STRUCTURING AND MODIFICATION OF OPEN SURFACE AREA GRAPHENE
NANOPLATELETS TO ENHANCE THE ENERGY DENSITY AND STORAGE
CAPACITY OF ELECTRODES FOR ELECTROCHEMICAL ENERGY STORAGE
APPLICATIONS
By

Debkumar Saha

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Materials Science and Engineering-Doctor of Philosophy
2014

ABSTRACT
STRUCTURING AND MODIFICATION OF OPEN SURFACE AREA GRAPHENE
NANOPLATELETS TO ENHANCE THE ENERGY DENSITY AND STORAGE
CAPACITY OF ELECTRODES FOR ELECTROCHEMICAL ENERGY STORAGE
APPLICATIONS
By

Debkumar Saha
Graphene nanoplatelets (GnP) are platelet shaped particles of graphite less than 5
nanometer in thickness and up to 50 microns in diameter consisting of a few layers of
graphene produced by the exfoliation of graphite. The resulting GnP has a high open
surface area consisting predominantly of the graphene basal plane. High open surface
area coupled with the properties of graphene is an important criterion for electrodes in
many electrochemical energy storage technologies as it provides many electrochemical
reaction sites. Unique platelet morphology of GnP offers the potential to structure selfstanding electrodes by doing away with inactive components such as current collectors.
Surface modification of GnP through the deposition of metal nanoparticles on high open
surface area has the potential to enhance the energy storage capacity due to the
participation of two different components.
In the first part of this research, GnP with different surface area, particle size and
structure combined with an organic electrolyte has been investigated as a Lithium-air
cathode. GnP with a surface area of 750 m 2/gm and submicron particle size was found
to deliver a higher discharge capacity with little overpotential compared to GnP with a
surface area of 120-150 m2/gm and 15 μm average particle size. Binder free, selfstanding, paper-like GnP electrodes have been investigated as higher energy density

alternative to metal current collector cathodes. A hybrid bilayer GnP paper composed of
two types of GnP was found to counter the negative effect of higher electrode loading
on discharge capacity by enhancing the net surface area of an electrode.
High surface area electrodes are an important criterion for high specific capacitance.
The second part of this research investigated GnP as an electric double layer capacitor
(EDLC) electrode material in an aqueous electrolyte. A hybrid bilayer GnP paper was
found to retain near-ideal double layer capacitive characteristics at a high scan rate of 1
V/sec. Specific capacitances of 66 F/gm and 27 F/gm were obtained at a current rate of
1 A/gm from 750 m2/gm GnP and the hybrid bilayer GnP paper respectively. The small
equivalent series resistance (ESR) and charge transfer resistance (Rct) of this GnP
paper electrode was found to be the cause of this desirable behavior.
The third part this research investigated a GnP composite system consisting of high
open surface area GnP combined with metal oxides. XRD and Scanning Electron
Microscopy found that when GnP substrate (120-150 m2/gm, 15 μm average particle
size) is introduced in the wet-chemical synthesis of MnO2, agglomeration is significantly
reduced while retaining the crystal structure. The same observation was made for two
types of manganese oxides, birnessite-MnO2 and γ-MnO2, both of which are of interest
as pseuodocapacitor electrode materials. The introduction of polyethylenimine
surfactant in the system changes manganese oxide morphology from a continuous
ribbon-like structure to nanoparticle clusters and the oxidation state of manganese is
reduced to form Mn3O4. The polymer acts both as a capping agent to restrict crystal
growth and as a reducing agent. The Mn3O4-GnP composite has potential as a higher
performing Lithium-ion battery anode.

Copyright by
DEBKUMAR SAHA
2014

‘To my parents Krishna Saha and Dipendra Prasad Saha’

v

ACKNOWLEDGMENTS

First of all, I would like to thank Professor Lawrence T. Drzal for choosing to be my
mentor and sticking by me through all the trials and tribulations during the course of my
Ph.D. program. I would also like to thank my Ph.D. committee members Professor Phil
Duxbury, Professor Jeff Sakamoto, and Professor Wei Lai for providing important
feedback on my research work that has enriched my dissertation.
Thanks to Composite Materials and Structures Center (CMSC) for making this facility
available for my research. Brian Rook at CMSC deserves special mention for always
being available and willing to help with setup of my experiments. Thanks to fellow
graduate students, postdocs, and visiting scholars at CMSC for help with miscellaneous
things during the course of my Ph.D. program.
Finally, I would like to thank my parents for always being encouraging, their blessings,
and their strong faith that has helped me navigate through all the trials of living abroad
by myself and succeed.

vi

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... xi
LIST OF FIGURES ....................................................................................................... xiii
Chapter 1
Introduction and Background ..................................................................................... 1
1.1 Graphene .................................................................................................................. 1
1.2 The Synthesis and Structure of Graphene Nanoplatelets ......................................... 2
1.3 Motivation and Objective of Research ....................................................................... 5
1.3.1 General Objective .............................................................................................. 5
1.3.2 Lithium-air Battery .............................................................................................. 6
1.3.2.1 Introduction to Lithium-air battery ............................................................... 6
1.3.2.2 Porous high surface area materials and GnP for Lithium-air cathode
fabrication ............................................................................................................... 9
1.3.2.3 Electrolytes ............................................................................................... 11
1.3.3 Electrochemical Capacitors.............................................................................. 12
1.3.3.1 Capacitor and electrochemical capacitor .................................................. 12
1.3.3.2 Electrochemical capacitors and batteries ................................................. 12
1.3.3.3 Applications of electrochemical capacitors ............................................... 14
1.3.3.4 Structure of double-layer and capacitance ............................................... 15
1.3.3.5 Challenges of EDLC electrodes and GnP ................................................. 17
1.3.4 GnP-Manganese Oxides Composites for Electrochemical Energy Storage
Applications............................................................................................................... 18
1.3.4.1 Formation of metal particles...................................................................... 18
1.3.4.2 Controlling particle size, aggregation of nanoparticles and its effect ........ 19
1.3.4.3 Strategies for the reduction of nanoparticle agglomeration ....................... 20
1.3.4.4 Manganese oxides.................................................................................... 22
1.3.4.5 Composites of GnP with manganese oxides and their applications .......... 22
REFERENCES ............................................................................................................. 24
Chapter 2
Investigation of Open Surface Area Graphene Nanoplatelet as a Potential Lithiumair Battery Cathode Material ...................................................................................... 28
2.1 Abstract ................................................................................................................... 28
2.2 Introduction ............................................................................................................. 28
2.3 Methods .................................................................................................................. 32
2.3.1 Synthesis and Characterization of GnP ........................................................... 32
2.3.2 Fabrication of Stainless Steel Wire Cloth Current Collector Supported Electrode
.................................................................................................................................. 33
2.3.3 Fabrication of Nickel Foam Supported Electrodes ........................................... 34
2.3.4 Structural Characterization of Electrodes ......................................................... 35
2.3.5 Setup for Lithium-air Battery Testing ................................................................ 36

vii

2.3.6 Electrochemical Performance Evaluation......................................................... 38
2.4 Results and Discussion ........................................................................................... 39
2.4.1 Structural Characterization ............................................................................... 39
2.4.2 Electrochemical Discharge Tests ..................................................................... 45
2.5 Conclusions ............................................................................................................. 57
2.6 Acknowledgements ................................................................................................. 58
REFERENCES ............................................................................................................. 59
Chapter 3
Graphene Nanoplatelet Based Paper as Binder Free Self-Standing Lithium-air
Battery Cathode: A High Energy Density Alternative to Metal Current Collector
Electrodes ................................................................................................................... 62
3.1 Abstract ................................................................................................................... 62
3.2 Introduction ............................................................................................................. 63
3.3 Methods .................................................................................................................. 64
3.3.1 Fabrication of GnP Paper................................................................................. 64
3.3.2 Structural Characterization of GnP Paper ........................................................ 67
3.3.3 Electrical Conductivity of GnP Paper ............................................................... 67
3.3.4 Thermo Gravimetric Analysis of GnP Paper .................................................... 67
3.3.5 Setup for Lithium-air Battery Testing ................................................................ 68
3.3.6 Electrochemical Performance Evaluation......................................................... 68
3.4 Results and Discussion ........................................................................................... 69
3.4.1 Structural Characterization of GnP Paper ........................................................ 69
3.4.2 Electrical Conductivity of GnP Paper ............................................................... 72
3.4.3 Thermo Gravimetric Analysis of GnP Paper .................................................... 72
3.4.4 Electrochemical Discharge Tests ..................................................................... 73
3.5 Conclusions ............................................................................................................. 76
3.6 Acknowledgements ................................................................................................. 76
REFERENCES ............................................................................................................. 77
Chapter 4
Graphene Nanoplatelet Based Hybrid Bilayer Paper as Binder Free Self-Standing
Lithium-air Battery Cathode: A High Energy Density Alternative to Metal Current
Collector Electrodes. .................................................................................................. 79
4.1 Abstract ................................................................................................................... 79
4.2 Introduction ............................................................................................................. 80
4.3 Methods .................................................................................................................. 80
4.3.1 Fabrication of Hybrid Bilayer GnP Paper ......................................................... 80
4.3.2 Structural Characterization of Hybrid Bilayer GnP Paper ................................. 83
4.3.3 Electrical Conductivity of GnP Paper ............................................................... 83
4.3.4 Thermo Gravimetric Analysis of GnP Paper .................................................... 83
4.3.5 Setup for Lithium-air Battery Testing ................................................................ 84
4.3.6 Electrochemical Performance Evaluation......................................................... 84
4.4 Results and Discussion ........................................................................................... 85
4.4.1 Structural Characterization of Hybrid Bilayer GnP Paper ................................. 85
4.4.2 Electrical Conductivity of GnP Paper ............................................................... 88

viii

4.4.3 Thermo Gravimetric Analysis of GnP Paper .................................................... 89
4.4.4 Electrochemical Discharge Tests ..................................................................... 90
4.5 Conclusions ............................................................................................................. 93
4.6 Acknowledgements ................................................................................................. 94
REFERENCES ............................................................................................................. 95
Chapter 5
High Surface Area Graphene Nanoplatelet Based Paper as Binder Free SelfStanding Electric Double-Layer Capacitor Electrode ............................................. 97
5.1 Abstract ................................................................................................................... 97
5.2 Introduction ............................................................................................................. 98
5.3 Methods .................................................................................................................. 99
5.3.1 Fabrication of GnP Paper................................................................................. 99
5.3.2 Structural Characterization of GnP Paper ...................................................... 100
5.3.3 Electrical Conductivity of GnP Paper ............................................................. 100
5.3.4 Fabrication of Capacitor Cell .......................................................................... 100
5.3.5 Electrochemical Performance Evaluation....................................................... 102
5.4 Results and Discussions ....................................................................................... 103
5.4.1 Structural Characterization of GnP Paper ...................................................... 103
5.4.2 Electrical Conductivity of GnP Paper ............................................................. 103
5.4.3 Cyclic Voltammetry of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP Paper
Electrode ................................................................................................................. 103
5.4.4 Capacitance Determination of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP
Paper Electrode ...................................................................................................... 110
5.4.5 Capacitance Determination of Grade-C-750 GnP .......................................... 114
5.4.6 Impedance Spectroscopy of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP
Paper Electrode ...................................................................................................... 117
5.5 Conclusions ........................................................................................................... 119
5.6 Acknowledgements ............................................................................................... 120
REFERENCES ........................................................................................................... 121
Chapter 6
Manganese Oxides for Electrochemical Energy Storage Applications: Effect of
Graphene Nanoplatelet Substrate and Polymer Surfactant ................................. 125
6.1 Abstract ................................................................................................................. 125
6.2 Introduction ........................................................................................................... 126
6.3 Method .................................................................................................................. 127
6.3.1 Materials ........................................................................................................ 127
6.3.2 Synthesis of Manganese Oxides without PEI ................................................. 127
6.3.3 PEI Mediated Manganese Oxide Synthesis ................................................... 131
6.3.4 Characterization of Composites ..................................................................... 131
6.4 Results and Discussion ......................................................................................... 131
6.4.1 Synthesis of Manganese Oxides without PEI ................................................. 131
6.4.2 Polyethylenimine (PEI) Mediated Manganese Oxides Synthesis ................... 146
6.4.2.1 Morphology of polyethylenimine (PEI) mediated manganese oxides ...... 146
6.4.2.2 XRD of polyethylenimine (PEI) mediated manganese oxides ................. 148

ix

6.4.2.3 Effect of polyethylenimine (PEI) on the morphology and crystal phase of
manganese oxides.............................................................................................. 149
6.4.2.4 Significance Mn3O4 synthesis through polyethylenimine (PEI) mediated
method and application....................................................................................... 149
6.5 Conclusion ............................................................................................................ 150
6.6 Acknowledgements ............................................................................................... 151
REFERENCES ........................................................................................................... 152
Chapter 7
Summary and Recommendation ............................................................................. 156
7.1 Summary ............................................................................................................... 156
7.2 Recommendation .................................................................................................. 158

x

LIST OF TABLES

Table 2-1: characteristics and performances of different types of electrodes in oxygen
atmosphere; OCV stands for open circuit voltage ......................................................... 47
Table 2-2: characteristics and performance of Super-C-65 electrode in oxygen
atmosphere; OCV stands for open circuit voltage ......................................................... 52
Table 3-1: compositions of dispersions for fabricating Grade-M-15 GnP papers (*0.5”
diameter discs were punched from the as fabricated papers to measure weight and
thicknesses; data is average and standard deviation from 9 samples) ......................... 66
Table 3-2: electrical conductivity of GnP papers (from table 3-1) ................................. 72
Table 3-3: comparison of electrochemical discharge performances of paper and metal
current collector electrodes and electrode specifications (for description of GnPM(1)/GnP-M(2) refer to table 3-1; for description of NiFoam/Grade-M-15 GnP refer to
table 2-1) ....................................................................................................................... 73
Table 4-1: compositions of dispersions for fabricating Grade-M-15/Grade-C-750 GnP
hybrid bilayer paper(*0.5” diameter discs were punched from the as fabricated papers to
measure weight and thicknesses; data is average and standard deviation from 9
samples)........................................................................................................................ 82
Table 4-2: comparison of electrochemical discharge performances of GnP-M(1) and
GnP-M(2) paper electrodes (chapter 3) and Grade-M-15/Grade-C-750 hybrid bilayer
GnP paper electrode ..................................................................................................... 92
Table 4-3: comparison of electrochemical discharge performances of metal current
collector electrodes (chapter 2) and GnP paper electrodes (chapter 3 and chapter 4) . 93
Table 5-1: rate performance, cycle performance and capacitance values of Grade-M15/Grade-C-750 hybrid bilayer GnP paper electrode; electrode weight for 1A/g
measurement: 5.3 mg; electrode weight for 10A/g measurement: 5.1 mg .................. 114
Table 5-2: current rate and capacitance values of GnP-M(1) paper electrode and
Grade-C-750 GnP; GnP-M(1) paper electrode weight: 3.3 mg ................................... 116
Table 6-1: composites and manganese oxides; MnOx-a refers to the synthesis of
manganese oxide without GnP substrate corresponding to Comp-a; same explanation
for MnOx-b ................................................................................................................... 130
Table 6-2: composites synthesized through PEI mediated method ............................ 131

xi

Table 6-3: various composites of manganese oxides with GnP, structures and their
potential applications ................................................................................................... 150

xii

LIST OF FIGURES

Figure 1-1: SEM image of Grade-M-15 GnP as received ............................................... 3
Figure 1-2: (a) a large aggregate of Grade-C-750 flakes as received; (b) TEM image of
a few Grade-C-750 GnP flakes [22] ................................................................................ 3
Figure 1-3: the edge surface (B) and the basal plane (A) of graphene layers in a GnP
[22] .................................................................................................................................. 4
Figure 1-4: schematic of cross section of a Lithium-air cathode..................................... 6
Figure 1-5: schematic for the formation of lithium-oxides and electron transport in
porous electrode materials and in GnP (arrow depicts electron transport direction) ....... 9
Figure 1-6: the formation of double-layer at an EDLC electrode .................................. 15
Figure 1-7: cumulative outer surface area of the particles is higher compared to the
outer surface area of the aggregate .............................................................................. 20
Figure 1-8: (a) electrostatic repulsion stabilization with positive charge; (b) combination
of steric stabilization with a cationic polymer and electrostatic repulsion stabilization ... 21
Figure 2-1: (a) doctor-blade film coater covered with aluminum foil; (b) heater; (c)
doctor-blade film applicator; (d) as received stainless steel wire cloth; (e) as prepared
electrode ....................................................................................................................... 34
Figure 2-2: setup for Lithium-air battery testing ............................................................ 37
Figure 2-3: (a) SEM surface image of Grade-M-15 GnP coated side of stainless steel
wire cloth; (b) SEM surface image of uncoated side of stainless steel wire cloth .......... 39
Figure 2-4: (a, b) FIB cross sectional images of Grade-M-15 GnP coated stainless steel
wire cloth ....................................................................................................................... 41
Figure 2-5: (a, b) SEM images of Grade-M-15 GnP coated nickel foam; (c, d) SEM
images of Grade-C-750 GnP coated nickel foam .......................................................... 43
Figure 2-6: electrochemical discharge performances of (a) Grade-M-15 GnP on SSWC;
(b) Grade-M-15 GnP on NiFoam; (c) Grade-C-750 GnP on NiFoam; 1, 2, and 3 shows
stages of discharge; η represents overpotential ............................................................ 46

