 saturated electrolytes at open
circuit potentials, 0.9 V, 0.7 V, 0.5 V and 0.3 V vs. RHE and a minimum of two scans
were measured at each potential. Three gas ionization detectors for the incident (I0),
transmission (It) and reference (It) [‘I’ represents the intensity of X-rays], chambers were
used. The cell was placed between an incident beam detector (a 15 cm in length gas
ionization detector filled with N2) and a transmittent beam detector (15 cm also N2). A Fe
reference foil was placed between transmittent and the third reference detector Iref (7 cm,
N2 filled) for the purpose of energy calibration and to correct for any beam drift that may
occur during the 12 hour life cycle of the beam. The ionization detector amplifiers were
140

Figure 5.1 Experimental set up for in-situ XAS measurements at X-3B. Courtesy: NE
University.

141

6

8

23

typically adjusted to a gain value of 10 – 10 V/A for optimal signal to noise (S/N).

Data analysis was performed using ATHENA and ARTEMIS software. Background
subtraction was performed using ATHENA using AUTOBCK algorithm followed by
normalization. The background subtraction steps include calibration of edge energy
(7112 eV for Fe Kα), alignment to standard reference scan to account for any drift in
beam energy and a post-edge normalization procedure to the aligned scans through a
cubic spline function. The spline function normalizes the oscillations over a specific
energy range between 25 to 250 eV with respect to E0. Typically, a Hanning type
-1

window for k-ranges between 2.5 to 12.5 Å was used for fitting the data.

5.5 Results and Discussions
5.5.1

Optimization of nominal Fe content

As explained in previous chapters, the DOE target for oxygen reduction activity of
-3

non-precious metal catalysts for the year 2015 has been set at 300 A cm at 0.8 V vs.
RHE, 80°C, and under O2 absolute pressures of 1 bar.

27,28

The targeted oxygen reduction
-3

activity is a mathematical product of the site density (number of active sites cm ), turnover frequency (number of electrons per active site per second at 0.8 V vs. RHE) and
electric charge of a single electron. The most straight forward approach to achieving the
2015 DOE target is to establish the active site structure and maximize the density of such
27,28

sites.

In our initial studies with pyridine-based MNC catalysts synthesized without Fe were
inactive for oxygen reduction (Fig. 2.6d), indicating the significance of transition metal in
142

promoting ORR. Through XPS analysis a correlation between current density at 0.6 V /
SHE and metal-pyridinic N co-ordination was established, though the highest
concentration achieved was very low (ca. 0.24 at%). Assuming that metal-pyridinic N is
the active site as established through XPS analysis, the nominal transition metal
concentration and nitrogen content should be increased in order to increase the site
density.
To optimize the optimal iron content, Melamine (6.3 wt%) was used as the model N
precursor and a series of catalysts were synthesized by varying the nominal iron metal
content, and the oxygen reduction activity was measured using rotating disc and rotating
ring-disc electrode studies, as detailed in Chapter 2, section 2.3.2. Figure 5.2a shows
mass-transfer corrected Tafel plots for catalysts synthesized using various metal loadings.
The activity increases when metal content was increased from 0.33 to 0.75 wt%, beyond
which the activity saturates. The Tafel slopes calculated between 0.9 and 0.75 V vs. RHE
varied between 60-70 mV/decade and increased with increasing metal content as shown
in Table 5.1. Figure 5.2b shows the observed kinetic current density at 0.8 V and bulk N
content for melamine-based catalysts as a function of nominal iron content. The kinetic
current maximizes at around 0.75 wt% Fe and the bulk N content remained constant at
around 4 wt% for catalysts with nominal Fe loading beyond 0.33 wt%. Only a fraction of
Fe atoms could be involved in the formation of active sites, and this might explain the
saturation beyond 0.75 wt % Fe. This hypothesis may also be confirmed from the fact
that crystalline Fe phases were detected in samples with higher metal loadings through Xray diffraction measurements.

143

Figure 5.2 a) Mass transfer corrected Tafel plots for Melamine based Fe-N-C catalysts
synthesized using varying nominal Fe loadings. b) Kinetic current and observed bulk N content
as function of nominal Fe loading. c) Kinetic current and observed percentage peroxide content
as function of nominal Fe loading.
-1

Conditions: O2-saturated, 1 N aqueous sulfuric acid, 40° Scan rate 0.5 mV s , 1200 rpm,
C.
Nominal 6.3 wt% N loading.

144

Figure 5.2c depicts the change in H2O2 production at 0.8 V vs. RHE with metal
content as observed through RRDE studies. A slight decrease in H2O2 yield was observed
by increasing the metal content to 0.75 wt % possibly due to the increase in reaction sites,
beyond which the peroxide yield saturates as well. This is also an indication that the
nature of the active sites hasn’t significantly changed with increased Fe loading beyond
0.75 wt%. Also, it is believed that, if more active reaction sites are present in the catalyst,
O2 undergoes a 2+2 electron reduction mechanism, where the peroxide formed due to 2e

-

-

transfer at one active site can further undergo another 2e reduction at another site, which
[8]

explains the decrease in peroxide yield with increasing Fe loading. Figure 5.3 shows the
X-ray diffraction patterns obtained for a series of catalysts based on melamine precursors.
A broad peak due to amorphous carbon (002) can be observed at around 25° for all the
catalysts. Diffraction peaks associated with iron metal and cementite (Fe3C) appear for
catalysts with more than 1.5% Fe. The formation of iron particles and iron carbides
promote the elimination of N as N2 gas and will thus reduce the nitrogen content of the
6,7

catalysts and hence reduce the oxygen reduction activity . Also, this supports the claim
that only part of the Fe atoms were involved in formation of active reaction sites.
In order to calculate the turn-over frequency, the bulk Fe content of the catalysts
should be known. Bulk Fe content determined through ICP-MS analysis for catalysts
synthesized using melamine precursor and the optimized Fe loading of 0.75 wt% are
shown in Table 5.2, and the results indicate that pyrolyzed catalysts contain close to
1 wt% Fe, well above what has been detected by XPS studies (~0.1%).