xiii

Figure 2-7: (a, b) SEM images of Super-C-65; (c) electrochemical discharge
performance of Super-C-65 on NiFoam ; 1, 2, and 3 shows stages of discharge; η
represents overpotential ................................................................................................ 50
Figure 2-8: electrochemical discharge potential plateaus of nickel foam current collector
electrodes with (a) Grade-C-750 GnP; (b) Super-C-65; and (c) Grade-M-15 GnP........ 53
Figure 2-9: electrochemical discharge performances of nickel foam current collector
electrodes at different active material loading with (a) Grade-M-15 GnP; (b) Super-C-65;
(c) Grade-C-750 GnP .................................................................................................... 54
Figure 2-10: electrochemical discharge performances of nickel foam current collector
electrodes in argon ........................................................................................................ 56
Figure 3-1: flowchart for Grade-M-15 GnP paper fabrication ....................................... 66
Figure 3-2: Grade-M-15 GnP paper exhibiting semi-flexibility ...................................... 70
Figure 3-3: surface SEM image of a Grade-M-15 GnP paper ...................................... 70
Figure 3-4: FIB cross-sectional image of a Grade-M-15 GnP paper (GnP-M(1) from
table 3-1) ....................................................................................................................... 71
Figure 3-5: thermo gravimetric analysis of as made and annealed GnP paper (GnPM(1) from table 3-1)....................................................................................................... 73
Figure 3-6: electrochemical discharge performances of (a) GnP-M (1) paper from table
3-1; (b) GnP-M (2) paper from table 3-1 ........................................................................ 75
Figure 4-1: flowchart for Grade-M-15/Grade-C-750 hybrid bilayer GnP paper fabrication
...................................................................................................................................... 82
Figure 4-2: (a) Grade-C-750 top layer of hybrid bilayer GnP paper along with the filter
paper; (b) Grade-M-15 bottom layer of hybrid bilayer GnP paper along with the filter
paper; (c) semi-flexible nature of hybrid bilayer GnP paper .......................................... 85
Figure 4-3: (a, b) Grade-M-15 bottom layer of hybrid bilayer GnP paper; (c, d) Grade-C750 top layer of hybrid bilayer GnP paper ..................................................................... 86
Figure 4-4: thermo gravimetric analysis of Grade-M-15/Grade-C-750 hybrid bilayer GnP
paper (as made and after annealing) ............................................................................ 89
Figure 4-5: advantage of a hybrid bilayer cathode structure ........................................ 91
Figure 4-6: electrochemical discharge performance of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper............................................................................................... 92

xiv

Figure 5-1: assembly of capacitor cell with hybrid bilayer GnP paper: (a) begin with four
sets of stainless steel plates and nut/bolts; (b-e) sandwich the electrodes in between
two stainless steel plates with a glass microfiber filter separating the electrodes, C-750
GnP layer faces the separator; (f) connection of sandwiched assembly to potentiostat
alligator clips; (g) the assembly is put upright in a beaker with electrolyte .................. 102
Figure 5-2: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode; 100 mV/sec scan rate; appearance of a knee
marked in with red arrow ............................................................................................. 105
Figure 5-3: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode; 1 V/sec scan rate; appearance of a knee marked
with red arrow .............................................................................................................. 106
Figure 5-4: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode; 25 mV/sec scan rate; appearance of a knee marked
with red arrow; ‘p’ stands for peak ............................................................................... 108
Figure 5-5: overlapped 9th and 10th cycle CV trajectory of a SS-plate blank EDLC cell;
100 mV/sec scan rate; appearance of a knee marked with red arrow ......................... 108
Figure 5-6: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15 GnP paper
electrode (GnP-M(1)); 100 mV/sec scan rate; appearance of a knee marked with red
arrow ........................................................................................................................... 109
Figure 5-7: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15 GnP paper
electrode (GnP-M(2)); 100 mV/sec scan rate; appearance of a knee marked with red
arrow ........................................................................................................................... 109
Figure 5-8: galvanostatic discharge/charge trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 10th cycle; 1 A/gm current rate ........................ 111
Figure 5-9: galvanostatic discharge/charge profile of a Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper electrode at 110th cycle; 1 A/gm current rate ................................. 112
Figure 5-10: galvanostatic discharge/charge trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 10th cycle; 10 A/g current rate ......................... 113
Figure 5-11: galvanostatic discharge/charge profile of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 110th cycle; 10 A/g current rate ....................... 114
Figure 5-12: galvanostatic discharge/charge trajectory of a Grade-M-15 GnP paper
electrode at 10th cycle; 1 A/g current rate .................................................................... 116

xv

Figure 5-13: impedance spectroscopy of a freshly prepared EDLC cell with Grade-M15/Grade-C-750 hybrid bilayer GnP paper electrode; the inset shows the initial part of
the impedance spectrum ............................................................................................. 118
Figure 5-14: impedance spectroscopy of a freshly prepared EDLC cell with GnP-M(1)
paper electrode; the inset shows the initial part of the impedance spectrum .............. 119
Figure 6-1: setup for GnP-manganese oxides composites synthesis ......................... 130
Figure 6-2: SEM images of ‘Comp-a’ from table 6-1: (a) low magnification image shows
presence of a continuous network-like structure on GnP surface; (b) high magnification
image shows that the structure grows vertically on GnP surface and forms a continuous
structure ...................................................................................................................... 132
Figure 6-3: XRD of composite ‘a’ (Comp-a) from table 6-1; diffraction peaks from
birnessite- MnO2 are marked with asterisks; inset shows the complete XRD spectrum ...
.................................................................................................................................... 134
Figure 6-4 XRD of MnOx-a from table 6-1; diffraction peaks from birnessite-MnO2 are
marked with asterisks .................................................................................................. 135
Figure 6-5: SEM image of GnP surface after reaction of GnP with KMnO4 only (without
MnSO4 as a reducing agent) corresponding to ‘Comp-a’ in table 6-1.......................... 137
Figure 6-6: SEM image of GnP surface after reaction of GnP with KMnO 4 only (without
MnSO4 as a reducing agent) corresponding to ‘Comp-b’ in table 6-1.......................... 138
Figure 6-7: TGA of GnP after reaction of GnP with KMnO 4 only (without MnSO4 as a
reducing agent) corresponding to ‘Comp-a’ in table 6-1 .............................................. 138
Figure 6-8: TGA of GnP after reaction of GnP with KMnO 4 only (without MnSO4 as a
reducing agent) corresponding to ‘Comp-b’ in table 6-1 .............................................. 139
Figure 6-9: SEM image of ‘MnOx-a’ from table 6-1 ..................................................... 140
Figure 6-10: SEM images of ‘Comp-b’ from table 1: (a) presence of two different
structures (left and right); (b) flat disk-like structure; (c) combination of acicular and flat
disk-like structure ........................................................................................................ 141
Figure 6-11 XRD of composite ‘b’ (Comp-b) from table 6-1; diffraction peaks from γMnO2 are marked with asterisks; inset shows the complete XRD spectrum ............... 144
Figure 6-12 XRD of MnOx-b from table 6-1; diffraction peaks from γ-MnO2 are marked
with asterisks ............................................................................................................... 144
Figure 6-13: SEM image of ‘MnOx-b’ from table 6-1 ................................................... 145

xvi

Figure 6-14: SEM images of ‘Comp-c’ from table 6-2: (a) low magnification image of
islands of clustered particles on GnP surface; (b, c) high magnification images of islands
of clustered particles on GnP surface.......................................................................... 146
Figure 6-15: XRD of composite ‘c’ (Comp-c) from table 6-2; diffraction peaks from
Mn3O4 are marked with asterisks; inset shows the complete XRD spectrum .............. 148

xvii

Chapter 1
Introduction and Background

1.1 Graphene
Graphene is a single layer of atomic carbons arranged in a honeycomb lattice [1]. It is
the building block for graphitic carbon materials of different dimensions such as: 0
dimension buckyball, 1 dimension carbon nanotube, and 3 dimension graphite.
Graphene has exceptional electrical, thermal and mechanical properties. Graphene has
received much attention recently due to the works by Geim and Novoselov [1, 21].
Large scale and economic production of graphene is difficult. There are a few methods
for the production of graphene or few layer graphene [2]. Scotch tape micromechanical
cleavage from highly oriented pyrolitic graphite (HOPG) and the reduction of graphite
oxide (GO) has received much attention among all the possible routes for graphene
production. The yield of graphene from scotch tape method is very low although the
quality of the graphene obtained is very high. The ‘GO route’ has high yield; however,
irreversible structural damages take place during the oxidation of graphite to graphite
oxide and its subsequent reduction to obtain graphene.

1

1.2 The Synthesis and Structure of Graphene Nanoplatelets
Particles with platelet morphology have large aspect ratio. Their width is much larger
compared to their thicknesses. xGnP® graphene nanoplatelets (GnP) are ultrathin
particles of graphite consisting of a few layers of graphene. They are synthesized
through proprietary exfoliation process and can be produced with thicknesses ranging
from less than 5 to 20 nanometers and width ranging from 1 to 50 microns [22]. This
method is non-oxidizing and preserves the pristine graphene surface of sp2 carbon
molecules [22]. The nanoplatelets have excellent electrical, thermal, and mechanical
properties and are also potential contenders for carbon nanotube (CNT) [22]. Various
grades of GnP are available with surface area as low as 50 m 2/gm and as high as 750
m2/gm depending upon the degree of exfoliation and particle size [22]. Two different
grades of GnP have been investigated in this research, Grade-M-15 GnP and Grade-C750 GnP. Grade-M-15 GnP has 120-150 m2/gm surface area, 15 µm average platelet
size, and is about 6 nm in thickness. Grade-C-750 GnP has 750 m2/gm surface area
and comes in the form of aggregates of very small flakes (less than 100 nm in
diameter). The aggregates may have sizes in the micron range. The materials show
perfectly graphitic structure as observed from X-Ray Diffraction and Raman
Spectroscopy methods [22]. Oxidation doesn’t take place until about 600°C as observed
from Thermo-gravimetric Analysis [22]. Figure 1-1 and Figure 1-2 are the Scanning
Electron Microscope images of Grade-M-15 GnP and Grade-C-750 GnP respectively.
There are abundant numbers of oxygen and nitrogen functional groups on the edge
surface of graphene layers [16]. The basal plane has many delocalized π-sites [16]. The
edge surface and the basal plane of graphene layers in a GnP are shown in figure 1-3.

2

Figure 1-1: SEM image of Grade-M-15 GnP as received
(a)

Figure 1-2: (a) a large aggregate of Grade-C-750 flakes as received; (b) TEM image
of a few Grade-C-750 GnP flakes [22]

3

Figure 1-2 (cont’d)

(b)

Figure 1-3: the edge surface (B) and the basal plane (A) of graphene layers in a
GnP [22]

4

1.3 Motivation and Objective of Research
1.3.1 General Objective
In recent times, GnP has been investigated as multifunctional additive for polymers and
composites with a myriad of objectives [3-9] and for the fabrication of nanoarchitectured high performance supercapacitor electrodes [10, 11]. Energy plays a key
role in worldwide development. State of technology of several energy storage
technologies have been discussed in details in literature [23]. Alternative energy and
renewable energy have gained much interest in recent times due to oil price volatility,
fossil fuel depletion and a growth in energy demand. Different energy technologies are
available as they find different applications due to their unique characteristics. Energy
storage systems can be broadly classified into two categories, thermal energy and
electrical energy [23]. Electricity is the preferred form of energy for many applications as
it is readily distributed over long distance through cables [23]. Electrical energy storage
refers to the capability of storing energy to produce electricity and release the energy at
the time of energy needs [23]. Electrical energy storage can broadly be classified into
three categories depending upon the principle of energy storage, electrochemical
systems; kinetic energy storage; and potential energy storage [23]. Electrochemical
systems are particularly of interest due to portability and for the potential of meeting
high power and energy density demands. Carbon nanomaterials, particularly graphene,
have received much attention in electrochemical energy storage technologies [14, 15,
18]. High surface area coupled with other properties of GnP (e.g. high electrical
conductivity; different particle size; corrosion and oxidation resistance [16]; chemical
characteristics of edge surface and basal plane) may be of significance in designing

5

electrodes for certain electrochemical energy storage technologies. Electrochemical
capacitors (also known as ‘supercapacitors’ or ‘ultracapacitors’) and Lithium batteries
are at the two ends of the energy density-power density spectrum [24]. Electrochemical
capacitors have the potential for higher power density capability; whereas, Lithium
batteries have higher potential for energy density capability. The objective of this
research has been to explore the value in different properties of GnP to design
electrodes for these electrochemical energy storage technologies.

1.3.2 Lithium-air Battery
1.3.2.1 Introduction to Lithium-air battery
Much interest has been generated recently in Lithium-air battery technology as it
promises to far exceed the energy densities of intercalation electrode based
technologies [12]. This can be attributed to the fact that the cathode oxidant oxygen is
not stored inside the battery but is continuously provided by transport from the
surroundings.

Electrolyte (Li+ ions)

Air (Oxygen)

GnP
Lithium oxides

Surface
passivation layer

Surface passivation
layer

Figure 1-4: schematic of cross section of a Lithium-air cathode

6

A Lithium-air battery consists of a lithium metal anode and a porous cathode separated
by an electrolyte [12, 20]. Figure 1-4 is a schematic of the porous air-cathode only.
Lithium-ions are formed at the metal anode; which are then transported across the
electrolyte into the pores of the cathode. Oxygen diffusing through the porous cathode
is dissolved in the cathode electrolyte and is reduced by the electrons rejected by
lithium metal during its ionization and transported through the external circuit to the
cathode side. The reduced oxygen combines with lithium-ion to form lithium-oxides as
reaction products which are deposited on the cathode particles; usually, carbon or
graphitic materials. This is how energy is stored in a Lithium-air battery. The oxide
formation reaction has to be reversible for a Lithium-air battery to be considered as a
rechargeable system. A surface passivation layer also develops on the outer surface of
the electrode which may be a combination of lithium-oxides and other products formed
through unwanted side reactions when a fresh electrode surface is exposed to an
electrolyte for the first time. The chemical nature of the lithium-oxides formed is
debatable. The following reactions have been found in literature [13]:

O +e →O

(1)

O + Li → LiO

(2)

2LiO → Li O + O

(3)

Reactions (1) to (3) take place during oxygen reduction and the formation of lithiumoxide. An alternative to this route has also been suggested:

LiO + Li + e → Li O

(4)

Li O + 2Li + 2e → 2Li O

(5)

7

For the system to be rechargeable, the following reaction has to take place:

Li O → 2Li + 2e + O

(6)

It is noteworthy that the component in air that takes part in reaction is oxygen only.
Thus, it is also appropriate to call this type of battery system as Lithium-oxygen battery.
The amount of energy stored in a Lithium-air battery is determined by the amount of
lithium-oxides that is formed. This is again determined by several factors [12, 20] such
as: the amount of available surface area for lithium-oxides deposition and the resistance
towards the formation of oxides (electron transport resistance within the electrode
particles and from the electrode particles to the current collector; lithium-ion diffusion
resistance in the bulk electrolyte and within the electrode; solubility of oxygen in the
electrolyte and its diffusion within the electrode etc.). Lower availability of surface area
and higher resistance results in early polarization of the electrode and reduces the
amount of energy that can be stored.
The amount of energy stored in a Lithium-air battery is measured in terms of discharge
capacity. Lithium-oxides are allowed to form at a constant current rate (galvanostatic)
and the Lithium-air battery is allowed to get polarized within a potential window. The
discharge capacity is represented in terms of the product of the current and the amount
of time the battery cell takes to get completely polarized. For the ease of comparison of
different materials, the discharge capacity is normalized with respect to the weight of the
electrode materials. If the unit of current is in mA (mill ampere), the unit of discharge
time is in hours (h), and the unit of electrode weight is gram (gm), then the unit of
discharge capacity would be (mA.h/gm).

8

1.3.2.2 Porous high surface area materials and GnP for Lithium-air cathode
fabrication
As already mentioned, high surface area is an important criterion for high discharge
capacity. However, in order to have high discharge capacity, the surface area must be
accessible to the electrolyte. Some high surface area materials have a large portion due
to micro and nano pores. The surface area of GnP predominantly consists of graphene
basal planes. Thus, the surface area of GnP may be considered as more open and
accessible compared to the surface area of porous materials. Open surface area
platelet shaped materials have potential advantage over porous materials in terms of
available surface area for lithium-oxides deposition and electron transport as well.
These hypothetical advantages are depicted in figure 1-5.

e-

eGnP

Porous electrode material

Lithium oxides

Lithium oxides

Figure 1-5: schematic for the formation of lithium-oxides and electron transport in
porous electrode materials and in GnP (arrow depicts electron transport direction)

9

One of the drawbacks of porous materials is that the pore orifice gets clogged with
reaction products depending upon the pore diameter [20]. This prevents lithium-ions
and oxygen from being transported inside the pore and utilize the inner surface area of
the pore for lithium-oxides deposition. The open surface area of GnP is not likely to
have any such limitation. It is also known that lithium oxides are insulating in nature [20].
If the surface of the electrode material is completely covered with lithium-oxides, the
electrode would polarize rapidly due to electron transport limitations. Early polarization
reduces discharge capacity. Even if the inner and outer surface of a porous material
(figure 1-5) is clean of any insulating products, high electron transport alone may not be
able to sustain lithium-oxides formation if the pore orifice has been clogged with
reaction products inhibiting any further reactant diffusion inside the pore. However, GnP
is likely to be able to sustain lithium-oxides formation as long as clean surface is
available and electron transport from one platelet to another is possible. Again, an
electrode packed with lower open surface area GnP would be polarized rapidly
compared to an electrode packed with higher open surface area variety due to lesser
availability of lithium-oxides deposition sites and high electron transport limitations
caused by insulating effects (assuming that other transport limitations arising from
electrode surface passivation and the internal structure of electrodes remain same for
both types of electrodes). Apart from the internal pores in the electrode material itself,
there are also inter-particle pores caused by the arrangement of electrode material
particles within the electrode structure. These pores affect the discharge capacity of an
air-cathode as well by determining the tortuosity within the cathode. Tortuosity affects
lithium-ion diffusion and dissolved oxygen diffusion within the bulk electrode. To

10

overcome the effect of tortuosity, it is necessary to design air-cathodes carefully through
structuring. GnP with flat platelet like morphology and different platelet sizes should
enable us to design electrodes easily. All of these reasons taken together clearly
demonstrate why is it appropriate to shift attention to platelet shaped GnP over porous
carbon materials for Lithium-air cathode fabrication.

1.3.2.3 Electrolytes
Finding the right electrolyte is a key challenge of this research [20, 25]. Various
categories of organic electrolytes have been investigated so far including propylene
carbonate (PC), ethylene carbonate/dimethyl carbonate (EC/DMC), tetraethylene glycol
dimethyl ether (TEGDME), dimethoxyethane (DME), and dimethyl sulfoxide (DMSO).
They are all prone to attack from oxygen radical during prolonged cycling. In this
research propylene carbonate (PC) based electrolyte has been used. The objective of
the research being to understand the effect of open surface area materials and the aircathode structure on discharge capacity only, electrodes have been discharged without
any cycling attempt to minimize the effect of electrolyte decomposition. The Lithium-air
battery system can be considered as a truly rechargeable system (according to reaction
6) only after a stable electrolyte has been discovered.
The potential of enhancing the energy density of a Lithium-air cathode by creating high
open surface area cathode materials through the exfoliation of graphite has been
investigated with Grade-M-15 GnP (120-150 m2/gm surface area, 15 μm average
particle size) and Grade-C-750 GnP (750 m2/gm surface area, aggregates of submicron platelets) in chapter 2. The designing of cathodes by taking advantage of

11

different particle sizes and surface areas with the objective of enhancing energy density
have been investigated in chapters 3 and 4.

1.3.3 Electrochemical Capacitors
1.3.3.1 Capacitor and electrochemical capacitor
A capacitor is formed by sandwiching two metallic or conductive plates with an
insulating material (known as dielectric) as the separator in between. Electric charge is
stored as energy in between the plates when a voltage is applied across the plates [23].
Electrochemical capacitors run on the same principle as traditional capacitors except
that the dielectric is replaced with an ion-conductor (i.e. electrolyte) and the plates are
replaced with high specific surface area electrodes [23]. Electrochemical capacitors are
also known as ‘supercapacitors’ or ‘ultracapacitors’. This is because of the fact that an
electrochemical capacitor can deliver a capacitance value six to nine orders of
magnitude higher compared to a traditional dielectric based capacitor [24]. The
necessity of electrochemical capacitors over other energy storage technologies such as
batteries stem from the necessity for having higher power capability compared to
batteries. This will be further clarified as the fundamental difference between battery
technologies and electrochemical capacitor technologies is discussed [24, 14, 15, 19].

1.3.3.2 Electrochemical capacitors and batteries
The fundamental difference between electrochemical capacitors and batteries stems
from the mechanism by which charge is stored. In batteries faradaic oxidation and
reduction of electrode materials take place. This results in a generation of charge;

12

however, at the expense of a bulk phase transformation (i.e. a change in oxidation
state) of the electrode materials. A high amount of energy can be stored through the
transformation of oxidation state. The kinetics of such transformation is slow which
results in the loss of power capability. Electrochemical capacitors can be classified in to
two categories, electric double-layer capacitors (EDLC) and pseudocapacitors. It has
already been mentioned that electrochemical capacitors are formed with two high
specific surface area electrodes that are separated by an electrolyte. In an EDLC
electrolyte ions attach themselves as equal and opposite charges on the electrodes
under an external electric field or when a current is passed from one electrode to the
other. High specific surface area of the electrode materials and nanometer scale
thickness of the double-layer results in much higher storage of energy compared to a
traditional dielectric based capacitor. The structure of double-layer and the charge
storage mechanism in a double-layer will be described further later in the discussion.
This type of energy storage doesn’t involve any change in the oxidation state of the
electrode materials which makes the electrode kinetics much faster compared to
batteries and delivers a high power capability. However, at the same time, the stored
energy would be somewhat less compared to batteries due to a surface storage of
charge only. In pseudocapacitors, potential-dependent faradaic charge transfer
becomes thermodynamically favorable in certain range of potential and charge transfer
takes place across the double-layer. In this way, pseudocapacitive behavior is
analogous to batteries. However, the faradaic reaction doesn’t propagate in the bulk of
the materials but takes place only at the electrode/electrolyte interface. Thus, change of
bulk oxidation state of the electrode materials doesn’t take place. Pseudocapacitive

13

materials exhibit higher energy density and relatively lower power capability compared
to double-layer capacitor materials due to partial oxidation state transformation. In
addition, it is possible to cycle an EDLC at least three orders of magnitude longer
compared to a battery due to no oxidation state transformation of the electrode
materials. Thus, it can be seen that capacitors, batteries, and electrochemical
capacitors are not completely different technologies but rather they fall in a continuum
and can be complementary to each other.

1.3.3.3 Applications of electrochemical capacitors
Electrochemical capacitors find applications mainly in three areas; hybrid energy
systems, uninterruptible power supply (UPS), and consumer electronics [24]. In hybrid
energy systems, batteries supply the average power output; whereas, additional peak
power is supplied by electrochemical capacitors. This prevents batteries from reaching
their power limit and enhances their cycle life. UPS are mostly used to supply power for
a short period of time to protect delicate instruments during power outage. Handheld
tools that require high power but only for a short period time (such as screwdrivers) are
good examples of consumer electronics that can use low energy density, high power
capability technology such as electrochemical capacitors. In brief, electrochemical
capacitors find applications in areas where they are used as complementary to batteries
to enhance power capability, to supply high power for a short duration of time, or to act
as a bridge between primary power support and auxiliary power support as a backup
during power outage.

14

1.3.3.4 Structure of double-layer and capacitance
The structure of the double-layer and the mechanism behind the enhancement of
capacitance in an EDLC compared to a traditional dielectric based capacitor has been
discussed in this section. The principle of EDLC has been discussed thoroughly by
Conway [35]. Figure 1-6 depicts the structure of double-layer at an EDLC electrode
surface.

a

b
Surface charge
Specifically adsorbed anion
Water molecules
Hydrated cation

Hydrated anion

Liquid bulk

a: inner Helmholtz plane
b: outer Helmholtz plane

Stern layer Diffuse layer
Figure 1-6: the formation of double-layer at an EDLC electrode
Energy is stored in an EDLC through the separation of charge at electrode-electrolyte
interface. Electrons from one electrode surface are transported through the external
circuit to the surface of the other electrode when an EDLC with a two electrode
configuration as described earlier is placed in an electrical field. The application of an
electric field establishes a potential difference between the electrodes. As a result of this

15

potential difference, surface charge densities of opposite signs (plus and minus) arise
on the electrode surfaces. The charge accumulated is then counter balanced by ions of
opposite signs (cations and anions). This creates a double-layer structure as seen in
figure 1-6. The electrode depicted in this figure has negative charge on it. The plane in
the immediate vicinity of the charged electrode is called the inner Helmholtz plane. This
plane has specifically adsorbed anions. The second plane, named outer Helmholtz
plane corresponds to the plane of hydrated cations. The difference between the inner
plane and the outer plane are that they are the closest distance that an anion or a
hydrated cation can approach towards a charged electrode surface. The inner
Helmholtz plane and the outer Helmholtz plane together constitutes stern layer. In this
layer the potential decreases linearly with distance. Beyond the outer Helmholtz plane is
the diffuse layer. This layer also has charge density. However, in this layer potential
drops exponentially with distance. The furthest point in the diffuse layer at which
potential reaches zero is considered as the boundary between the diffuse layer and the
bulk liquid. Both stern layer and diffuse layer contribute capacitance. If the Helmholtz
type compact double-layer capacitance in the stern layer is designated as CH and the
capacitance contribution from the rest of the charge density in the diffuse layer is
designated as Cdiff , then the overall double-layer capacitance delivered Cdl would be:

1

1

+

1

( )

The specific capacitance of a capacitor is directly proportional to the specific surface
area of an electrode (S) and inversely proportional to the separation of the plates (D)
[24]. The specific surface area of a conventional material used as an EDLC electrode is
likely to be at least three orders of magnitude higher than the surface area of the plates

16

used in a conventional dielectric based capacitor. However, thickness of the doublelayer in an EDLC is likely to be at least three orders of magnitude lower compared to the
separation of plates in a conventional dielectric based capacitor. This explains why an
EDLC electrode would have several orders of magnitude higher specific capacitance
compared to a conventional dielectric based capacitor. This also explains the
importance of high specific surface area materials for the fabrication of EDLC
electrodes.

1.3.3.5 Challenges of EDLC electrodes and GnP
A good EDLC material needs to meet three basic criteria [19]: (i) high surface area; (ii)
high electronic conductivity; (iii) and low resistance to ion transport inside the electrode
structure. A high surface area allows for more charge separation to occur at the
electrode/electrolyte interface. Ideally, an EDLC electrode should be completely
polarizable. Ideal polarizability implies a linear change in potential with the application of
a current or an instantaneous current response under the influence of an electric field.
High electronic conductivity is an essential requirement for high power capability which
is the major purpose of an electrochemical capacitor. Ion transport resistance inside the
electrode structure has to be low for a fast separation of charge, which is again a
criterion for a high power capability. Ion transport inside an EDLC electrode is
determined by the electrode porosity as well as the internal porosities in the electrode
materials. High surface area, high electrical conductivity and the possibility of structuring
of electrodes make GnP an attractive choice as EDLC electrode. In addition, porous
high surface area materials complicate ion transport inside the internal porosities and

17

the mechanism of charge separation inside small pores is yet to be understood [19].
GnP being a high surface area material without any internal pores, has an advantage
over porous high surface area materials in terms of facilitation ion transport. The
potential of Grade-C-750 GnP (750 m2/gm surface area, aggregates of sub-micron
platelets) and Grade-M-15 GnP (120-150 m2/gm surface area, 15 μm particle size) and
the designing of high performing EDLC electrodes have been investigated in chapter 5
with Cyclic Voltammetry (CV), galvanostatic charge-discharge technique, and
impedance spectroscopy technique (EIS).

1.3.4 GnP-Manganese Oxides Composites for Electrochemical Energy Storage
Applications
1.3.4.1 Formation of metal particles
The formation mechanism of metal particles has been discussed in details by Do [16].
Existing techniques for metal particles synthesis can be divided into three categories,
top-down; phase-transformation; and bottom-up. Bottom-up is considered to be the
most versatile, economical and easy to perform procedure. Bottom-up corresponds to a
wet chemical reduction of metal salts or the decomposition of metastable organometallic
compounds. Nucleation, particle growth, and the interaction between particles can be
controlled with relative ease in liquid media. The general mechanism of metal
nanoparticles synthesis involves electron transfer between a reducing agent and an
oxidized metallic species. Metal atoms that are formed by electron transfer reduction
reaction are insoluble in the solution. These metal atoms collide with each other as well
as metal ions to form clusters called ‘embryos’. The first stable metallic phase in such a

18

synthesis system is called ‘nuclei’. Embryos grow to a critical size to form the nuclei. In
the next stage, further addition of the remaining metal atoms causes diffusional growth
of the nuclei to form unstable primary particles. If more metal atoms are provided in the
system, they will continuously diffuse to the surface of the primary particles to form submicrometer size metal particles as well as their aggregates that are even larger in
dimension.

1.3.4.2 Controlling particle size, aggregation of nanoparticles and its effect
Effective synthesis methods are required to produce metal nanoparticles of controlled
size and narrow size distribution. The kinetics of metal particles nucleation process and
its subsequent growth determine particle size and distribution. Synthesis process of
small sized free metal particles in a solution can be controlled by the nature of metal salt
and reducing agent, reaction temperature, and the reaction time [16]. However, small
particles have high surface energy which makes them thermodynamically metastable or
unstable [26, 27]. Small particles tend to sinter together to reduce their net surface area
and thus reduce the surface energy to attain higher stability [28] as shown in figure 1-7.
Many unique properties of nanoparticles can be attributed to their high surface area.
This is because more atoms are located at the surface in nanomaterials due to their
small size compared to bulk materials [29]. Aggregation causes loss in unique
properties of nanomaterials.

19

Aggregation

Figure 1-7: cumulative outer surface area of the particles is higher compared to the
outer surface area of the aggregate
1.3.4.3 Strategies for the reduction of nanoparticle agglomeration
The phenomenon of coarsening of small particles to grow into larger particles is known
as Ostwald ripening [30, 31]. Stabilization strategies for free metal particles in solution
are required to prevent coarsening through aggregation. There are two methods by
which this stabilization can be achieved [16]: electrostatic repulsion and steric
stabilization. Electrostatic repulsion is the method of kinetic stabilization of metal
nanoparticles to overcome Van der Waals force. Electrostatic repulsion takes place due
to the chemisorption of charged species such as H+ or OH-. Thus, the suitable media for
this type of stabilization would be polar organics or aqueous solution (strongly
dependent on its pH value). Steric stabilization is the method of thermodynamic
stabilization of metal nanoparticles in a solution. Steric stabilization is achieved through
the surface modification of metal nanoparticles with a stabilizer such as a polymer. The
principle of stabilization through electrostatic repulsion and steric stabilization are
different [32]. Electrostatic repulsion results in an enhancement of activation energy of
metal nanoparticles and thus preventing their agglomeration. Thermodynamic
stabilization means that the overall energy of the polymer coated product is lowered
compared to the energy of an unmodified metal nanoparticle. Many polymers are either

20

cationic or anionic. Thus, polymer coated metal nanoparticles may experience
electrostatic repulsion as well. Stabilization can be achieved through a combination of
steric and electrostatic stabilization [33]. The phenomena of electrostatic repulsion,
steric stabilization and their combination are shown in figure 1-8.

(a)

+
+

+
+
+

+

Electrostatic

+

Repulsion

+
+

+

+
+

+
+

+

+

+

+

Electrostatic

(b)

Repulsion

+

+

+

+

+

Steric stabilized
with polymer
coating

+

Figure 1-8: (a) electrostatic repulsion stabilization with positive charge; (b)
combination of steric stabilization with a cationic polymer and electrostatic repulsion
stabilization

Another possible solution to the problem of aggregation of small metal particles is to
apply them to a substrate surface. Depending upon the application of the nanoparticles,
the support substrate should have certain desired properties. One such support
substrate is graphite [16]. The basal plane of graphite has many delocalized π-electron
sites. In aqueous media, the π-sites interact with water followed by electrostatic
attraction with metal ions which are later converted to metal atoms [16]. Thus, π-sites
act as anchors for metal particles deposition. The interaction between π-sites and metal
particles acts as a resistance against the coarsening of metal particles and results in
uniform distribution.

21

1.3.4.4 Manganese oxides
Manganese oxide is an interesting metal oxide structure with many crystal phases.
More than 30 manganese oxide crystal phases occur in a wide variety of geological
settings [17]. In natural systems three different oxidation states of Mn can be seen: MnO
(II), Mn2O3 (III), and MnO2 (IV). Manganese oxides can be found in multivalent state as
well, such as Mn3O4 (II, III) (a combination of MnO and Mn2O3). The building block for
manganese oxides structures is the MnO6 octahedra [17]. Different types of structural
arrangements can be obtained through the sharing of edges and corners of these
octahedra. The structures can be broadly classified in to two categories: chain or tunnel
structures and layered structures. The tunnels and the inter-layer spaces may be filled
with water molecules or a wide assortment of cations. These structures can further be
classified depending upon the arrangements of the chain/tunnel structures or layered
structures and the types of cations associated with them. MnO2 may have various
polymorph structures as well such as α, β, γ, and δ [34].

1.3.4.5 Composites of GnP with manganese oxides and their applications
Composites of graphene with metal oxides have found applications as electrodes in
Lithium-ion batteries and as electrodes in pseudocapacitors [18]. A graphene substrate
for the metal oxides serves mainly three purposes: (i) contributes additional capacity to
enhance energy density; (ii) enhances overall electrical conductivity of the electrodes to
enhance power capability; and (iii) accommodates strain in the metal oxide during
electrode cycling to improve cycle life. GnP has the potential to be an alternative
substrate for the reduction of agglomeration manganese oxides due to its high open

22

surface area and platelet morphology. As already mentioned reduction of agglomeration
preserves the properties of nanomaterials. In general, reduction of agglomeration
reduces electrical resistivity and shortens diffusion length of ions in this type of
electrochemical applications. GnP can also create electrically conductive network
among manganese oxide particles to enhance the overall electrical conductivity of the
composite. Wet chemical synthesis and characterization of composites of manganese
oxides combined with GnP that have the potential to find applications as Lithium-ion or
pseudocapacitor electrodes have been investigated in chapter 6 with x-ray diffraction
(XRD) and scanning electron microscopy (SEM) techniques. The effect of polymer
surfactant on controlling of manganese oxide morphology in the composite has also
been investigated.