145

29

It should be

Table 5.1 Summary of electrochemical and structural parameters obtained for Melamine
based Fe-N-C catalysts synthesized using varying nominal Fe loading.
-3

Nominal metal
Content / wt%

Ik / A cm
(@ 0.8 V)

Bulk N / wt %

0.33

2.1 ± 0.2

0.75

Tafel slope
(V dec )

H2O2 / %
(@ 0.8 V)

2.57

0.063 ± 0.01

10.5 ± 0.4

6.3 ± 0.2

4.03

0.061 ± 0.02

6.3 ± 0.3

1.33

4.5 ± 0.1

4.21

0.071 ± 0.05

6.7 ± 0.3

1.50

5.2 ± 0.2

3.69

0.075 ± 0.024

5.9 ± 0.5

146

-1

noted that the bare Ketjen carbon support also contains trace iron, ~0.24% wt%. Similarly,
a catalyst sample that was leached in sulfuric acid at 80°C for five hours still retained
80% of its original iron content. Such leaching steps are typically used to remove iron,
but in this case are ineffective. These results indicate that substantial iron content remains
in the catalyst material post-fabrication. The fact that most Fe is inaccessible via XPS
measurements suggests that this iron is buried under other materials and is not within the
1-5 nm penetration depth of XPS. The extent to which such iron may participate in
catalysis is explained in the following sections.
Assuming all iron determined (0.79 wt %) through ICP-bulk analysis participated in
the formation of active reaction sites, the turn-over frequency of the MNC catalyst can be
calculated using the following relation.

C N ρ
I k (0.8V ) = Fe a cat TOF (0.8V )e −1 = SD TOF(0.8V )e −1
100 M Fe

(5.4)

-3

where Ik is the volumetric kinetic current density at 0.8 V (A cm ), CFe is the bulk Fe
23

content (wt%). Na is the Avagadro number (6.023 x10
-1

-

-1

mol ), MFe is the molar mass of

Fe (55.8 g mol ), e is the charge of the electron (1.69 x10

-19

C) and SD is the volumetric

-3

-3

site density (Sites cm ), ρcat is the density of the porous catalyst (0.4 g cm ). The site
density calculated using the iron content of 0.79 wt % amounted to 0.34 x 10

20

-3

sites cm .

-3

Using the RDE volumetric current density of 2.4 A cm at 0.8 V, the turn-over frequency
calculated was close to 0.44 e
-3

Chapter 3) of 15 A cm

-1

-1

site s

-1

and MEA volumetric current density (Ref:

at 0.8 V (as measured in a single fuel cell), the turn-over
-1

-1 -1
.

frequency was calculated to be 2.7 e site s

147

However, to achieve the 2015 target of

Figure 5.3 XRD patterns of catalysts synthesized by varying nominal Fe content, N
precursor-Melamine.

148

Table 5.2 ICP-MS characterization of MNC catalysts and carbon precursors
Fe content
(wt%)

Materials
Ketjen Black Precursor

0.24

Melamine Precursor

0.16

Fe/Melamine/KB Catalyst

0.99

Fe/Melamine/KB Catalyst (acid
leached)

0.79

149

-3

300 A cm , if the turn-over frequency is maintained at 2.7 e
20

-1

-1 -1

site s

and the site

-3

density has to be increased to 7 x 10 sites cm .

5.5.2

Metal – Centered oxygen reduction activity
30

Recently, Thorum et al.

examined the oxygen reduction activity of carbon

supported transition metal macrocycles in the presence of small-molecule poisons such
ascyanide. Their experiments established that such catalysts showed a reduced oxygen
reduction activity in the presence of cyanide, suggesting that active sites include a
transition metal. We carried out similar poisoning studies on MNC catalysts, where the
concentration of transition metal is ~⅕ that of metal macrocycles. Polarization curves
recorded in oxygen-saturated 0.1 M NaOH in the presence and absence of cyanide anions
are shown in Fig. 5.4. A 25% decrease in oxygen reduction kinetic activity (0.8 V/RHE)
is observed with the introduction of 10 mM KCN. Cyanide is known to complex with
iron and hence impedes access by oxygen.

31

If iron participates in catalysis, such

poisoning would result in decreased oxygen reduction activity. The observed 25%
reduction in activity provides additional evidence that transition metals may participate in
oxygen reduction.
However, the exact structure of the active site still needs to be established. Through
XANES (Near edge X-ray absorption spectroscopy), the co-ordination geometry around
Fe metal and via. EXAFS (Extended X-ray absorption fine structure spectroscopy), the
Fe-C and Fe-N distances between each carbon shell can be assessed.