23

REFERENCES

24

REFERENCES

1. Graphene: status and prospects, Science, 324, 1530 (2009)
2. Ph.D. Dissertation, Jinglei Xiang, Michigan State University (2012)
3. Reduction in percolation threshold of injection molded high density
polyethylene/exfoliated graphene nanoplatelets composites by solid state ball milling
and solid state shear pulverization, J. Appl. Poly. Sci., 124, 525-535 (2012)
4. Multifunctional high density polyethylene nanocomposites produced by incorporation
of exfoliated graphene nanoplatelets 2: crystallization, thermal and electrical properties,
Polymer Composites, 33, 636-642 (2012)
5. Graphene nanoplatelet paper as a light weight composite having excellent electrical,
thermal conductivity and gas barreier properties, Carbon, 50, 1135-1145 (2012)
6. Improving electrical conductivity and mechanical properties of high density
polyethylene through incorporation of paraffin wax coated exfoliated graphene
nanoplatelets and multi-wall carbon nanotubes, Composites: Part A, 42, 1840-1849
(2011)
7. Investigation of exfoliated graphite nanoplatelets (xGnP) in improving thermal
conductivity of paraffin wax-based phase change material, Solar Energy and Solar
Cells, 95, 1811 (2011)
8. Multifunctional high density polyethylene nanocomposites produced by incorporation
of exfoliated graphene nanoplatelets 1: morphology and mechanical properties, Polymer
Composites, 31 (6), 1091-1098 (2010)
9. Mechanical and electrical property enhancement in exfoliated graphene
nanoplatelet/liquid crystalline polymer nanocomposites, Composites: Part A, 42, 371375 (2011)
10. Multilayered nano-architecture of variable sized graphene nanosheets for enhanced
supercapacitor electrode performance, ACS Applied Materials & Interfaces, 2 (8), 22932300 (2010)
11. Multilayered nano-architecture of graphene nanosheets and polypyrrole nanowires
for high performance supercapacitor electrodes, Chem. Materials, 22, 5667 (2010)
12. Lithium-oxygen batteries-limiting factors that affect performance, Journal of Power
Sources, 196, 4436-4444 (2011)

25

13. The pursuit of rechargeable non-aqueous lithium-oxygen battery cathodes, Current
Opinion in Solid State and Materials Science, 16, 178-185 (2012)
14. Graphene for energy conversion and storage in fuel cells and supercapacitors,
Nano Energy, 1, 534-551 (2012)
15. Nanostructured carbon for energy storage, 1, 195-220 (2012)
16. Ph.D. Dissertation, In Hwan Do, Michigan State University (2006)
17. Manganese oxide minerals: crystal structures and economic and environmental
significance, Proc. Natl. Acad. Sci. USA, 96, 3447-3454 (1999)
18. Graphene/metal oxide composite electrode materials for energy storage, Nano
Energy, 1, 107-131 (2012)
19. Materials for electrochemical capacitors, Nature Materials, 7, 845-854 (2008)
20. Critical review of Li/air batteries, Journal of The Electrochemical Society, 159 (2),
R1-R30 (2012)
21. Nature Materials, 6, 183-191 (2007)
22. Available from http:// www.xgsciences.com
23. Energy storage: applications and challenges, Solar Energy Materials and Solar
Cells, 120, 59-80 (2014)
24. Energy storage in electrochemical capacitors: designing functional materials to
improve performance, Energy Environ. Sci., 3, 1238-1251 (2010)
25. Graphene for advanced Li/S and Li/air batteries, J. Mater. Chem. A., 2, 33-47 (2014)
26. High surface energy of free nanoparticles, Physical Review Letters, 91 (10), 106102
(2003)
27. Nanomaterials: an introduction to synthesis, properties and application, Wiley-VCH
(2013)
28. Nanomaterials synthesis, properties, and application, CRC Press (1998)
29. Critical size and surface effect of the hydrogen interaction of palladium clusters, Eur.
Phys. J., D 8, 245-250 (2000)
30. Ostwald ripening with size-dependent rates: similarity and power law solutions,
Journal of Chemical Physics, 117 (17), 8042 (2002)

26

31. Kinetic equations for Ostwald ripening, Physica, 134 A, 323-338 (1986)
32. Encyclopedia of supramolecular chemistry, Marcel Dekker Inc. (2004)
33. Preparation and organization of nanoscale polyelectrolyte-coated
nanoparticles, Advanced Functional Materials, 13 (3), 183-1888 (2003)

gold

34. Manganese oxides for Lithium batteries, Prog. Solid St. Chem., 25, 1-71 (1997)
35. Electrochemical supercapacitors, scientific fundamentals and technological
applications, B.E. Conway, Springer (1999)

27

Chapter 2
Investigation of Open Surface Area Graphene Nanoplatelet
as a Potential Lithium-air Battery Cathode Material

2.1 Abstract
Two different grades of graphene nanoplatelet (GnP), Grade-M-15 (120-150 m2/gm
surface area, 15 µm average particle size) and Grade-C-750 (750 m2/gm surface area,
aggregates of sub-micron platelets) were investigated as potential materials for Lithiumair battery cathode fabrication. Stainless steel wire cloth and nickel foam were used as
current collectors. Galvanostatic discharge tests were performed to measure the
discharge capacity of each material. Performance of Grade-C-750 GnP was found to be
better compared to Grade-M-15 GnP and comparable to Super-C-65 (62 m2/gm surface
area), a highly porous carbon black material.

2.2 Introduction
The total surface area of a powder material is considered to be the sum of the surface
areas of pores of all sizes. Porosity of porous carbon powder has been discussed in
details by Tran et al. [1]. High percentage of small pores result in a material with high
surface area compared to a material with larger pores. A powder material generally has
two types of particles, primary particles and secondary particles which are formed by
the aggregation of the previous type. There are also two types of pores, internal pores
that are formed within the primary particles and interstitial pores that are formed inside

28

the aggregates. Thus, there are two types of surfaces, external surfaces and internal
surfaces. External surfaces are the outer surfaces of primary and secondary particles
and cracks with width-to-depth ratio greater than 1. The walls of all cracks, pores and
cavities that are deeper compared to their width constitute internal surfaces. The
internal and interstitial pores can be classified into three categories according to the
orifice diameter, micro pores (<20 Å); meso pores (>20Å and less than 50 nm) and
macro pores (>50 nm). Solid lithium-oxides are formed on the external surface of the
cathode carbon as well as on the walls inside the micro, meso and macro pores in a
Lithium-air cathode. Although a high surface area provides more active sites to aid
electrochemical reactions, porous high surface area carbons do not always yield the
highest specific capacities. This phenomenon has been illustrated by several research
works. Tran et al. [1] investigated surface areas associated with pore sizes larger than 5
Å, 10 Å and 20 Å in different lab prepared high surface area activated carbon materials.
It was found that the discharge time increased as the average pore diameter and the
surface area associated with pore sizes >20 Å increased. Zhang et al. [2] found in their
work that M-30 carbon with very high specific surface area [2500-3200 m2/gm] delivers
lower capacity compared to Super P carbon that has a much lower specific surface area
[62 m2/gm) [3]. Mesocellular carbon foam has been investigated as Lithium-air cathode
by Yang et al. [4]. They found that mesocellular carbon foam delivers about 40% higher
discharge capacity compared to Super P at an equivalent current density. Since
mesocellular carbon foam has smaller pore size and higher surface area compared to
Super P, this result seems to agree with the premise that high surface area materials
with small pores tend to deliver higher capacity compared to materials with large pores

29

and low surface area. However, this might be due to the fact that mesocellular carbon
foam has a much higher pore volume (1.45 cm 3/gm) compared to Super P (0.32
cm3/gm) [5]. They also found that XC-72 and activated carbon (AC) having much higher
surface area and smaller pore size, delivers lower specific capacity compared to Super
P. Carbon nanotube (CNT) has a surface area comparable to that of Super P but much
smaller pore size and delivers only one third the specific capacity of the later one. Park
et al. [5] have investigated several carbon black materials as Lithium-air cathodes
including Ketjen black and Super P. They found that Ketjen black EC600JD with highest
surface area delivered the highest specific capacity among five materials that were
investigated. However, it should also be mentioned that Ketjen black EC600JD had the
highest pore volume among all of them. The pore sizes of the materials were not
reported. Several commercial carbon materials have been investigated as cathodes for
Lithium-air batteries by Xiao et al. [6]. Ketjen black with pore sizes centered around 30
Å but also as high as 150 Å, high pore volume (7.651 cm 3/gm) and high surface area
(2672 m2/gm) displayed highest specific capacity. The general finding of their research
was that materials with high mesopore volume and narrow mesopore size distribution
shows better performance. Shitta-Bey et al. [7] and Mirzaeian et al. [8, 9] have
investigated polymer derived activated carbons and carbon aerogels as Lithium-air
cathodes. All of these researches unanimously agree that a combination of high pore
volume; large pore size; high surface area and a trade-off among them are the keys to
high discharge capacities. Thus, both availability and accessibility of reaction sites are
important factors to be considered for a high discharge capacity.

30

GnP can have high open surface area without any internal pores. It is appropriate to pay
attention to this material in light of the utilization of reaction sites, reduction of electrode
polarization and its potential in electrode structuring as discussed already in chapter 1.
Graphene nanosheets have been investigated as Lithium-air cathodes [10, 14].
However, graphene nanosheets have several disadvantages. Complicated and time
consuming synthesis procedure and the inability to synthesize at a large scale are just a
few to name. The possibility of enhancing Lithium-air cathode discharge capacity by
creating high open surface area through the exfoliation of graphite has been
investigated in this research.
Super-C-65 (62 m2/gm surface area) is a conductive carbon black synthesized by
TIMCAL Graphite and Carbon. According to the manufacturer it is one of a kind carbon
black material with low surface area compared to other carbon black materials that have
surface areas in thousands of square meters per gram of material [12]. Visibly, this
material obtained from TIMCAL is highly porous (figure 2-7) in spite of its low surface
area. Since GnP and Super-C-65 are at two extremes (i.e. one without any porosity, but
high surface area and the other highly porous, but very low surface area), Super-C-65
has been investigated as a porous baseline in this research to understand the effect of
surface area-porosity relationship on the discharge performance of a Lithium-air
cathode.
Finding stable electrolyte is a key research challenge for Lithium-air batteries [15]. Early
research of Lithium-air batteries was conducted mostly with organic carbonates
because a majority of research on Lithium metal anode was conducted using such
electrolytes [16]. Propylene carbonate based electrolytes particularly found interest due

31

to low volatility [15]. Many research articles have been published in last few years using
this category of electrolyte [1, 2, 4, 6, 9, 13, 17-19]. However, it is now known that in
oxygen rich atmosphere propylene carbonate comes under attack from oxygen radicals
and mostly alkyl carbonates are formed instead of lithium oxides [16]. In spite of the
instability of propylene carbonate, this category of electrolyte is still being used as long
as the study is restricted to only first discharge without any recharge or cycling [20]. In
addition, even if alkyl carbonates are formed, their formation is likely to be governed by
the transport of reactants determined by the electrode structure. Thus, it should still be
possible to discern electrode material or electrode structural differences with this type of
systems. Various alternatives of organic carbonates have also been studied; however,
an electrolyte perfectly stable in oxygen atmosphere is yet to be found [15]. The
objective of this research being investigation of first cycle discharge capacity alone,
propylene carbonate has been chosen as the electrolyte solvent.

2.3 Methods
2.3.1 Synthesis and Characterization of GnP
Two grades of GnP were investigated in this research, Grade-M-15 (120-150 m2/gm
surface area, 15 µm average particle size) and Grade-C-750 (750 m2/gm surface area,
aggregates of sub-micron platelets). Synthesis and characterization of these materials
are as described in chapter 1.

32

2.3.2 Fabrication of Stainless Steel Wire Cloth Current Collector Supported
Electrode
Super corrosion resistant type 316 stainless steel wire cloth (woven) was obtained from
McMaster Carr. The specifications were as follows: (200×200 mesh size; wire diameter
0.0016”; opening width 0.0034”; 46.2% open area). A 3”×3” piece was cut from the as
obtained 12”×12” piece for electrode fabrication. A tape casting doctor-blade film coater
(MTI Corporation) and a doctor-blade film applicator were utilized to coat Grade-M-15
GnP on the stainless steel wire cloth. The top of the doctor-blade film coater was
covered with aluminum foil and vacuum suction was applied to keep it flat. The 3”×3”
piece of wire cloth was attached on top of the aluminum foil with high temperature tape
on all four sides leaving only about 2.5”×2.5” to be coated with GnP. Slurry was
prepared by mixing 0.6 gm GnP, 0.067 gm polyvinylidene fluoride (PVDF) (Kynarflex
2001) and 12 ml N-Methyl-2-pyrrolidione (NMP) (anhydrous, 99.5%, sigma-aldrich) in a
stainless steel mixing vial with a ball mill (SPEX sample prep, 8000D mixer mill). The
slurry was poured on one edge of the wire cloth and then coated uniformly with the help
of the doctor-blade film applicator. The coating speed was adjusted to obtain optimum
uniformity. The GnP coated stainless steel wire cloth was dried overnight on the doctorblade film coater at room temperature followed by a second drying at an elevated
temperature of 120°C for 8 hours with the help of the attached heater. Residual NMP
was removed by an additional step of overnight vacuum oven drying at 120°C to obtain
the final electrode. Only Grade-M-15 GnP was used to fabricate electrode through this
process as the Grade-C-750 GnP particles are too small to be coated on a stainless
steel wire cloth. The thickness of coating can be adjusted by tuning the amount of GnP

33

and NMP in the slurry and by adjusting the film applicator. Higher thickness results in
brittleness of the coated film. 0.5” diameter pieces were punched from the as prepared
2.5”×2.5” electrode and used as cathodes in a Lithium-air test cell as described later.
Figure 2-1 shows the accessories for electrode fabrication and the as prepared
electrode.
(a)

(d)

(b)

(c)

(e)

Figure 2-1: (a) doctor-blade film coater covered with aluminum foil; (b) heater; (c)
doctor-blade film applicator; (d) as received stainless steel wire cloth; (e) as prepared
electrode

2.3.3 Fabrication of Nickel Foam Supported Electrodes
Nickel foam (INCOFOAM®) was obtained from NOVAMET. The specifications of the
foam were as follows: 800 m2/gm areal density; 800 µm pore size and 2.2 mm
thickness. 20 ml N-Methyl-2-pyrrolidione (NMP) (anhydrous, 99.5%, sigma-aldrich) was
taken in a vial followed by the addition of GnP and polyvinylidene fluoride (PVDF)
(Kynarflex 2001) to prepare a slurry through sonication. The amount of NMP and GnP

34

can be adjusted to obtain a desired slurry viscosity. 0.5” diameter pieces were punched
from an as obtained Nickel foam mat, immersed in the slurry and then sonicated for 1
minute. The sonication allows for slurry penetration in the foam structure. The
penetrated slurry coats the cell walls of the foam with GnP. The slurry coated foams are
first air dried under a hood overnight followed by a second drying in a vacuum oven at
120°C for 12 hours for complete removal of NMP. Both grades of GnP (Grade-M-15 and
Grade-C-750) were used to fabricate cathodes. It was observed that the amount of GnP
coated on a foam increases as the concentration of GnP is increased in the slurry.
However, the correlation among slurry concentration, sonication time and the amount of
GnP that is coated on foam was not clear. Slurry coating on nickel foam is not a wellcontrolled process. It was difficult to prepare electrodes with equivalent amount of
material coating for different grades of materials. An electrode with Super-C-65 was
also prepared following the same method.

2.3.4 Structural Characterization of Electrodes
The electrodes were characterized with surface and internal structure imaging.
Scanning electron microscope (SEM) images of surfaces of both types of electrodes
(GnP coated on stainless steel wire cloth and GnP coated on nickel foam) and the
internal structure of the GnP coating on stainless steel wire cloth by cutting with a
focused ion beam (FIB) were captured with a Carl Zeiss Auriga 39 microscope.

35

2.3.5 Setup for Lithium-air Battery Testing
Figure 2-2 depicts the setup of a Lithium-air test cell. There are three components in a
Lithium-air test cell, lithium metal (counter), electrolyte soaked separator and a cathode
(air-electrode). The components are stacked inside a cell body as depicted in figure 2-2.
A spring keeps the component parts under constant pressure. The stainless steel plates
at the top and bottom of the cell body act as electrical connectors to a potentiostat for
the cell stack. The test cell is assembled inside an mBraun inert gas glove box and is
filled with argon gas at the time of assembly. The cell body itself has two valves
(indexed 2 and 3) for oxygen to flow in and flow out. The cell body with valves is
connected to an oxygen cylinder through a cylinder valve (indexed 1). The other end of
the cell is connected to a mineral oil bath. The cell body is purged with oxygen
(Industrial grade, 99.5% purity) before discharge begins. This step removes the argon
gas trapped inside the cell body. After oxygen purging, the outlet valve (indexed 3) is
closed and the cell atmosphere remains exposed to the oxygen cylinder through the
inlet valve (indexed 2) and the cylinder valve (indexed 1). The opening and closing
sequence of various valves are indicated in parenthesis. The oxygen pressure is
maintained at 15 psi, slightly positive to the atmospheric pressure. After a cell is
assembled, about 45 minutes is allowed for all cell parts to be wetted completely. This is
followed by purging with and exposure of cell components to oxygen as described
above. The cell components are exposed to oxygen for about 45 minutes to allow for
complete saturation of the electrolyte with oxygen and cell potential stabilization before
electrochemical discharge tests are performed. The cell components remain exposed to
oxygen for the duration of the electrochemical discharge tests.

36

Stainless steel plate
connected to
potentiostat

Compression spring

Air-cathode
Electrolyte
soaked separator
Lithium metal

Stainless steel plate
connected to
potentiostat

Figure 2-2: setup for Lithium-air battery testing

37

2.3.6 Electrochemical Performance Evaluation
All electrodes used in this research were 0.5” diameter circular discs. All experiments
were carried out with same electrolyte chemistry, 1(M) LiPF6 (Sigma-Aldrich, battery
grade, >99.99% trace metal basis) in propylene carbonate (Sigma-Aldrich, anhydrous,
99.7%). Electrochemical discharge tests were performed with an Arbin BT200
instrument at a current density of 0.05 mA/cm2 of geometrical surface area of electrode.
The potential cut-off limit for discharge tests was 2 V. Glass microfiber filter (Whatman)
was used as separator in all assembled cells. 0.5” diameter lithium counter was
punched from lithium foil (Alfa Aesar, 0.75mm thick, 99.9% metal basis). Separators
were soaked with 150 µl electrolyte in each cell. The nickel foam current collector
electrodes were immersed in the electrolyte in addition to wetting the separator. The
stainless steel wire cloth current collector electrode was not immersed in the electrolyte
separately. The nickel foam having a much higher thickness, it was necessary to have it
wetted separately to ensure complete wetting across the thickness. The amount of
electrolyte that is soaked up from the separator is enough to wet the stainless steel wire
cloth current collector electrode as it is much thinner compared to the nickel foam.
Electrochemical discharge tests of all electrodes were performed in inert argon
atmosphere as well to check for background capacity contribution from unwanted
faradaic reaction between GnP and the electrolyte. Stainless steel wire cloth and nickel
foam without GnP were electrochemically discharged in oxygen atmosphere to check
for background capacity contribution from the current collectors.

38

2.4 Results and Discussion
2.4.1 Structural Characterization

(a)

Figure 2-3: (a) SEM surface image of Grade-M-15 GnP coated side of stainless steel
wire cloth; (b) SEM surface image of uncoated side of stainless steel wire cloth

39

Figure 2-3 (cont’d)

(b)

Figure 2-3 shows the SEM images of stainless steel wire cloth current collector
electrodes with Grade-M-15 GnP. Figure 2-3 (a) is the coated side of the stainless steel
wire cloth. The stainless steel current collector is not visible as it lies underneath. Figure
2-3 (b) is the image of the uncoated side of the stainless steel wire cloth. It seems from
Figure 2-3 (a) that GnP is sitting in the form of a uniform film on the current collector.
Figure 2-3 (b) reveals that GnP has filled up the openings in between the wires.

40

(a)

Figure 2-4: (a, b) FIB cross sectional images of Grade-M-15 GnP coated stainless
steel wire cloth

41

Figure 2-4 (cont’d)

(b)

Figure 2-4 (a, b) are the Focused Ion Beam (FIB) cross sectional images of the GradeM-15 GnP coated stainless steel wire cloth current collector. The GnP coating was cut
from the coated side of the current collector with an ion beam to capture the cross
sectional image. The beam was focused in such a way so as to avoid cutting through
the stainless steel wire cloth. It was cut up to a certain depth to investigate how the
nanoplatelets orient themselves within the electrode structure which is not revealed
from the surface images. The cut is made in the form of a trapezoid well. It is clear from
the images that the nanoplatelets do not exactly stack up on each other. The
nanoplatelets are oriented quite irregularly. This irregularity results in open spaces in
between the nanoplatelets and thus making the electrode porous.

42

(a)

(b)

Figure 2-5: (a, b) SEM images of Grade-M-15 GnP coated nickel foam; (c, d) SEM
images of Grade-C-750 GnP coated nickel foam

43

Figure 2-5 (cont’d)
(c)

(d)

44

Figure 2-5 (a, b) are the SEM images of Grade-M-15 GnP coated on a nickel foam
current collector. In figure 2-5 (a) the grey areas indicate deposition of GnP. There are
also clean areas which appear as white in the image. The pores within the foam don’t
get clogged with GnP and remain open as is seen from the images. Figure 2-5 (b)
shows a magnified image of an area of the foam where GnP has deposited uniformly.
Figure 2-5 (c) shows an area on nickel foam with Grade-C-750 GnP deposit. Magnified
image of the deposit area (figure 2-5 (d)) doesn’t indicate the presence of large micron
sized aggregates. This implies that sonication of GnP in NMP during the fabrication of
electrode de-aggregates the GnP platelets to a certain extent.

2.4.2 Electrochemical Discharge Tests
Figure 2-6 shows the electrochemical discharge performances of stainless steel wire
cloth current collector (with Grade-M-15 GnP) and nickel foam current collector (with
Grade-M-15 and Grade-C-750 GnP) cathodes. Grade-C-750 GnP on nickel foam
current collector displays the best performance obtained from a GnP electrode. It
delivers a discharge capacity value of 688 mAh/gm (for a GnP loading of 4.5 mg on a
nickel

foam

current

collector

weighing

101

mg).

Electrochemical

discharge

performances obtained from different electrodes along with the electrode specifications
are listed in table 2-1. Note that the acronyms for stainless steel wire cloth and nickel
foam are SSWC and NiFoam respectively.

45

(a)
1
η
2
3

(b)
1
η
2
3

Figure 2-6: electrochemical discharge performances of (a) Grade-M-15 GnP on
SSWC; (b) Grade-M-15 GnP on NiFoam; (c) Grade-C-750 GnP on NiFoam; 1, 2, and
3 shows stages of discharge; η represents overpotential

46

Figure 2-6 (cont’d)

(c)

1

η
2
3

Current
collector

Current
collector
weight (mg)

Active material/
weight (mg)

SSWC

21.1

NiFoam

101

NiFoam

101

Grade-M-15
GnP/3.4
Grade-M-15
GnP/2.7
Grade-C-750
GnP/4.5

Discharge
capacity
(mAh/gm of
active material)
285

OCV of cells
before
discharge (V)

228

3.16

688

3.25

3.10

Table 2-1: characteristics and performances of different types of electrodes in
oxygen atmosphere; OCV stands for open circuit voltage

47

The kinetics of discharge process in a Lithium-air cathode is complicated. Discharge
process is affected by several different parameters [13] such as: (i) the porosity inside
the electrode (i.e. available sites for lithium oxides deposition and a path for electrolyte
to make a continuous network); (ii) concentration of oxygen in air-cathode at the
cathode/oxygen interface; (iii) current density; (iv) depth from air-cathode/oxygen
interface into the electrode; and (v) the effective diffusion coefficient of oxygen. A
discharge process involves three steps (figure 2-6). There are two rapid cell potential
drops at the beginning and end of a discharge process. In between these two steps, the
potential remains nearly constant or drops relatively slowly compared to the first and the
last step. Each of these steps has their own interpretation. Initially, there is a rapid
potential drop at the beginning of the discharge process determined by the internal
resistance of the electrode. Potential drop in the middle step is electrochemical reaction
rate dependent. The electrochemical reaction rate is determined by the concentration of
oxygen inside the air-cathode and the availability of reaction product deposition sites.
The electrochemical reaction rate won’t be uniform, if the concentration of oxygen inside
the air-cathode is not uniform or sites for reaction product deposition is lacking. To keep
a constant current discharge (i.e. constant reaction rate), there must be a significant
driving force, known as the overpotential. This results in a continuous potential drop
during the discharge process. The last step of discharge, where potential decreases
rapidly, is caused by a pinch-off effect due to closing off of the electrode from oxygen
gas. Most likely, electrode surface passivation causes oxygen from diffusing to the
electrode material particles. In figure 2-6, approximately the beginning of the third step

48

(i.e. where the knee begins to appear) has been taken as the reference point for
comparison of overpotential.
It can be seen from table 2-1 that Grade-C-750 GnP cathode has the highest open
circuit potential. Initial potential drop in figure 2-6 is same for all three electrodes. The
open circuit potential of Grade-C-750 GnP cathode being higher, it begins to discharge
at a higher electrode potential. Relatively lower surface area Grade-M-15 GnP cathodes
display higher overpotential during their course of discharge due to a higher degree of
polarization compared to relatively higher surface area Grade-C-750 GnP cathode. It is
also interesting to note that two Grade-M-15 GnP electrodes (on nickel foam and
stainless steel wire cloth current collectors) that have different macroscopic structures
deliver similar discharge performance with similar degree of overpotential over the
course of their discharge. This may further be indicative of the fact that higher
polarization of lower surface area Grade-M-15 GnP can be attributed to its surface area
only rather than any electrode structural parameter. Higher overpotential of Grade-M-15
GnP results in a lower discharge capacity in both types of electrodes compared to
Grade-C-750 GnP. It is also interesting to note that the Grade-C-750 GnP discharges
around 2.9 V. Li2O2 and Li2O has been reported to form around 2.95 V and 2.91 V [25].
High overpotential results in less average discharge potential and less power capability
of a cell.
Figure 2-7 and table 2-2 show the structure and electrochemical discharge
performances of a Super-C-65 Lithium-air cathode. The discharge capacity value is
nearly same for Grade-C-750 GnP. Although the average discharge potential of SuperC-65 is lower, it doesn’t display a rapid rise in overpotential over its course of discharge

49

as it happens for Grade-M-15 GnP electrode. High porosity and high surface area, both
being factors that reduce polarization, Grade-C-750 GnP and Super-C-65 display lower
overpotential over the course of their discharge and deliver higher discharge capacity
compared to Grade-M-15 GnP.
(a)

Figure 2-7: (a, b) SEM images of Super-C-65; (c) electrochemical discharge
performance of Super-C-65 on NiFoam ; 1, 2, and 3 shows stages of discharge; η
represents overpotential

50

Figure 2-7 (cont’d)

(b)

(c)

1

η
2
3

51

Current
collector

Current
collector
weight (mg)

Active material/
weight (mg)

NiFoam

101

Super-C-65/3.1

Discharge
capacity
(mAh/gm of
active material)
621

OCV of cells
before
discharge (V)
3.20

Table 2-2: characteristics and performance of Super-C-65 electrode in oxygen
atmosphere; OCV stands for open circuit voltage
A direct comparison of GnP Lithium-air cathode to other researched materials is difficult
due to different research conditions [21, 22]. Specific capacities of different carbon
materials in carbonate based electrolytes have been reported to be in the range of 414
to 2500 mAh/g (at 0.1 mA/cm2 current rate, 2.0 V cut-off limit) [4] and in the range of 579
to 2600 mAh/g depending upon surface area, pore volume, and pore diameter [23].
Graphene nanosheet has delivered capacity values of 2359 mAh/g [14] and 8705.9
mAh/g [24] at current rates of 50 mA/g and 75 mA/g in carbonate based electrolytes.
Whereas these values appear to be high, Ketjen Black (KB) has been reported to
deliver a capacity of 851 mAh/g at a current density of 0.05 mA/cm 2 in an electrolyte of
lithium bis (trifluoromethanesulfonyl) imide in ethylene carbonate/propylene carbonate
(1:1 weight ratio) [6]. Hierarchical arrangement of functionalized graphene sheet has
been reported to deliver a capacity of 15000 mAh/g in ether based electrolyte [26].
Figure 2-8 represents the discharge capacities of three grades of materials and
discharge overpotentials relative to each other (all on nickel foam current collector).
Grade-C-750 GnP shows the least overpotential and highest discharge capacity;
whereas, Grade-M-15 GnP shows the highest overpotential and least discharge
capacity.

52

Figure 2-8: electrochemical discharge potential plateaus of nickel foam current
collector electrodes with (a) Grade-C-750 GnP; (b) Super-C-65; and (c) Grade-M-15
GnP

Figure 2-9 compares discharge profiles and discharge capacities of three types of
electrodes with nickel foam current collector (Grade-M-15 GnP, Grade-C-750 GnP and
Super-C-65) at different active material loading values. The findings agree with other
research reports [5, 11] in that the discharge capacity is reduced with increasing active
material loading and the discharge profile shows more overpotential. For Grade-M-15
GnP the discharge capacity reduces from 228 mAh/gm at 2.7 mg of active material to
167 mAh/gm at 3 mg active material. A discharge capacity of 621 mAh/gm is obtained
at a loading of 3.1 mg of active material for Super-C-65. The value is reduced to 531
mAh/gm at an active material loading of 3.6 mg. The same trend is observed for GradeC-750 GnP with a discharge capacity of 688 mAh/gm at 4.5 mg of active material
loading and a capacity of 586 mAh/gm when the loading is increased to 5.2 mg. It

53

should also be noted that among all the materials investigated in table 2-1, Grade-C750 GnP has the best performance in spite of having the highest active material
loading.

(a)

(b)

Figure 2-9: electrochemical discharge performances of nickel foam current collector
electrodes at different active material loading with (a) Grade-M-15 GnP; (b) Super-C65; (c) Grade-C-750 GnP

54

Figure 2-9 (cont’d)

(c)

Figure 2-10 shows electrochemical discharge performances of nickel foam current
collector electrodes in Argon. The discharge profiles do not have any peak or plateau
indicating that faradaic reactions do not take place on GnP or Super-C-65 surface in
absence of oxygen. Thus, the discharge performances obtained in oxygen atmosphere
can be completely attributed to the presence of oxygen in the system. Discharge time of
Grade-C-750 GnP is higher as the GnP electrodes behave as electric double-layer
capacitor electrodes and the surface area of Grade-C-750 is much higher leading to a
high amount of ion adsorbed ions.
The discharge capacity obtained from current collectors (nickel foam and stainless steel
wire cloth) in oxygen without GnP or Super-C-65 was negligible. Both current collectors
discharged only for 30 seconds compared to hours of discharge time with GnP or
Super-C-65.

55

Figure 2-10: electrochemical discharge performances of nickel foam current collector
electrodes in argon

56

2.5 Conclusions
Surface area and porosity of cathode carbon in determining the electrochemical
performance of Lithium-air cathode has been a long researched subject. The potential
of enhancing Lithium-air cathode discharge capacity by creating high open surface area
through the exfoliation of graphite has been investigated in this research. Two different
grades of GnP (of different surface areas, platelet sizes, and morphology) have been
investigated. Sub-micron platelets of GnP with a surface area of 750 m 2/gm exhibits a
flat discharge potential plateau with very little overpotential and delivers a discharge
capacity of 688 mAh/gm at a low current density of 0.05 mA/cm 2. The performance is
much better compared to 15 μm platelets of GnP with 120-150 m2/gm surface area and
is nearly similar to its highly porous but low surface area baseline material Super-C-65.
Since Grade-C-750 GnP and Super-C-65 have nearly opposite characteristics (one with
high surface area and no internal porosity and the other with high porosity but low
surface area) and both of them display enhancement of discharge capacity over GradeM-15 GnP (a material with low surface area and no internal porosity), we can conclude
that a combination of surface area and porosity would be required to obtain even higher
discharge capacity from a GnP electrode. A direct comparison of the performance of
GnP to that of porous high surface area materials and graphene nanosheets is not
practical as all investigations have been carried out under different conditions. There is
example of prior research where porous carbon black material such as Ketjen Black has
delivered discharge capacity value similar to that of high surface area Grade-C-750
GnP under equivalent condition. However, most porous high surface area materials and
graphene nanosheets have been known to deliver much higher discharge capacity

57

values compared to GnP. Since GnP doesn’t possess any internal porosities, the
possibility of enhancing discharge capacity of GnP depends on the ability to create even
higher surface area through exfoliation and keeping the platelets separated from each
other through careful electrode structuring to create to create a porous network
throughout the electrode.

2.6 Acknowledgements
The author acknowledges the support Composite Material and Structures Center
(CMSC) research staff in accomplishing this work.

58

REFERENCES

59

REFERENCES

1. Investigation of the gas-diffusion-electrode used as lithium/air cathode in nonaqueous electrolyte and the importance of carbon material porosity, Journal of Power
Sources, 195, 2057-2063 (2010)
2. Discharge characteristics of a non-aqueous electrolyte Li/O2 battery, Journal of
Power Sources, 195, 1235-1240 (2010)
3. Characterization of the Lithium/Oxygen organic electrolyte battery, Journal of The
Electrochemical Society, 149 (9), A1190-A1195 (2002)
4. Preparation of mesocellular carbon foam and its application for lithium/oxygen
battery, Electrochemistry Communications, 11, 1127-1130 (2009)
5. Electrochemical performances of Lithium-air cell with carbon materials, Bull. Korean
Chem. Soc., 31 (11), (2010)
6. Optimization of air electrode for Li/Air batteries, Journal of The Electrochemical
Society, 157 (4), A487-A492 (2010)
7. The electrochemical performance of phenol-formaldehyde based activated carbon
electrodes for lithium/oxygen batteries, Journal of The Electrochemical Society, 159 (3),
A315-A320 (2012)
8. Preparation of controlled porosity carbon aerogels for energy storage in rechargeable
lithium oxygen batteries, Electrochimica Acta , 54, 7444-7451 (2009)
9. Characterization capacity loss of lithium oxygen batteries by impedance
spectroscopy, Journal of Power Sources, 195, 6817-6824 (2010)
10. Li-Air rechargeable battery based on metal-free graphene nanosheet catalysts, ACS
Nano, 5 (4), 3020-3026 (2011)
11. High-capacity Lithium-Air cathodes, Journal of The Electrochemical Society, 156 (1),
A44-A47 (2009)
12. Available from http://www.timcal.com/
13. Lithium-air batteries using SWNT/CNF buckypapers as air electrodes, Journal of
The Electrochemical Society, 157 (8), A953-A956 (2010)
14. Graphene nanosheets as cathode catalysts for lithium-air batteries with an
enhanced electrochemical performance, Carbon, 50, 727-733 (2012)

60

15. Making Li-air batteries rechargeable: material challenges, Adv. Funct. Mater., 23,
987-1004 (2013)
16. A critical review on lithium-air battery electrolytes, Phys. Chem. Chem. Phys., 16,
2801-2822 (2014)
17. Identifying capacity limitations in the Li/Oxygen battery using experiments and
modeling, Journal of The Electrochemical Society, 158 (3), A343-A351 (2011)
18. Preparation, characterization and electrochemical catalytic properties of hollandite
Ag2Mn8O16 for Li-air batteries, Journal of The Electrochemical Society, 159 (3), A310A314 (2012)
19. α-MnO2/carbon nanotube/carbon nanofiber composite catalytic air electrodes for
rechargeable Lithium-air batteries, Journal of The Electrochemical Society, 158 (7),
A822-A827 (2011)
20. Evaluation and electrochemical analyses of cathodes for lithium-air batteries,
Journal of Power Sources, 239, 132-143 (2013)
21. Lithium-air batteries: survey on the current status and perspectives towards
automotive applications from a battery industry standpoint, Adv. Energy Mater., 2, 780800 (2012)
22. Carbon-based electrodes for Lithium air batteries: scientific and technological
challenges from a modelling perspective, ECS Journal of Solid State Science and
Technology, 2 (10), M3084-M3100 (2013)
23. Electrochemical performances of Lithiu-air cell with carbon materials, Bull. Kor.
Chem. Soc., 31, 3221-3224 (2010)
24. Superior energy capacity of graphene nanosheets for a nonaqueous Lithium-oxygen
battery, Chem. Commun., 47(33), 9438-9440 (2011)
25. Critical aspects in the development of Lithium-air batteries, J. Solid State
Electrochem., 17, 1793-1807 (2013)
26. Hierarchically porous graphene as a Lithium-air battery electrode, Nano Lett., 11,
5071-5078 (2011)

61

Chapter 3
Graphene Nanoplatelet Based Paper as Binder Free SelfStanding Lithium-air Battery Cathode: A High Energy Density
Alternative to Metal Current Collector Electrodes

3.1 Abstract
Binder free, self-standing paper-like structures have the potential to be used as
electrodes in energy storage technologies. Self-standing structures without the support
of a metal current collector impart flexibility to the electrodes in addition to enhancing
energy density by doing away with electrochemically inactive metal current collectors.
Graphene nanoplatelet (GnP) based papers prepared by the vacuum filtration of GnP
dispersions in water have been investigated as Lithium-air battery cathodes in this
research. Grade-M-15 (120-150 m2/gm surface area, 15 µm average particle size) GnP
has been investigated to prepare GnP paper electrodes of different thicknesses.
Galvanostatic discharge tests were performed to measure the discharge capacity of
each paper. Although the thickness and loading of a GnP paper electrode can be tuned
to match the discharge capacity of a metal current collector electrode, the capacity can’t
be retained at higher electrode thickness and loading level.