150

Figure 5.4 Polarization curves of melamine – based Fe-N-C MNC catalysts: O2-saturated,
-1

0.1 M aqueous sodium hydroxide, 40° Scan rate 0.5 mV s , 1200 rpm, Nominal 0.75
C.
wt% Fe loading.

151

5.6 X-ray Absorption Spectroscopy
5.6.1

EXAFS data analysis

The raw data obtained during XAS measurements has to be reduced first to remove
background signals, the zero point of energy E0 has to be identified and then the data is
normalized and analyzed using the XAFS equation. The individual steps of data
reduction were performed using ATHENA software. Figure 5.5-a shows the
representative spectrum obtained in O2-saturated electrolyte at 0.9 V (OCP) and 5.5-b
and c show the pre-edge subtracted and normalized data. Background subtraction
removes both the variation in the absorbance with energy caused by the other atoms in
the sample (the near-linear variation seen before the edge between ca. 7000 and 7100 eV
in Figure 5.5-a) and the smooth variation in µ past the absorption edge (beyond 7100 eV),
corresponding to the absorption of the free atom. The zero point of the energy, E0 is
usually taken as the inflection point in the absorption edge, as shown later in Figure 5.5-a.
-1

This allows the energy of the incident photon χ(E), to be converted to k -space (Å

)

using equation 5.2. Subsequently, the absorption spectra µ(E) is normalized to read from
0 to 1, so that it represents the absorption of one X-ray. Normalization places the
measured spectrum on a per-absorber-atom basis, thereby taking into account the
concentration of the sample, and is division of the absorption data by the magnitude of
the edge step at 30 eV above the absorption edge. Once the EXAFS spectrum is isolated,
the data is be fitted to the EXAFS equation 5.1. Fourier transformation of the EXAFS
using equation 5.2 gives the radial structure function. Figure 5.5-c represents the resulting
2

k Fourier transformed spectrum. The EXAFS is usually oscillatory and quickly decays
with k. To emphasize the oscillations, χ(k) is often multiplied by a power of k.
152

32-34

3

Figure 5.6-a and b shows the Fourier transformed (k weighted) EXAFS spectra
obtained for Melamine based Fe-N-C catalysts under ex-situ conditions in the absence of
electrolyte and in-situ conditions in O2-saturated 0.1 M HClO4 at room temperature,
measured at open circuit potential (0.88 V vs. RHE). In both the Fourier transformed
spectra, a large feature can be observed at ~ 2 Å and the spectra obtained under ex-situ
conditions indicate an additional shoulder at around 2.2 Å. The peak intensities are a
combination of disorder in the neighboring distance σ2 and the co-ordination number N
explained in EXAFS equation 5.3.
Using FEFF input files obtained from published crystallographic parameters, the
spectra were fitted to first and second co-ordination shells. FEFF is a program for abinitio calculations of phase shifts and back-scattering amplitudes of XAFS and XANES
spectra for a cluster of atoms. The FEFFIT program fits the experimental EXAFS data to
the theoretical standards calculated using FEFF and includes multiple scattering pathways
during the fit. The EXAFS spectrum was fitted assuming Fe-Nx (x = 2,3,4) as the first coordination shell and Fe-Fe as the second coordination shell for data collected under exsitu conditions and Fe-O as the second coordination shell for the in-situ data. The results
of the fits for the EXAFS data (equation 5.3) collected under ex-situ and insitu conditions
at different electrochemical potentials are summarized in Table 5.2, where N represents
the co-ordination number, R represents the inter-atomic distance between the metal and
2

neighboring atom, E0 is the zero point of energy and σ represents the mean-square
displacement in R. The primary peak in R observed between 1 and 3 Å in the ex-situ data

153

Figure 5.5 a) XAFS data showing pre-edge line. b) Pre-edge subtracted and Normalized
2

spectra. c) K weighted Fourier transforms of XAFS data. Conditions: O2-saturated 0.1M HClO4,
0.9 V vs RHE in-situ Kα Edge XANES.

154

shown in Fig. 5.6-a arises due to Fe surrounded by 4 nitrogen atoms in square-planar
geometry and the shoulder at 2.2 Å arises from second shell interactions of trace Fe
present on the surface catalysts. It should be noted that the catalyst under study has not
been subjected to leaching in acid to remove excess metal before collecting the spectra.
Also, fitting of the ex-situ data showed absence of any adsorbed oxygen as expected.
Inset in Fig. 5.6-a shows that data fits with adsorbed oxygen included did not fit the
experimental spectra very well. The shoulder at 2.2 Å disappears in the spectra collected
under in-situ conditions measured in 0.1 M HClO4 as excess metal would not have
survived the acidic environment. It can be noted from Table 5.3 that the Fe-N bond
length increases with decreasing electrochemical potential. Also, a minor expansion in
the peak height can be seen in Fig. 5.6-b as compared to 5.6-a in order to accommodate
the O atom in axial position and hence a slight increase in bond-length of Fe-N from 1.97
to 2.08 Å can be witnessed as shown in Table 5.3, comparing ex-situ fits at OCP and insitu fits at OCP (0.88 V). From the table, it can be seen that the average Fe-O interatomic distances increase with increase in potential, owing to oxidation of the active sites
at those potentials.
Figure 5.7 a,b,c shows the EXAFS data fits with Fe-N co-ordination of 2, 3 and 4
respectively and the results of the fits are presented in table 5.4. It can be seen that the
spectra can be fitted well for co-ordination of 3 and 4, however, Titov et. al.