62

3.2 Introduction
In chapter 2 it has been found that it is possible to reduce air-cathode polarization and
enhance discharge capacity by using high open surface area material for electrode
fabrication. However, it has also been seen that the enhancement of discharge capacity
is not proportional to the increment of open surface area which is likely related to the
fact that the open surface area GnP platelets do not remain completely separated from
each other in the electrode. The importance of material porosity over surface area alone
has been established as well. A higher discharge capacity enhances the energy density
of an electrode. The implication of all of these observations is that careful structuring of
electrode is necessary. In this research, GnP paper fabricated by taking advantage of
the discrete platelet morphology of large particle size Grade-M-15 GnP, has been
investigated as electrode. Wu has used GnP papers to fabricate light weight composites
[1]. These are binder free, self-standing and somewhat flexible structures made of GnP
possessing high conductivity. The thickness of these paper-like structures is easily
controllable. Paper-like electrode structures fabricated with carbon nanotube (CNT) and
carbon nanofiber (CNF) have already been investigated in Lithium-air and Zinc-air
batteries [2-5]. In general, carbon nanomaterials and graphene based films have been
attracting much attention for flexible energy storage devices in recent times [7, 8]. Not
only that paper-like structures are flexible and offer easy manipulation of electrode
structures, it is also possible to do away with metal current collectors. This results in an
overall enhancement of energy density of an electrode.

63

3.3 Methods
3.3.1 Fabrication of GnP Paper
Paper-like structures have been fabricated by the vacuum filtration of dispersions of
GnP. Grade-M-15 (120-150 m2/gm surface area, 15 µm average particles size) GnP
has been used to fabricate paper in this research. GnP is hydrophobic in nature. The
particles aggregate and don’t disperse well in water. Polyethylenimine (PEI) is a
hydrophilic polymer with primary, secondary and tertiary amino groups and an overall
positive charge in neutral aqueous solution. This cationic polymer can be adsorbed on
GnP surface due to thermodynamic driving forces and thus reduce the hydrophobic
interface between GnP and water or other polar solvents. At pH<10, the positively
charged polymer chain also induces electrostatic repulsion of the particles in water. The
combination of steric and electrostatic stabilization results in a good dispersion of the
GnP in water [6]. The dispersion is sonicated for 2 minutes and stirred overnight to
ensure a strong adsorption of PEI on GnP surface. It is then filtered through a 0.65 µm,
DVPP, Durapore membrane filter (Millipore) under vacuum suction force to obtain the
GnP paper. The vacuum pump is always turned on after the dispersion has been
poured on top of the filter and not otherwise. This results in a better alignment of the
nanoplatelets in the paper. The water with excess PEI filters through the filter paper,
while the GnP is deposited on top of the filter. Although the water filters out in less than
a minute, the vacuum suction is applied for about 5 minutes to ensure complete
drainage of water. Following the filtration, the vacuum pump is turned off and the filter
paper with GnP deposit is removed slowly. The GnP deposit on filter paper is dried
overnight under a hood in air. After overnight drying, the GnP deposit can be peeled off

64

from the filter in the form of a whole paper. The platelets hold together in the form of a
paper through intertwining with each other and through weak van der waals force. The
diameter of the filter paper is 47 mm and the diameter of the GnP paper is slightly less
since the filter is pressed on the edge by a Buchner funnel on top. The as made GnP
paper is likely to have residual PEI on its surface. It is necessary to anneal the paper to
get rid of the PEI. Annealing is done in an air furnace at about 450°C for 90 minutes.
Since the decomposition temperature of PEI is 340°C, it is believed that this annealing
treatment would remove the excess PEI from the GnP paper surface. The annealed
papers were used as is without any pressing. Papers with different thicknesses can be
prepared by tuning the amount of dispersion filtered at a time. In order to get papers of
different thicknesses, different volumes of dispersions were prepared keeping the
concentration of GnP and PEI in the dispersions constant. All other conditions for paper
fabrication such as sonication and stirring times remained constant. The compositions
of the dispersions used for fabricating GnP papers are listed in table 3-1. The
corresponding GnP papers are designated as GnP-M(1) and GnP-M(2). Figure 3-1 is a
flowchart for Grade-M-15 GnP paper fabrication.
Grade-C-750 GnP cannot be used to fabricate a paper following the same method due
to morphological differences. Aggregates of Grade-C-750 GnP cannot hold together in
the form of a paper while being peeled off from the filter; whereas, discrete platelets of
Grade-M-15 GnP can come off as an integrated paper structure.

65

Grade

GnPM(1)
GnPM(2)

Average
weight (mg)
(0.5”
diameter
discs)*

Average
thickness
(μm) (0.5”
diameter
discs)*

Dispersion
volume (ml.)

GnP amount
(gm)

PEI amount
(gm)

100

0.1

0.1

3.1±0.1

61±3

200

0.2

0.2

6.3±0.1

112±5

Table 3-1: compositions of dispersions for fabricating Grade-M-15 GnP papers (*0.5”
diameter discs were punched from the as fabricated papers to measure weight and
thicknesses; data is average and standard deviation from 9 samples)

Add GnP and PEIN to water
Sonicate for 2 minutes
Stir overnight to prepare GnP dispersion
Pour dispersion on a filter paper and turn on vacuum pump
Continue vacuum filtration for 5 minutes and turn off vacuum pump
Remove filter paper with GnP deposit on it
Dry filter paper/GnP deposit inside a hood overnight at normal temperature
Peel GnP deposit in the form of a paper from the filter
Anneal GnP paper at 450°C to get rid of PEI
Figure 3-1: flowchart for Grade-M-15 GnP paper fabrication

66

3.3.2 Structural Characterization of GnP Paper
The GnP papers were characterized with surface and internal structure imaging.
Surface images were obtained with Carl Zeiss EVO LS25 variable pressure scanning
electron microscope (SEM). Internal structure was analyzed by cutting with a focused
ion beam (FIB) microscope (Carl Zeiss Auriga 39).

3.3.3 Electrical Conductivity of GnP Paper
Electrical conductivity of the GnP paper was measured by 4-point probe technique with
a Keithly 2400 Source Meter. Rectangular strips were cut from the GnP papers to make
measurements. The surface of the rectangular strip was probed with four probes. The
meter can measure resistance (R) of the paper strip only. The electrical conductivity of
the paper is obtained through the following set of equations.

⍴= R × (A/L)

(1)

⍴ is the electrical resistivity of the paper strip; A is (thickness × width) of the paper strip;
and L is the distance between two probes. The electrical conductivity (σ) of the paper
strip is then be given by

σ (1/⍴)

(2)

3.3.4 Thermo Gravimetric Analysis of GnP Paper
Thermo gravimetric analysis (TGA) of GnP paper was performed in the temperature
range of 25°C-800°C at a scan rate of 20°C/minute. Both annealed and as made papers
with supposedly residual PEI were analyzed with TGA.

67

3.3.5 Setup for Lithium-air Battery Testing
The setup for battery testing is same as discussed in chapter 2. Since the electrode in
this research is a self-standing GnP paper, which is fragile, it was not directly pressed
under the spring as done with stainless steel wire cloth or nickel foam current collector
electrodes. A mesh was placed in between the compression spring and the GnP paper
electrode. However, the mesh is not an integral part of the electrode and will not be
required if the cell design is improvised where a direct compression under a spring is
not necessary. Any discussion that involves energy density considerations, the mesh is
not taken into account for paper based electrodes as it has been used for the protection
of the paper only, specific to the cell design used in this research. However, for metallic
current collector electrodes, stainless steel wire cloth and nickel foam have been taken
into consideration as they are integral parts of the electrodes and are used to hold
together the powder based electro-active materials.

3.3.6 Electrochemical Performance Evaluation
0.5” diameter circular discs were punched from annealed GnP papers and used as
electrodes. Both types of papers as discussed in table 3-1 were investigated as Lithiumair cathodes. All experiments were carried out with the same electrolyte chemistry, 1(M)
LiPF6 (Sigma-Aldrich, battery grade, >99.99% trace metal basis) in propylene carbonate
(Sigma-Aldrich, anhydrous, 99.7%). Electrochemical discharge tests were performed
with an Arbin BT200 instrument at a current density of 0.05 mA/cm 2 of geometrical
surface area of electrode. The potential cut-off limit for discharge tests was 2 V. Glass
microfiber filter (Whatman) was used as separator in all assembled cells. 0.5” diameter

68

lithium counter was punched from lithium foil (Alfa Aesar, 0.75mm thick, 99.9% metal
basis). Separators were soaked with 150 µl electrolyte in each cell. The amount of
electrolyte that is soaked up from the separator in to the paper electrode is enough to
carry out an experiment. Electrochemical discharge tests were performed after allowing
time for wetting of all cell components and saturation of electrolyte with oxygen as
discussed in chapter 2. Electrochemical discharge test in inert argon atmosphere was
also performed to check for background capacity contribution from unwanted faradaic
reaction between GnP and electrolyte.

3.4 Results and Discussion
3.4.1 Structural Characterization of GnP Paper
Figure 3-2 is a camera image of a GnP paper exhibiting semi-flexibility. Figure 3-3 is a
surface SEM image of a GnP paper showing platelets in a single layer on top of the
paper. The interior structure of a GnP paper (GnP-M(1) from table 3-1) is seen in the
FIB cross sectional image (figure 3-4). The bottom part of the image in figure 3-4 is the
side of the paper that is attached to the filter before it is peeled off. The top part of the
image corresponds to the top part of the GnP paper. It is clear from figure 3-4 that the
GnP platelets are more aligned at the bottom part of a paper; whereas, it is more
disoriented on top. This is associated with the flow pattern of water during the filtration
of dispersion as found previously [6].

69

Figure 3-2: Grade-M-15 GnP paper exhibiting semi-flexibility

Figure 3-3: surface SEM image of a Grade-M-15 GnP paper

70

Figure 3-4: FIB cross-sectional image of a Grade-M-15 GnP paper (GnP-M(1) from
table 3-1)

Although GnP paper electrode structure is easy to control compared to metal current
collector electrodes, the thickness/density of the paper varies slightly over its as made
area depending upon the precision of fabrication and other effects as discussed above.
The difference can be eliminated by pressing; however, it is not a wise choice when the
paper is used for electrode purpose as that would make the path of electrolyte ion and
oxygen diffusion more tortuous.

71

3.4.2 Electrical Conductivity of GnP Paper
Electrical conductivity measurements were performed on both types of GnP papers as
discussed in table 3-1 following equation 1 and 2. The electrical conductivity values
corresponding to each paper have been listed in table 3-2. The values are absolute, not
average. The deviation from paper to paper is less than 5%. It can be seen that the
electrical conductivity values of the two categories of papers are almost same.

Category of paper
GnP-M(1)
GnP-M(2)

Electrical conductivity (S/cm)
432
456

Table 3-2: electrical conductivity of GnP papers (from table 3-1)

3.4.3 Thermo Gravimetric Analysis of GnP Paper
Thermo gravimetric analysis of as made and annealed GnP paper (GnP-M(1) from table
3-1) is compared in figure 3-5. There is a continuous decrease in the as made sample
wt% before GnP starts to oxidize at 600°C; whereas, the annealed sample remains
stable up to 600°C. The weight loss prior to 600°C is a combined effect of adsorbed
water loss and residual PEI loss due to heating. It is clear that the as made sample
retains about 95% of its sample weight even at 600°C. Thus, the amount of residual PEI
in the as made sample is very small, which is in agreement with previous reports [1, 6].
Similar effect of annealing was found for GnP-M(2) paper as well.

72

Figure 3-5: thermo gravimetric analysis of as made and annealed GnP paper (GnPM(1) from table 3-1)

3.4.4 Electrochemical Discharge Tests
Figure 3-6 (a, b) are the electrochemical discharge performances of GnP-M(1) and
GnP-M(2) paper cathodes respectively from table 3-1. A comparison of electrochemical
discharge performances of different paper cathodes and Grade-M-15 GnP on nickel
foam has been made in table 3-3.
Electrode type

Current
collector
weight
(mg)

GnP
weight
(mg)

Thickness
(µm)

GnP-M(1)
GnP-M(2)
NiFoam/GradeM-15 GnP

0
0
101

2.7
6.4
2.7

58
106
-

Discharge
capacity
(mAh/gm of
active
material)
248
40
228

OCV of
cells before
discharge
(V)
3.26
3.18
3.16

Table 3-3: comparison of electrochemical discharge performances of paper and
metal current collector electrodes and electrode specifications (for description of
GnP-M(1)/GnP-M(2) refer to table 3-1; for description of NiFoam/Grade-M-15 GnP
refer to table 2-1)

73

It can be seen that the discharge capacity of GnP-M(1) paper (248 mAh/gm of active
material) is equivalent to that of Grade-M-15 GnP on nickel foam (228 mAh/gm of active
material) current collector (table 3-3). For both categories of electrodes, the amount of
active material is 2.7 mg; however, GnP-M(1) paper doesn’t have any additional weight
from current collector. Considering the overall weight of the electrode including the
current collector, GnP-M(1) paper cathode would have higher energy density. In order
to understand the effect of GnP paper structure on discharge capacity further, a second
discharge experiment was performed with GnP-M(2) paper cathode which had almost
twice the thickness and more than twice the amount of active material compared to the
GnP-M(1) paper cathode. The discharge capacity, as seen from table 3-3, has been
reduced by many folds. Overpotential during the discharge process due to various
transport limitations, as discussed in chapters 1 and 2, would be higher for GnP-M(2)
paper due to its higher thickness and highly tortuous internal structure as seen in figure
3-4. This causes a reduction in discharge capacity when normalized to the electrode
weight as much of the material remains unutilized.

74

(a)

(b)

Figure 3-6: electrochemical discharge performances of (a) GnP-M (1) paper from
table 3-1; (b) GnP-M (2) paper from table 3-1

75

Electrochemical discharge tests of Grade-M-15 GnP in argon doesn’t show any peak or
plateau that indicates a faradaic reaction between Grade-M-15 GnP and the electrolyte
as discussed already in chapter 2. Thus, the discharge capacity obtained in oxygen
atmosphere can be completely attributed to the presence of oxygen in the system.

3.5 Conclusions
Large particle size Grade-M-15 GnP can be structured to form self-standing paper-like
Lithium-air cathode. The amount of material and the thickness of Grade-M-15 GnP
paper cathode can be tuned to match the discharge capacity of a nickel foam current
collector Lithium-air cathode. Thus, a Grade-M-15 GnP paper cathode makes a
significant gain over a nickel foam cathode in terms of energy density by doing away
with high weight and electrochemically inactive metal current collector. Discharge
capacity is reduced with increasing paper electrode thickness and weight due to a
higher degree of polarization. It is necessary to innovate electrode designs that can
retain or improve the discharge capacity at a higher electrode weight so that the current
can be delivered for a longer time that meets the need of practical applications. One
possible solution may be to come up with a method to incorporate large surface area
Grade-C-750 GnP in a paper-like electrode structure.

3.6 Acknowledgements
The author acknowledges the support of Composite Materials and Structures Center
(CMSC) research staff at Michigan State University in accomplishing this work.

76

REFERENCES

77

REFERENCES

1. Graphene nanoplatelet paper as a light-weight composite with excellent electrical and
thermal conductivity and good gas barrier properties, Carbon, 50 (3), 1135-1145 (2012)
2. Lithium-air batteries using SWNT/CNF buckypapers as air electrodes, Journal of The
Electrochemical Society, 157 (8), A953-A956 (2010)
3. Preparation, characterization and electrochemical catalytic properties of Hollandite
Ag2Mn8O16 for Li-air batteries, Journal of The Electrochemical Society, 159 (3), A310A314 (2012)
4. α-MnO2/carbon nanotube/carbon nanofiber composite catalytic air electrodes for
rechargeable Lithium-air batteries, Journal of The Electrochemical Society, 158 (7),
A822-A827 (2011)
5. Paper like free-standing hybrid single-walled carbon nanotubes air electrodes for
zinc-air batteries, J. Solid State Electrochem., 16, 1585-1593 (2012)
6. Ph.D Dissertation, Jinglei Xiang, Michigan State University (2012)
7. Carbon nanomaterials for flexible energy storage, Materials Research Letters, 1 (4),
175-192 (2013)
8. Graphene films for flexible organic and energy storage devices, J. Phys. Chem. Lett.,
4, 831-841 (2013)

78

Chapter 4
Graphene Nanoplatelet Based Hybrid Bilayer Paper as Binder
Free Self-Standing Lithium-air Battery Cathode: A High
Energy Density Alternative to Metal Current Collector
Electrodes

4.1 Abstract
Enhancement of energy density of Lithium-air cathode by replacing metal current
collector electrode with graphene nanoplatelet (GnP) paper electrode has been
investigated in chapter 3. Grade-M-15 (120-150 m2/gm surface area, 15 µm average
particle size) GnP can be formed into paper-like structures through a vacuum filtration
method. It is necessary to innovate a new design to incorporate Grade-C-750 GnP (750
m2/gm surface area, aggregates of sub-micron platelets) in a paper-like structure as
Grade-C-750 GnP delivers much higher discharge capacity compared to Grade-M-15
GnP. In addition, as it has been mentioned in chapter 3, it is necessary to retain or
improve the discharge capacity of an electrode at high electrode thickness and loading
level in order to be able to deliver current for a long time that meets the needs of
practical applications. A hybrid bilayer Grade-M-15/Grade-C-750 GnP paper has been
fabricated and investigated as Lithium-air cathode in this research. This category of
GnP paper has the potential to reverse the phenomenon of discharge capacity
reduction at high electrode thickness and loading level.

79

4.2 Introduction
A strategy for enhancing the energy density of a Lithium-air cathode through the
fabrication of a GnP paper cathode by taking advantage of the discrete platelet
morphology of large particle size Grade-M-15 GnP has been demonstrated in chapter 3.
Grade-C-750 is an agglomerate of sub-micron platelets. The aggregates can’t hold
together with weak van der waals force in the form of a paper-like structure when being
peeled off from a filter as it does for the discrete large platelets of Grade-M-15 GnP.
Many researchers have used commercial carbon cloth as the current collector for high
surface area porous carbons [1-4]. Slurries of high surface area carbon have been
coated on carbon cloth followed by drying to remove the solvent. Slurry coating is a
complex and time consuming process. An alternative could be to prepare a hybrid
bilayer paper through a two-step vacuum filtration method where one component acts
as the substrate for the deposition of the other component. This hybrid bilayer paper,
when used as Lithium-air cathode, has the potential to deliver higher discharge capacity
compared to a Grade-M-15 GnP paper electrode of equivalent dimension due to a net
surface area gain by the incorporation of Grade-C-750 GnP.

4.3 Methods
4.3.1 Fabrication of Hybrid Bilayer GnP Paper
High surface area Grade-C-750 GnP has been incorporated in a paper-like structure by
fabricating a hybrid bilayer GnP paper. In this type of paper there are two layers with
two compositions, one layer composed of Grade-M-15 GnP and a second layer
composed of Grade-C-750 GnP. First, Grade-M-15 GnP is deposited on a filter paper

80

as described in chapter 3. However, the deposit is not peeled off in the form of a GnP
paper after drying overnight inside a hood at normal temperature. A separate dispersion
of Grade-C-750 GnP is prepared. The filter paper, with dry Grade-M-15 GnP deposit on
top, is used to filter the second dispersion of Grade-C-750 GnP. This results in a second
deposit of Grade-C-750 GnP on top of the Grade-M-15 GnP deposit. The filtration of a
Grade-C-750 GnP dispersion takes longer compared to the filtration of Grade-M-15
GnP dispersion on a fresh filter paper. The composition of each dispersion for the
fabrication of hybrid bilayer GnP paper is same as that of a GnP-M(1) paper fabricated
in chapter 3 (table 3-1). The bilayer deposit is dried overnight inside a hood. After
overnight drying, the deposit can be peeled off in the form of a hybrid bilayer paper. The
paper is annealed in an air furnace to get rid of residual PEI as described in chapter 3.
Ideally, a hybrid bilayer paper electrode having the same diameter as that of a GnPM(1) paper electrode should weigh twice that of the GnP-M(1) paper, which is about 6.2
mg. The compositions of dispersions used for the fabrication of hybrid bilayer composite
paper are listed in table 4-1. Figure 4-1 is a flowchart for hybrid bilayer GnP paper
fabrication.

81

Add Grade-M-15 GnP and PEI to water
Sonicate for 2 minutes
Stir overnight to prepare Grade-M-15 GnP dispersion
Pour dispersion on a filter paper and turn on vacuum pump
Continue vacuum filtration for 5 minutes and turn off vacuum pump
Remove filter paper with Grade-M-15 GnP deposit on it
Dry filter paper/Grade-M-15 GnP deposit inside a hood overnight at
normal temperature

Prepare a Grade-C-750 GnP dispersion similar to Grade-M-15 GnP
dispersion
Filter the Grade-C-750 GnP dispersion on top of the Grade-M-15
deposit
Dry filter paper with Grade-M-15 and Grade-C-750 deposit inside a
hood overnight at normal temperature
Peel GnP deposit in the form of a hybrid bilayer paper from the filter
Anneal GnP paper at 450°C to get rid of PEI
Figure 4-1: flowchart for Grade-M-15/Grade-C-750 hybrid bilayer GnP paper
fabrication

GnP Grade
M-15

Dispersion
volume (ml.)
100

GnP amount
(gm)
0.1

PEI amount
(gm)
0.1

C-750

100

0.1

0.1

Average
weight (mg)
(0.5”
diameter
discs)*

Average
thickness
(μm) (0.5”
diameter
discs)*

5.2±0.2

123±11

Table 4-1: compositions of dispersions for fabricating Grade-M-15/Grade-C-750 GnP
hybrid bilayer paper(*0.5” diameter discs were punched from the as fabricated
papers to measure weight and thicknesses; data is average and standard deviation
from 9 samples)

82

It can be seen that the weight of a 0.5” diameter disc punched from a Grade-M15/Grade-C-750 hybrid bilayer GnP paper weigh about 5.2 mg, almost 1 mg less than
the estimated value of 6.2 mg. This can be attributed to the loss of some material from
the top Grade-C-750 layer during the processing of hybrid bilayer GnP paper.

4.3.2 Structural Characterization of Hybrid Bilayer GnP Paper
The hybrid bilayer GnP paper was characterized with surface and internal structure
imaging. Scanning electron microscope (SEM) images of surface and the images of
internal structure by cutting with a focused ion beam (FIB) were captured with a Carl
Zeiss Auriga 39 microscope.

4.3.3 Electrical Conductivity of GnP Paper
Electrical conductivity of the GnP paper was measured by 4-point probe technique with
a Keithly 2400 Source Meter. The method for measuring electrical conductivity is same
as described in chapter 3.

4.3.4 Thermo Gravimetric Analysis of GnP Paper
Thermo gravimetric analysis (TGA) of GnP paper was performed in the temperature
range of 25°C-800°C at a scan rate of 20°C/minute. Both annealed and as made papers
with supposedly residual PEI were analyzed with TGA.

83

4.3.5 Setup for Lithium-air Battery Testing
The setup for battery testing is same as discussed in chapter 2. The self-standing hybrid
bilayer GnP paper electrode was not pressed directly under a compression spring but
with a mesh in between the spring and the electrode as discussed in chapter 3.

4.3.6 Electrochemical Performance Evaluation
0.5” diameter circular discs were punched from annealed hybrid bilayer GnP paper and
used as electrodes. All experiments were carried out with the same electrolyte
chemistry, 1(M) LiPF6 (Sigma-Aldrich, battery grade, >99.99% trace metal basis) in
propylene carbonate (Sigma-Aldrich, anhydrous, 99.7%). Electrochemical discharge
tests were performed with an Arbin BT200 instrument at a current density of 0.05
mA/cm2 of geometrical surface area of electrode. The potential cut-off limit for discharge
test was 2 V. Glass microfiber filter (Whatman) was used as separator in all assembled
cells. 0.5” diameter lithium counter was punched from lithium foil (Alfa Aesar, 0.75mm
thick, 99.9% metal basis). Separators were soaked with 150 µl electrolyte in each cell.
The amount of electrolyte that is soaked up from the separator in to the paper electrode
is enough to carry out an experiment. Electrochemical discharge tests were performed
after allowing time for wetting of all cell components and saturation of electrolyte with
oxygen as discussed in chapter 2. Electrochemical discharge test in inert argon
atmosphere was also performed to check for background capacity contribution from
unwanted faradaic reaction between GnP and electrolyte.

84

4.4 Results and Discussion
4.4.1 Structural Characterization of Hybrid Bilayer GnP Paper
Figure 4-2 shows the camera images of top and bottom layers of a hybrid bilayer GnP
paper along with the semi-flexible nature of the paper. The top part of the hybrid bilayer
paper is dark in color (Grade-C-750 GnP). The bottom part of the hybrid bilayer is grey
in color (Grade-M-15 GnP).
(a)

(b)

(c)

Figure 4-2: (a) Grade-C-750 top layer of hybrid bilayer GnP paper along with the
filter paper; (b) Grade-M-15 bottom layer of hybrid bilayer GnP paper along with the
filter paper; (c) semi-flexible nature of hybrid bilayer GnP paper

85

Figure 4-3 shows SEM images of the top and bottom layers of Grade-M-15/Grade-C750 hybrid bilayer GnP paper. Figure 4-3 (a, b) show Grade-M-15 GnP platelets in a
single layer at the bottom of the hybrid bilayer paper. Figure 4-3 (c, d) show the GradeC-750 GnP deposit in the top layer. Figure 4-3 (c) shows mud-crack formation in the top
Grade-C-750 GnP layer. The mud-cracks are developed during the drying process.
Since Grade-C-750 GnP is of aggregate morphology, they don’t hold together in a
uniform layer upon drying as it happens for platelet shaped Grade-M-15 GnP. However,
there should be good enough contact in the thickness direction.
(a)

Figure 4-3: (a, b) Grade-M-15 bottom layer of hybrid bilayer GnP paper; (c, d)
Grade-C-750 top layer of hybrid bilayer GnP paper

86

Figure 4-3 (cont’d)
(b)

(c)

87

Figure 4-3 (cont’d)
(d)

The internal structure of the Grade-M-15 GnP layer would be same as that of the
internal structure of GnP-M(1) paper discussed in chapter 3 (figure 3-4). The internal
structure of the Grade-C-750 layer was not investigated with FIB cutting. It can be seen
from the high magnification image of the top Grade-C-750 GnP layer (figure 4-3(d)) that
the aggregates are just randomly stacked on top of each other.

4.4.2 Electrical Conductivity of GnP Paper
Electrical conductivity of a Grade-M-15/Grade-C-750 hybrid bilayer GnP paper was
measured to be around 263 S/cm. Electrical conductivity of the hybrid bilayer GnP
paper is much lower compared to un-hybridized GnP papers of any thickness as listed
in table 3-2 of chapter 3. This is because the Grade-C-750 GnP top layer doesn’t have

88

any or little electrical conductivity in the in-plane direction as the particles are not well
connected with each other. However, this layer would still have current capability as the
particles connected with each other in the thickness direction and are deposited on
Grade-M-15 GnP layer with high electrical conductivity.

4.4.3 Thermo Gravimetric Analysis of GnP Paper
Figure 4-4 shows the thermo gravimetric analysis of a Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper, as made and after annealing at 450°C. The annealed paper remains
stable up to 600°C before oxidation of GnP begins. The as made paper shows a
continuous weight loss up to 600°C; which is a combination of loss of adsorbed water
and residual PEI. The weight% retained at 600°C is less than that of GnP-M(1) paper
(figure 3-5 of chapter 3) for the as made sample. This is a combined effect of higher
amount of adsorbed water and a higher amount of residual PEI.

Annealed

As made

Figure 4-4: thermo gravimetric analysis of Grade-M-15/Grade-C-750 hybrid bilayer
GnP paper (as made and after annealing)

89

4.4.4 Electrochemical Discharge Tests
Figure 4-6 shows the electrochemical discharge performance of a Grade-M-15/GraceC-750 hybrid bilayer GnP paper. The hybrid bilayer GnP paper delivers a discharge
capacity of 349 mAh/gm from a 5.1 mg electrode. A comparison of electrochemical
discharge performances of GnP-M(1) and GnP-M(2) papers (chapter 3) and Grade-M15/Grade-C-750 hybrid bilayer GnP paper has been made in table 4-2. Compared to the
discharge capacity of a GnP-M(1) paper, the performance of the hybrid bilayer GnP
paper is higher by 100 mAh/gm. Grade-M-15/Grade-C-750 hybrid bilayer GnP paper
delivers a higher discharge capacity in spite of the fact that its electrode loading is
almost twice that of GnP-M(1) paper electrode. High electrode loading reduces
discharge capacity as discussed already in chapter 2 and chapter 3. This trend has
been reversed though hybridization as high surface area Grade-C-750 GnP provides
more sites for reaction product deposition. Having a large normalized discharge
capacity from a small amount of material doesn’t solve practical problems. It is
important to be able to retain or improve normalized discharge capacity with higher
amount of active material. It is the only way to deliver current for a long time that meets
the needs of a practical application. Figure 4-5 depicts the advantage of a hybrid bilayer
cathode structure. However, it is also important to note that the discharge profile of
Grade-C-750 GnP on Grade-M-15 GnP paper substrate is not flat plateau-like as is
observed with a nickel foam current collector (figure 2-6 (c), chapter 2). Also,
considering the fact that in this type of hybrid paper electrode both Grade-M-15 layer
and Grade-C-750 layer are supposed to deliver discharge capacity and the fact that
Grade-C-750 GnP alone on a nickel foam current collector delivers a discharge capacity

90

of 688 mAh/gm, a much higher discharge capacity would be expected from the hybrid
paper electrode. The use of Grade-M-15 paper as a substrate, which has already been
seen to easily become polarized, might have affected the discharge capacity that can
be harnessed from Grade-C-750 GnP. Also, it is possible that the de-agglomeration of
Grade-C-750 GnP aggregates can be achieved to a much lesser extent through
sonication in water/PEI solution compared to sonication in NMP as can be seen from
the comparison of electrode structures in figure 2-5 (chapter 2) and figure 4-3 (d) (this
chapter). In spite of these potential drawbacks, the Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper is advantageous from energy density perspective due to low current
collector weight.

Electrolyte (Li+ ions)

Air (Oxygen)

1 High surface area

Grade-C-750 GnP
aggregates
Grade-M-15 GnP
paper layer

2 Support surface

2

1

Figure 4-5: advantage of a hybrid bilayer cathode structure

91

Red arrows
indicate
electron
transport
directions

Figure 4-6: electrochemical discharge performance of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper
Electrode type

Current
collector
weight
(mg)

GnP
weight
(mg)

Thickness
(µm)

GnP-M(1)
GnP-M(2)
Grade-M15/Grade-C-750
bilayer

0
0
0

2.7
6.4
5.1

58
106
142

Discharge
capacity
(mAh/gm of
active
material)
248
40
349

OCV of cells
before
discharge
(V)
3.26
3.18
3.38

Table 4-2: comparison of electrochemical discharge performances of GnP-M(1) and
GnP-M(2) paper electrodes (chapter 3) and Grade-M-15/Grade-C-750 hybrid bilayer
GnP paper electrode

Table 4-3 compares the discharge capacities and electrode weights of various metal
current collector and paper-like electrodes. The advantage of paper-like electrode
structures over metal current collector electrodes is clearly revealed from table 4-3 with
Grade-M-15/Grade-C-750 hybrid bilayer GnP paper showing the highest gain in terms
of energy density.

92

Electrode type

Current
collector
weight (mg)

GnP
weight
(mg)
2.7

Discharge
capacity
(mAh/gm of
active material)
228

OCV of cells
before
discharge
(V)
3.16

NiFoam/Grade-M-15
GnP
NiFoam/Grade-C-750
GnP
GnP-M(1)
GnP-M(2)
Grade-M-15/Grade-C750 bilayer

101
101

4.5

688

3.25

0
0
0

2.7
6.4
5.1

248
40
349

3.26
3.18
3.38

Table 4-3: comparison of electrochemical discharge performances of metal current
collector electrodes (chapter 2) and GnP paper electrodes (chapter 3 and chapter 4)

4.5 Conclusions
High surface area, small particle size Grade-C-750 GnP aggregates can be
incorporated in a paper-like structure by hybridizing it with large particle size Grade-M15 GnP. This hybridization is achieved with a bilayer structure fabricated through a twostep vacuum filtration technique. This is an easier alternative to slurry casting. The trend
of discharge capacity loss with higher thickness and higher amount of material loading
as observed for Grade-M-15 GnP paper electrode is reversed with a hybrid bilayer
Grade-M-15/Grade-C-750 GnP paper electrode. Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper electrode has the highest normalized discharge capacity among all
paper-like electrodes. Thus, it should have made the highest gain in terms of energy
density by doing away with metal current collectors. Overall, the research discussed in
chapters 2, 3, and 4 convey the idea that it is possible to enhance the discharge
capacity of a Lithium-air cathode by creating high surface area through the exfoliation of
graphite and that design strategies can be adopted to enhance the energy density of an
electrode. However, to make the performance of GnP comparable to that of porous high

93

surface area materials, innovative electrode fabrication strategies need to be adopted
that can keep the platelets separated from each other. The highest obtainable surface
area through the exfoliation of graphite so far is 750 m 2/gm (Grade-C-750 GnP).
Enhancement of discharge capacity would also depend on the ability to form high
surface area in thousands of meters square per gram range to make the discharge
capacity comparable to most porous high surface area materials.

4.6 Acknowledgements
The author acknowledges the support of Composite Materials and Structures Center
(CMSC) research staff in accomplishing this work.