17

has

demonstrated using density functional theory that the structure and energetics of Fe-N2
and Fe-N4 are more stable in comparison with Fe-N3. Therefore, ignoring the possibility
of Fe-N3 sites in the catalyst, we observe that Fe-N4 co-ordination represents

155

Figure 5.6 Fe Kα non-phase corrected Fourier transformed EXAFS spectra and data fits of
melamine –based Fe-N-C catalysts under (a) ex-situ conditions (Inset shows fits with Fe-O
shell included)
vs. RHE).

(b) In-situ 0.1 M HClO4, room temperature, Open circuit potential (0.9 V

156

Table 5.3 Fe K-edge EXAFS fit results for Melamine based Fe-N-C catalysts under
ex-situ and in-situ (O2 saturated 0.1M HClO4, room temperature).
Potential /V

Shell

N

R /Å

E0/ eV

σ /Å

Fe-N

3.8

1.97

3.45

0.011

Fe-Fe

0.5

2.91

4.17

0.011

Fe-N

3.8

2.08

0.47

0.013

Fe-O

1.6

1.95

-2.92

0.010

Fe-N

3.8

2.09

0.52

0.013

Fe-O

1.8

1.93

-2.05

0.010

Fe-N

3.8

2.08

0.37

0.014

Fe-O

0.1

2.06

-3.32

0.010

Fe-N

3.8

2.19

2.32

0.015

Fe-O

1.9

1.99

-3.89

0.010

Fe-N

3.8

2.24

4.23

0.014

Fe-O

1.7

2.03

-3.02

0.010

Fe-N

3.8

2.28

3.12

0.013

Fe-O

0.8

2.03

-2.89

0.010

Reduced
χ2

16

Ex-situ

1.0

23

0.9

0.88 (OCP)

21

14

0.7

8

0.5

0.3
Error limits: N ± 10%, R ± 0.05 Å, σ ± 0.005Å

157

12

14

predominantly the nature of active site structure for melamine based Fe-N-C catalysts and
the co-ordination did not change under all potentials considered, which is consistent with
17,19,23

other literature reports.

A fit of adequate quality was not obtained for Fe-N2 and

that is reflected in the unrealistically high values of E0 as shown in Table 5.4, Column 4.
However, it should be noted these catalysts are likely to be highly heterogeneous and a
bulk technique such as XAS will average over all the heterogeneity. Hence it is difficult
to ascertain if the entire material follows Fe-N co-ordination of 4 and there is a possibility
of presence of Fe-N2 sites as well. Ziegelbauer et al. observed similar results for their Coporphrin catalysts and conclude that their catalyst should be considered as a mixture of
Co-N, Co-N2, Co-N3 and Co-N4 and true nature of the catalytic site somewhat remains
ambiguous. They reported that majority of the Co-N4 active sites in the catalysts broke
down to Co-N2 for pyrolysis temperatures 800°C and beyond.

23

XANES is a local atomic probe that gives information about the formal oxidation
state and co-ordination geometry of the metal center. In-situ Fe Kα XANES spectra of
melamine based electrocatalysts measured under Oxygen saturated 0.1 M HClO4 together
with Fe, Fe2O3, and Fe(III) pthalo-cyanin standards are shown in Fig. 5.7-a. Fe(III)
pthalo-cyanin has four nitrogen atoms in the square planar environment and a pre-edge
transition at 7112.5 eV (Fe 1s to 3d transition) is due to slight disruption in center of
symmetry due to oxygen ligand coordinated to the metal center. It can be seen that the
pre-edge features (~7112-7120 eV) differ significantly to that of Fe reference foil
suggesting the absence of metallic Fe particles on the surface. Even if traces of metallic
Fe particles were present in the surface of the catalysts, the highly acidic electrolyte
158

Figure 5.7a,b,c. Fe Kα non-phase corrected Fourier transformed EXAFS spectra and data fits
of melamine –based Fe-N-C catalysts for Fe-N coordination of 2, 3 and 4. Conditions: 0.1 M
HClO4, room temperature, Open circuit potential (0.9 V vs. RHE).

159

Table 5.4 Fe K-edge EXAFS fit results for Melamine based Fe-N-C catalysts under in-situ (O2
saturated 0.1M HClO4, room temperature) for varying Fe-N co-ordination
Shell

N

R /Å

E0/ eV

σ /Å

Fe-N

3.8

2.09

0.52

0.013

Fe-O

1.8

1.93

-2.05

0.010

Fe-N

3.0

2.08

1.81

0.014

Fe-O

2.3

2.06

-7.91

0.010

Fe-N

1.8

2.19

14.0

0.015

Fe-O

1.9

1.99

-36.2

0.010

Reduced
χ2

Error limits: N ± 10%, R ± 0.05 Å, σ ± 0.005 Å

160

23

21

14

Figure 5.8 a) In situ Fe Kα XANES spectra of Melamine based catalysts measured at
different potentials under O2 together with standards b) First derivative analysis of the
Kα XANES spectra. Conditions: 0.1M HClO4, room temperature.

161

would have leached off the particles from their surface.
The effect of potential on the XANES spectra can be seen in Fig. 5.7a. With increase
in potential, the adsorption energy edge (~7122 - 7130 eV) shifts to higher values,
indicating oxidation of Fe at high potentials. However, comparison of the adsorption edge
and white line intensity or edge inflection features (~7125 - 7135 eV) to the standards
shown in Fig 5.8-a reveal an oxidation state of +2 for Fe in the catalysts for potentials
upto 0.7 V (region of mixed electrode potential due to kinetics and mass transport) and an
oxidation state of +3 for potentials over 0.7V (purely kinetically controlled). Also, a
23

change in edge inflection is observed with increasing electrode potential.

The white line

intensity is believed to arise from 1s to 4p (7117 eV) electron transitions and is a measure
35,36

of the partial density of the final states above the Fermi level.

The white line intensity

levels off at around 0.9 V signifying that a full component of an adsorbed oxygen
35,36

molecule is present at 1 V.

The changes in pre-edge and adsorption edges can be

more clearly visualized as their derivatives as shown in Fig. 5.7b. Comparing the XANES
region of the catalysts with the standards indicate that the metal center in the catalyst in
terms of its oxidation state is reminiscent of reduced Fe

2+

3+

at 0.3 V and oxidized Fe

at

0.9V and has an O atom in the axial position.
5.7 Conclusions
The effect of transition metal towards generating active sites in Metal-NitrogenCarbon catalysts has been studied and an optimal nominal Iron content of ca. 0.75 wt %
was established for melamine and bi-pyridine based catalysts. Electrochemical studies in
the presence of anions such as cyanide in alkaline environment indicated a reduction in
162

oxygen reduction activity, suggesting that Fe based sites are part of the active sites and
participate in oxygen reduction reaction. In-situ X-ray Absorption Spectroscopy (XAS)
analysis on melamine based catalysts under O2 saturated 0.1 M HClO4 indicated the
absence of metallic Fe particles on the surface. Comparison of the adsorption edge and
white line features of XANES reveal an oxidation state of +2 for Fe in the catalysts for
potentials up to 0.7 V (region of mixed electrode potential due to kinetics and mass
transport) and an oxidation state of +3 for potentials over 0.7 V (purely kinetically
controlled). In-situ EXAFS analysis of the catalysts provided an insight into the active
site to be predominantly Fe metal co-ordinated to 4 nitrogen atoms with the possibility of
presence of Fe-N2 sites. These results suggest that Fe forms part of the catalytic active
site and participates in oxygen reduction reaction mechanism.

163

REFERENCES

164

5.8 References
1.

F. Jaouen, E. Proietti, M. Lefevre, R. Chenitz, J.-P. Dodelet, G. Wu, H. T. Chung,
C. M. Johnston, and P. Zelenay, "Recent advances in non-precious metal catalysis
for oxygen-reduction reaction in polymer electrolyte fuel cells", Energy &
Environmental Science 4, 114 (2011).

2.

E. Proietti, F. Jaouen, M. Lefevre, N. Larouche, J. Tian, J. Herranz, and J. P.
Dodelet, "Iron-based cathode catalyst with enhanced power density in polymer
electrolyte membrane fuel cells", Nature Communications 2 (2011).

3.

M. Lefevre, E. Proietti, F. Jaouen, and J.-P. Dodelet, "Iron-Based Catalysts with
Improved Oxygen Reduction Activity in Polymer Electrolyte Fuel Cells", Science
324, 71 (2009).

4.

G. Wu, K. L. More, C. M. Johnston, and P. Zelenay, "High-Performance
Electrocatalysts for Oxygen Reduction Derived from Polyaniline, Iron, and
Cobalt", Science 332, 443 (2011).

5.

M. Lefevre, J. P. Dodelet, and P. Bertrand, "Molecular oxygen reduction in PEM
fuel cell conditions: ToF-SIMS analysis of Co-based electrocatalysts", Journal of
Physical Chemistry B 109, 16718 (2005).

6.

Y. Nabae, S. Moriya, K. Matsubayashi, S. M. Lyth, M. Malon, L. B. Wu, N. M.
Islam, Y. Koshigoe, S. Kuroki, M. A. Kakimoto, S. Miyata, and J. Ozaki, "The
role of Fe species in the pyrolysis of Fe phthalocyanine and phenolic resin for
preparation of carbon-based cathode catalysts", Carbon 48, 2613 (2010).

7.

J. Ozaki, N. Kimura, T. Anahara, and A. Oya, "Preparation and oxygen reduction
activity of BN-doped carbons", Carbon 45, 1847 (2007).

8.

M. C. M. Alves, J. P. Dodelet, D. Guay, M. Ladouceur, and G. Tourillon, "Origin
of the electrocatalytic properties for O2 reduction of some heat-treated
polyacrylonitrile and phthalocyanine cobalt compounds adsorbed on carbon-black
as probed by electrochemistry and x-ray absorption-spectroscopy", Journal of
Physical Chemistry 96, 10898 (1992).

9.

J. A. R. Van Veen, H. A. Colijn, and J. F. van Baar, "On the effect of a heat
treatment on the structure of carbon-supported metalloporphyrins and
phthalocyanines", Electrochimica Acta 33, 801 (1988).

10.

E. Yeager, "Electrocatalysts for O2 reduction", Electrochimica Acta 29, 1527
(1984).

165

11.

Z. Chen, D. Higgins, and Z. W. Chen, "Nitrogen doped carbon nanotubes and
their impact on the oxygen reduction reaction in fuel cells", Carbon 48, 3057
(2010).

12.

T. Enoki, Y. Kobayashi, and K. I. Fukui, "Electronic structures of graphene edges
and nanographene", International Reviews in Physical Chemistry 26, 609 (2007).

13.

T. Iwazaki, R. Obinata, W. Sugimoto, and Y. Takasu, "High oxygen-reduction
activity of silk-derived activated carbon", Electrochemistry Communications 11,
376 (2009).

14.

S. M. Lyth, Y. Nabae, S. Moriya, S. Kuroki, M. Kakimoto, J. Ozaki, and S.
Miyata, "Carbon Nitride as a Nonprecious Catalyst for Electrochemical Oxygen
Reduction", Journal of Physical Chemistry C 113, 20148 (2009).

15.

N. P. Subramanian, X. G. Li, V. Nallathambi, S. P. Kumaraguru, H. ColonMercado, G. Wu, J. W. Lee, and B. N. Popov, "Nitrogen-modified carbon-based
catalysts for oxygen reduction reaction in polymer electrolyte membrane fuel
cells", Journal of Power Sources 188, 38 (2009).

16.

S. Entani, S. Ikeda, M. Kiguchi, K. Saiki, G. Yoshikawa, I. Nakai, H. Kondoh,
and T. Ohta, "Growth of nanographite on Pt(111) and its edge state", Applied
Physics Letters 88 (2006).

17.

A. Titov, P. Zapol, P. Kral, D. J. Liu, H. Iddir, K. Baishya, and L. A. Curtiss,
"Catalytic Fe-xN Sites in Carbon Nanotubes", Journal of Physical Chemistry C
113, 21629 (2009).

18.

H. J. Choi, G. Kwag, and S. Kim, "Electrochemical and XAFS investigation of
nitrite reduction by heat-treated mu-oxo derivative of iron phthalocyanine
supported on high area carbon", Journal of Electroanalytical Chemistry 508, 105
(2001).

19.

M. Bron, J. Radnik, M. Fieber-Erdmann, P. Bogdanoff, and S. Fiechter, "EXAFS,
XPS and electrochemical studies on oxygen reduction catalysts obtained by heat
treatment of iron phenanthroline complexes supported on high surface area carbon
black", Journal of Electroanalytical Chemistry 535, 113 (2002).

20.

R. W. Joyner, J. A. R. Vanveen, and W. M. H. Sachtler, "Extended X-ray
absorption fine-structure (EXAFS) study of cobalt-porphyrin catalysts supported
on active-carbon", Journal of the Chemical Society-Faraday Transactions I 78,
1021 (1982).

21.

B. Van Wingerden, J. A. R. Van Veen, and C. T. J. Mensch, "An extended X-ray
absorption fine structure study of heat-treated cobalt porphyrin catalysts

166

supported on active carbon", Journal of the Chemical Society, Faraday
Transactions 1: Physical Chemistry in Condensed Phases 84, 65 (1988).
22.

H. Niwa, K. Horiba, Y. Harada, M. Oshima, T. Ikeda, K. Terakura, J. Ozaki, and
S. Miyata, "X-ray absorption analysis of nitrogen contribution to oxygen
reduction reaction in carbon alloy cathode catalysts for polymer electrolyte fuel
cells", Journal of Power Sources 187, 93 (2009).

23.

J. M. Ziegelbauer, T. S. Olson, S. Pylypenko, F. Alamgir, C. Jaye, P. Atanassov,
and S. Mukerjee, "Direct spectroscopic observation of the structural origin of
peroxide generation from co-based pyrolyzed porphyrins for ORR applications",
Journal of Physical Chemistry C 112, 8839 (2008).

24.

D. C. Koningsberger, B. L. Mojet, G. E. van Dorssen, and D. E. Ramaker, "XAFS
spectroscopy; fundamental principles and data analysis", Topics in Catalysis 10,
143 (2000).

25.

A. E. Russell and A. Rose, "X-ray absorption Spectroscopy of low temperature
fuel cell catalysts", Chemical Reviews 104, 4613 (2004).

26.

B. K. Teo, EXAFS: basic principles and data analysis (Springer-Verlag, 1986).

27.

“Hydrogen, Fuel Cells & Infrastructure Technologies Program Multi-Year
Research, Development, and Demonstration Plan,” (2007).

28.

H. A. Gasteiger, S. S. Kocha, B. Sompalli, and F. T. Wagner, "Activity
benchmarks and requirements for Pt, Pt-alloy, and non-Pt oxygen reduction
catalysts for PEMFCs", Applied Catalysis B-Environmental 56, 9 (2005).

29.

R. Kothandaraman, V. Nallathambi, K. Artyushkova, and S. Calabrese Barton,
"Non-precious oxygen reduction catalysts prepared by high-pressure pyrolysis for
low-temperature fuel cells", Applied Catalysis B-Environmental 92, 209 (2009).

30.

M. S. Thorum, J. M. Hankett, and A. A. Gewirth, "Poisoning the Oxygen
Reduction Reaction on Carbon-Supported Fe and Cu Electrocatalysts: Evidence
for Metal-Centered Activity", Journal of Physical Chemistry Letters 2, 295
(2011).

31.

J. Burgess and M. V. Twigg, in Encyclopedia of Inorganic Chemistry (John Wiley
& Sons, Ltd, 2006).

32.

G. G. Li, F. Bridges, and C. H. Booth, "X-ray-absorption fine-structure standards
- a comparison of experiment and theory", Physical Review B 52, 6332 (1995).

167

33.

I. Sagi, M. D. Wirt, E. F. Chen, S. Frisbie, and M. R. Chance, "Structure of an
intermediate of coenzyme-B12 catalysis by EXAFS - Ccobalt(II)-B12", Journal
of the American Chemical Society 112, 8639 (1990).

34.

M. D. Wirt, I. Sagi, and M. R. Chance, "Formation of a square-planar Co(I) B-12
intermediate - implications for enzyme catalysis", Biophysical Journal 63, 412
(1992).

35.

S. Kim, I. T. Bae, M. Sandifer, P. N. Ross, R. Carr, J. Woicik, M. R. Antonio, and
D. A. Scherson, "In situ XANES of an iron porphyrin irreversibly adsorbed on an
electrode surface", Journal of the American Chemical Society 113, 9063 (1991).

36.

S. Kim, D. Tryk, I. T. Bae, M. Sandifer, R. Carr, M. R. Antonio, and D. A.
Scherson, "In Situ Extended X-ray Absorption Fine Structure of an Iron
Porphyrin Irreversibly Adsorbed on an Electrode Surface", The Journal of
Physical Chemistry 99, 10359 (1995).

168

6. SUMMARY AND FUTURE DIRECTIONS
The overall objective of the work is to develop non-precious metal-based materials
for oxygen reduction reaction that would meet DOE’s targets for application in proton
exchange membrane fuel cells intended for automotive applications.

The major

challenges towards recognizing this objective include: 1. Understanding the structure of
active reaction sites for oxygen reduction 2. Designing catalysts with more uniform
reaction sites, to increase the turn-over frequency (TOF), which translates to the number
of sites transferred per site per unit time to meet the DOE targets.
As demonstrated in Chapter 2, active Fe-N-C catalysts for oxygen reduction reaction
in a proton exchange membrane fuel cell were synthesized using a novel high-pressure
pyrolysis technique. In comparison to the use of flowing nitrogen gas at atmospheric
pressure, introducing the nitrogen precursor as a high-pressure gas allowed for
amplification and control of nitrogen activity during pyrolysis. Using this high-pressure
pyrolysis approach, Fe-N-C catalysts were synthesized using Ketjen carbon black, Iron
acetate and bi-pyridine as carbon, iron and nitrogen precursors respectively. Increasing
nominal nitrogen content led to increased surface nitrogen content and oxygen reduction
activity up to 10.3 wt.% nominal N; beyond which catalyst mass rose steeply due to
excess carbon deposition from bi-pyridine. A two-fold increase in surface nitrogen
content was observed for catalysts with 10.3 wt.% nominal N, which correlated with
increased open-circuit potential and improved ORR kinetics in the potential region of
interest for operating fuel cells. The current density obtained at 0.65 V/SHE increases
with increasing pyridinic-Fe nitrogen content as determined by XPS analysis,

169

demonstrating that Iron co-ordinated to pyridinic nitrogen structures form part of active
reaction sites for oxygen reduction.
In Chapter 3, the significance of increasing a key property of nitrogen precursors for
metal-nitrogen-carbon (MNC) catalysts, namely the Nitrogen/Carbon (N/C) ratio was
demonstrated. By increasing the N/C ratio of the nitrogen precursor increased the
accessible active site density by reducing carbon deposition in the pores of the carbon
support during pyrolysis. The most active catalysts were obtained in this work using
melamine as the nitrogen precursor, having a N/C ratio of 2. Kinetic current density as
-3

high as 15 A cm

at 0.8 ViR-free and over 100 h of stable current at 0.5 V were

demonstrated in single fuel cells measurements with melamine based Fe-N-C cathode
catalysts. A 33% increase in BET surface area was observed for the melamine catalysts
as compared to bipyridine, indicating that increased surface accessibility contributes to
improved activity. Similarly, nitrogen retention of these catalysts, as measured by CHN
combustion analysis, increased with increasing N/C ratio. A 50 % increase in N retention
was observed for melamine catalysts as compared to bipyridine. Composition analysis
through XPS indicated that increase in pyridinic N sites and Pyridinic Fe-N sites
contributed to improved activity. In conclusion, increase in surface accessibility of the
catalyst, nitrogen content in the form of pyridinic and Pyridinic N co-ordinated to Fe, led
to improved oxygen reduction activity.
In Chapter 4, we looked into carbon-free nitrogen precursors such as ammonia in an
attempt to completely eliminate carbon interference during generation of active sites
during pyrolysis. Ammonia generating nitrogen precursors such as urea and ammonium

170

carbamate were used as N precursors in this study. These precursors etched and increased
the surface area of the catalysts during pyrolysis by creating porosity in the support,
while generating various nitrogen bearing functionalities on the carbon support. N2
adsorption isotherms indicated a 33 % increase in total surface area, 18 % increase in
microporosity and 30% increase in mesoporosity for ammonium carbamate based
catalysts in comparison with bare ketjen black carbon support. Active catalysts obtained
using ammonium carbamate as the nitrogen source indicated an onset potential close to
-2

0.95 V/RHE and current densities close to 12 mA cm

at 0.8 V/RHE in RDE

measurements, which is twice as high in comparison with melamine-based catalysts
addressed in Chapter 3.

XPS analysis confirmed the correlation between oxygen

reduction activity to pyridinic and pyridinic Fe-sites. Single fuel cell measurements
indicated comparable activity to that of melamine and volumetric current density close to
3

13 A/cm at 0.8 V/RHE was achieved using ammonium carbamate based catalysts.

In Chapter 5, the effect of transition metal towards generating active sites in MetalNitrogen-Carbon catalysts has been studied. Electrochemical studies performed in the
presence of poisoning anions such as cyanide in alkaline environment indicated a 25%
decrease in oxygen reduction activity, suggesting that Fe based sites are part of the active
sites and participate in oxygen reduction reaction. In-situ Fe Kα XANES spectra of
melamine-based electrocatalysts measured under Oxygen saturated 0.1 M HClO4
indicated the absence of metallic Fe particles on the surface. Comparison of the
adsorption edge and white line features revealed an oxidation state of +2 for Fe in the
catalysts for potentials up to 0.7 V (region of mixed electrode potential due to kinetics

171

and mass transport) and an oxidation state of +3 for potentials over 0.7 V (purely
kinetically controlled). In-situ EXAFS analysis of the catalysts provided an insight into
the active site to be Fe metal coordinated to 4 nitrogen atoms.
3

In this work, reasonable oxygen reduction activity, ~ 15 A /cm at 0.8 V/RHE and
over 100 h of stable performance has been demonstrated using the Fe-N-C catalysts and
Fe metal coordinated to 4 nitrogen atoms in pyridinic form has been established as the
active site structure of such Fe-N-C catalysts through EXAFS and XPS analyses. The
turn-over frequency for the catalysts was calculated to be 0.8 e
20

density was found to be 1 x 10

20

-1 -1

site s and the site

-3

sites cm . However, to achieve the 2015 target, the turn-

over frequency has to be increased to 5.2 e
increased to 7 x 10

-1

-1

-1 -1

site s

and the site density has to be

-3

sites cm . In order to maximize the number of active reaction sites

and to achieve a uniform distribution of the sites to increase catalyst utilization, the
following steps can be followed.
1. Carbon supports with well- defined porosity should be considered. Ordered
1

mesoporous carbon (with pore sizes from 2 nm to 50 nm), synthesized by casting
2,3

ordered mesoporous silica templates

or by directly templating triblock copolymer

4

structures have been explored extensively as catalyst supports due to their large
surface area, with uniformly distributed three-dimensionally interconnected
mesopores. With their higher surface area and pore volumes, these supports will
allow for better catalyst utilization and efficient transport of reactants and products.
During the high temperature pyrolysis process, addition of aromatic nitrogen

172

precursors such as Melamine will promote graphitization of the mesporous support
2,4,5

and will improve the conductivity of the catalyst.

Higher edge exposure during

graphitization will also maximize the catalytically active Fe-N sites on these supports.

2. To achieve the targeted sites density of 7 x 10

20

-3

sites cm , the number of Fe-N sites

on carbon should be increased, which requires increasing the metal and nitrogen
nominal loading with respect to carbon loading during synthesis. However, any
excess metal particles on the surface might render the catalyst unstable during fuel
cell operation due to leaching of Fe metal, promoting formation of peroxide radicals,
as explained in Chapter 5. To avoid this, as the first step, N can be doped into the
graphitic structure of mesoporous carbon through carbonization, followed by iron
6

incorporation in the core of the N doped mesoporous support. By controlling the
porosity of the support, mass transport can be regulated and by incorporating metal in
7

the core, stability of the material may be improved.

This dissertation presents in detail a novel synthetic procedure for Metal-Nitrogencarbon catalysts, methods to analyze the electro-chemical oxygen reduction activity of
the catalysts and the significant role of nitrogen and transition metal was analyzed in
detail. Collaborative efforts with University of New mexico and North eastern university
led to important conclusion with regard to the nature of the reaction active centers. These
low costs catalysts find direct application in Proton Exchange Membrane Fuel cells for
transportation applications, where there is a huge drive to improve the economy of the
fuel cell by reducing the costs associated with state-of the art Platinum based catalysts.

173

REFERENCES

174

6.1 References
1.

Y. Guo, J. He, T. Wang, H. Xue, Y. Hu, G. Li, J. Tang, and X. Sun, "Enhanced
electrocatalytic activity of platinum supported on nitrogen modified ordered
mesoporous carbon", Journal of Power Sources 196, 9299 (2011).

2.

J. Lee, J. Kim, and T. Hyeon, "Recent Progress in the Synthesis of Porous Carbon
Materials", Advanced Materials 18, 2073 (2006).

3.

Y. Shao, J. Liu, Y. Wang, and Y. Lin, "Novel catalyst support materials for PEM
fuel cells: current status and future prospects", Journal of Materials Chemistry 19,
46 (2009).

4.

P. Yang, D. Zhao, B. F. Chmelka, and G. D. Stucky, "Triblock-Copolymer-Directed
Syntheses of Large-Pore Mesoporous Silica Fibers", Chemistry of Materials 10,
2033 (1998).

5.

Y. Shao, G. Yin, J. Wang, Y. Gao, and P. Shi, "Multi-walled carbon nanotubes
based Pt electrodes prepared with in situ ion exchange method for oxygen
reduction", Journal of Power Sources 161, 47 (2006).

6.

G. Wu, D. Li, C. Dai, D. Wang, and N. Li, "Well-Dispersed High-Loading Pt
Nanoparticles Supported by Shell−Core Nanostructured Carbon for Methanol
Electrooxidation", Langmuir 24, 3566 (2008).

7.

Y. Li, S. Liu, L. Yao, W. Ji, and C.-T. Au, "Core-shell structured iron nanoparticles
for the generation of COx-free hydrogen via ammonia decomposition", Catalysis
Communications 11, 368 (2010).

175