94

REFERENCES

95

REFERENCES

1. Electrochemical performance of a non-aqueous rechargeable lithium-air battery,
Ionics, 19, 1791-1793 (2013)
2. Studies of Li-air cells utilizing Dimethyl Sulfoxide-based electrolyte, J. Electrochem.
Soc., 2, A259-A267 (2013)
3. An improved high-performance lithium-air battery, Nature Chemistry, 4, 579-585
(2012)
4. Rechargeable Lithium/TEGDME-LiPF6/O2 battery, J. Electrochem. Soc., 3, A302A308 (2011)

96

Chapter 5
High Surface Area Graphene Nanoplatelet Based Paper as
Binder Free Self-Standing Electric Double-Layer Capacitor
Electrode

5.1 Abstract
High surface area carbon and graphitic materials are of interest for electrochemical
capacitor electrode fabrication. Electrode flexibility and enhancement of electrode
energy density are much researched subjects. Hybrid bilayer Grade-M-15/Grade-C-750
graphene nanoplatelet (GnP) paper combined with aqueous KOH electrolyte has been
investigated as electric double-layer capacitor (EDLC) electrode in this research. Cyclic
voltammetry (CV); impedance spectroscopy and galvanostatic charge/discharge
experiments have been carried out to determine the capacitive characteristics. The
hybrid bilayer paper displays excellent capacitive characteristics attributed to its low
equivalent series resistance (ESR). Individual capacitance contributions from Grade-M15 GnP and Grade-C-750 GnP have been determined and the potential role played by
each of the component in determining the capacitive characteristics of the composite
paper has been discussed.

97

5.2 Introduction
Self-standing

thin

film

architectures

are

considered

as

potentially

attractive

electrochemical capacitor electrodes. High power density, resulting from low equivalent
series resistance, is an advantage of thin film architectures. Another advantage of this
type of structure is that additional supporting current collector is not required for the
electro-active material. Various types of carbon nanotube (CNT) based thin films have
been investigated as electrochemical capacitor electrodes [1-7]. Some of these thin
films are self-standing and paper-like [4-7]; whereas, others are supported on a
substrate [1-3]. Multilayer aligned graphene nanosheet structures have attracted much
attention both as EDLC [8, 9] and as pseudocapacitor [10]. EDLC and pseudocapacitor
with excellent performance are often obtained through the hybridization of different
materials and with hierarchical structures as a result of optimization of electrode kinetics
and surface area [1, 5-7, 10-12]. Primarily three categories of carbon materials have
been investigated as EDLC electrode materials, high surface area activated carbons;
CNT; and graphene nanosheet based materials [22, 23]. A traditional dielectric based
capacitor delivers a capacitance value in the microfarad range [22]. CNT based EDLC
electrodes deliver a capacitance in the range of 15 to 135 F/gm and activated carbons
in the range of 40 to 140 F/gm [22]. Pristine graphene based EDLC electrodes deliver a
capacitance of about 100 F/gm [22]; however, hybridization with other carbon materials
may enhance the specific capacitance value [23]. Although activated carbons have high
surface area, small particle size and poor electrical conductivity restrict their capacitive
performance [22]. CNT possess higher electrical conductivity compared to activated
carbons; however, the surface area may not satisfy the requirement of an

98

electrochemical capacitor [22]. Graphene is also an attractive EDLC electrode material
both due to its high surface area and high electrical conductivity; however,
agglomeration of few layered graphene is a problem that needs to be resolved to attain
high effective surface area [22]. As already mentioned, hybrid structures of different
carbon materials such as graphene nanosheet-CNT [26] and graphene nanosheetcarbon spheres [12] have the potential to deliver high capacitance due to structural
optimization. Similarly, a combination of large platelet size of Grade-M-15 GnP and high
surface area of aggregates of sub-micron platelets of Grade-C-750 GnP could have the
potential to deliver high capacitance by taking advantage of their unique properties.
Such a combination could be realized by using a Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper as EDLC electrode where the Grade-M-15 particles act as potentially
highly electronically conductive substrate attributed to their large particle size and the
ability to hold together through Van der Waals force for potentially highly capacitive
Grade-C-750 particles attributed to their high surface area.

5.3 Methods
5.3.1 Fabrication of GnP Paper
Two types of GnP papers were investigated as EDLC electrodes in this research.
Grade-M-15 GnP (120-150 m2/gm surface area, 15 µm average particle size) paper was
fabricated according to the GnP-M(1) composition following the same method described
in chapter 3 (table 3-1). Grade-M-15/Grade-C-750 (750 m2/gm, aggregates of submicron platelets) hybrid bilayer GnP paper was prepared following the same method
described in chapter 4.

99

5.3.2 Structural Characterization of GnP Paper
The GnP papers were characterized with surface and internal structure imaging.
Scanning electron microscope (SEM) images of surface and the images of internal
structure by cutting with a focused ion beam (FIB) were captured with a Carl Zeiss
Auriga 39 microscope.

5.3.3 Electrical Conductivity of GnP Paper
Electrical conductivities of the GnP papers were measured by 4-point probe technique
with a Keithly 2400 Source Meter. The method for measuring electrical conductivity is
same as described in chapter 3.

5.3.4 Fabrication of Capacitor Cell
The method for the fabrication of capacitor cell is illustrated in figure 5-1. 0.5” diameter
disc electrodes were punched from the GnP papers. The electrodes were sandwiched
in between two stainless steel plates with a glass microfiber filter (Whatman) separating
the electrodes. Grade-C-750 layer faced the separator while investigating the Grade-M15/Grade-C-750 hybrid bilayer GnP paper as electrode. It is important to make sure that
the electrodes overlap on each other when they are sandwiched in between the
stainless steel plates. The plates were sandwiched by tightening with nut and bolts. Two
additional plates as shown in figure 5-1 were used as electrical connectors to the
alligator clips coming from the potentiostat. The assembly was immersed in electrolyte
just enough for the electrodes to be wet. The electrolyte was taken in a glass beaker
with diameter slightly larger than the width of the square stainless steel plates. The

100

assembly was kept upright in the beaker with the sandwiched plates holding the
electrodes at the bottom. The electrodes were placed towards the bottom part of the
sandwiched assembly so that the electrodes could be wet with minimal immersion. The
separator was pre-wet with electrolyte prior to sandwiching of the plates. Potassium
hydroxide, KOH (99.98%, metal basis, Alfa Aesar) was chosen as electrolyte salt. The
electrolyte was prepared by dissolving the salt in reverse osmosis water to the strength
of 6M. Aqueous electrolyte was chosen since organic electrolytes show poor ionic
conductivity and thus are not suitable to meet high power demand although their
operating voltage window is wider [13]. A two electrode configuration was chosen since
a two electrode system represents a real capacitor device [13].

101

(a)

(b)

(c)

(d)

(e)

(f)

To
alligator
clips

(g)

Figure 5-1: assembly of capacitor cell with hybrid bilayer GnP paper: (a) begin with
four sets of stainless steel plates and nut/bolts; (b-e) sandwich the electrodes in
between two stainless steel plates with a glass microfiber filter separating the
electrodes, C-750 GnP layer faces the separator; (f) connection of sandwiched
assembly to potentiostat alligator clips; (g) the assembly is put upright in a beaker with
electrolyte

5.3.5 Electrochemical Performance Evaluation
Three types of electrochemical tests were performed. Cyclic voltammetry was
performed in the potential range of 0-1 V at scan rates of 25 mV/sec; 100 mV/sec and 1
V/sec. Galvanostatic charge/discharge tests were performed in the potential range of 01 V at 1 A/gm current rate (normalized with respect to the weight of one electrode only).
Impedance spectrum was obtained in the frequency range of 10 5 Hz-10 mHz at 10 mV

102

amplitude. After assembly, a cell was rested for 30 minutes before beginning
electrochemical tests. A 30 minutes time was allowed in between two cyclic
voltammetry measurements at different rates.

5.4 Results and Discussions
5.4.1 Structural Characterization of GnP Paper
All structural characterization discussion remains same as discussed in chapter 3 for
Grade-M-15 GnP paper and chapter 4 for Grade-M-15/Grade-C-750 hybrid bilayer GnP
paper.

5.4.2 Electrical Conductivity of GnP Paper
The electrical conductivity values of the GnP papers remain same as discussed in
chapter 3 for Grade-M-15 and chapter 4 for Grade-M-15/Grade-C-750 hybrid bilayer
GnP papers. The Grade-M-15 GnP paper investigated in this research corresponds to
the composition GnP-M(1) (table 3-1, chapter 3) and the electrical conductivity value
should be read accordingly.

5.4.3 Cyclic Voltammetry of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP Paper
Electrode
The charge held on each side of the electrode interface in an EDLC depends on the
applied electrode potential [24]. This dependence can be measured directly by the ratio
of differential change in charge to that of the differential change in electrode potential.
This type of differential relationship give more resolved information about the electrode

103

than an integral information (i.e. the ratio of total charge held to that of the total applied
electrode potential). For example, presence of pseudocapacitance would result in a
peak due to higher charge transfer. This is why cyclic voltammetry (CV) is considered
as a valuable technique for studying double-layer behavior [24] where the electrode
surface is charged and then discharged at a certain potential rate within the potential
window of the electrolyte. Ideally, this type of charge storage is purely electrostatic [13].
The instantaneous current value is dependent on the applied potential perturbation. This
is called ‘capacitive current’. However, the current is independent of the total applied
electrode potential. The sign of the current should reverse immediately upon the
reversal of potential perturbation without any delay. Thus, an ideal double-layer
capacitive behavior is exhibited by a rectangular CV characteristic [13]. Movement of
ions in the bulk of the electrolyte is different from the movement inside the electrode.
The ease of charging and discharging of an electrode determines the time constant for
equilibration of current upon application of a potential perturbation [13]. Thus, materials
with high resistivity would not show a rapid current response and result in an oblique CV
trajectory rather than a rectangular profile [13].
CV profile of a two electrode EDLC with Grade-M-15/Grade-C-750 hybrid bilayer GnP
paper is shown in figure 5-2. It can be seen from figure 5-2 that the CV trajectory is
nearly rectangular without any obvious redox peaks indicating near-ideal double-layer
capacitive behavior at a scan rate of 100 mV/sec. At a scan rate as high as 1 V/sec, the
near-rectangular shape is still maintained without much distortion (figure 5-3). The
EDLC was cycled over several cycles. In figure 5-2 and figure 5-3, 9th and 10th cycle CV
trajectories were overlapped to show the stability of the EDLC behavior. Excellent

104

capacitive characteristics of the EDLC can be attributed to a rapid current response of
the electrodes due to several factors. This includes accessibility of electrolyte and fast
diffusion of electrolyte ions within the electrode structure, small equivalent series
resistance (ESR), and small charge transfer resistance [8, 11, 12, 14-19]. Practical
capacitors deviate from pure capacitance in that the capacitance is linked in series with
an element exhibiting ohmic behavior. This ohmic component is known as ESR [24]. All
of these characteristics can be attributed to either one or both of the components in the
hybrid bilayer paper electrode as will be revealed later in the discussion. Only a few
researches have investigated CV trajectory at a scan rate as high as 1 V/sec [8, 10]. A
near-ideal double-layer capacitive performance at a much lower scan rate has been
accepted as satisfactory in many prior researches [11, 14, 15, 17].

Figure 5-2: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C750 hybrid bilayer GnP paper electrode; 100 mV/sec scan rate; appearance of a
knee marked in with red arrow

105

Figure 5-3: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C750 hybrid bilayer GnP paper electrode; 1 V/sec scan rate; appearance of a knee
marked with red arrow
The CV characteristics of a carbon/carbon symmetric capacitor have been discussed in
literature [13]. Figure 5-4 shows the CV trajectory of an EDLC with Grade-M-15/GradeC-750 hybrid bilayer GnP paper electrode at a scan rate of 25 mV/sec. A peak appears
at the end of charge segment (indexed as ‘p’), in contrast to the trajectory which is
nearly rectangular. The peak is present in the CV trajectory at 100 mV/sec and 1 V/sec
scan rate as well. However, it is more clearly distinguishable at 25 mV/sec scan.
Capacitive current (level-off current) at 25 mV/sec scan rate is much lower compared to
100 mV/sec scan rate and 1 V/sec scan rate in addition to the fact that the resolution of
data is higher at low scan rate. This makes the peak stand out at the lower scan rate. A
peak at the end of charge cycle for carbon/carbon symmetric capacitor has been
reported in literature [13]; however, its significance has not been clearly discussed.

106

There might be another explanation to this phenomenon. Breakdown field strength of
dielectric materials for high-voltage devices has been a widely researched subject [28].
Effect of dielectric geometry such as thickness; porosity; grain size; impurity effects;
surface finish effects and temperature effects have been discussed. Figure 5-5 is the
CV trajectory of an electrolyte soaked glass microfiber separator pressed in-between
two stainless steel plates without any GnP electrodes. The electrolyte soaked separator
can be considered as equivalent to that of a traditional dielectric and this configuration
can be considered as a traditional capacitor. In this CV trajectory a knee appears
approximately around 600 mV indicating a breakdown. Insertion of a paper electrode
fabricated with 120-150 m2/gm surface area Grade-M-15 GnP (GnP-M(1) paper, table
3-1, chapter 3) at the interface of stainless steel plate and electrolyte soaked glass
microfiber separator doesn’t affect the breakdown voltage significantly (figure 5-6).
Insertion of a paper electrode fabricated with 120-150 m2/gm surface area Grade-M-15
GnP with a net higher surface area due to its higher thickness (GnP-M(2) paper, table
3-1, chapter 3) increases the breakdown voltage to a relatively higher limit (figure 5-7).
The effect of high surface area of electrode in enhancing the breakdown voltage limit is
clearly demonstrated from the CV trajectory of Grade-M-15/Grade-C-750 hybrid bilayer
GnP paper (figure 5-2). The breakdown voltage has been enhanced from 600 mV for an
assembly with no electrode to almost 800 mV for an assembly with high surface area
bilayer electrode. High breakdown field strength is an important characteristic of EDLC
compared to traditional capacitors which has been acknowledged in literature [27] but
not clearly illustrated as done in this work.

107

p

Figure 5-4: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15/Grade-C750 hybrid bilayer GnP paper electrode; 25 mV/sec scan rate; appearance of a knee
marked with red arrow; ‘p’ stands for peak

Figure 5-5: overlapped 9th and 10th cycle CV trajectory of a SS-plate blank EDLC
cell; 100 mV/sec scan rate; appearance of a knee marked with red arrow

108

Figure 5-6: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15 GnP paper
electrode (GnP-M(1)); 100 mV/sec scan rate; appearance of a knee marked with red
arrow

Figure 5-7: overlapped 9th and 10th cycle CV trajectory of a Grade-M-15 GnP paper
electrode (GnP-M(2)); 100 mV/sec scan rate; appearance of a knee marked with red
arrow

109

5.4.4 Capacitance Determination of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP
Paper Electrode
Galvanostatic charge/discharge experiments were carried out at different current rates
and capacitance was measured from the discharge part of the charge/discharge
trajectory. A charged EDLC electrode may show a current dependent ohmic potential
drop upon discharge [24]. This appears as a break (also known as iR drop) in the
discharge trajectory at the beginning of the discharge process determined by the ESR
of the electrode [25]. An EDLC with Grade-M-15/Grade-C-750 hybrid bilayer GnP paper
electrode was charged and discharged over several cycles at different current rates to
determine the rate dependence of capacitance; the stability of the electrode; and to
investigate the discharge trajectory.
Figure 5-8 and figure 5-9 show the galvanostatic charge/discharge trajectory of an
EDLC with Grade-M-15/Grade-C-750 hybrid bilayer GnP paper electrode at 10th cycle
and 110th cycle at 1A/g (normalized with respect to the weight of a single electrode)
current rate. The discharge trajectory is almost a straight-line. The discharge trajectory
doesn’t show any iR potential drop at the beginning of the discharge process which
implies that the ESR of the paper electrode is low. This can be related well with the
retention of a rectangular CV trajectory at a high scan rate determined by the small ESR
of the electrode. The capacitance was calculated from the discharge curve using the
following equations [8, 15, 20].

(1)
(2)

110

Where,

is the capacitance of cell in farads (F);

(A) and

is the discharge current in ampere

is the slope of the discharge curve in volts/sec.

is the specific capacitance of an electrode in farads/gm (F/gm) of active material;
is the weight of active material in an electrode.

1

Cell potential (Volts)

0.8

0.6

0.4

0.2

0
250

255

260

265

270

275

280

285

Discharge time (Seconds)

Figure 5-8: galvanostatic discharge/charge trajectory of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 10th cycle; 1 A/gm current rate

111

Figure 5-9: galvanostatic discharge/charge profile of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 110th cycle; 1 A/gm current rate

Equation 1 and equation 2 yield a specific capacitance value of 27 F/gm both at 10 th
cycle and at 110th cycle at 1A/gm current rate. No capacitance reduction over 100
cycles implies that the hybrid bilayer paper electrode structure is highly stable. Figure 510 and figure 5-11 show the galvanostatic charge/discharge trajectory of an EDLC with
Grade-M-15/Grade-C-750 hybrid bilayer GnP paper electrode at 10 th cycle and 110th
cycle at 10 A/gm (normalized with respect to the weight of a single electrode). Again,
the discharge trajectory is almost a straight-line. The discharge trajectory doesn’t show
any iR potential drop at the beginning of the discharge process as an effect of tenfold
current rate. Equation 1 and equation 2 yield a specific capacitance value of 23 F/g at
10th cycle and at 110th cycle at 10A/g current rate. This result implies a capacitance

112

reduction of only 15% as a result of tenfold increase of current rate. The current rates,
capacitance values and cycle performances are listed in table 5-1.

Figure 5-10: galvanostatic discharge/charge trajectory of a Grade-M-15/Grade-C750 hybrid bilayer GnP paper electrode at 10th cycle; 10 A/g current rate

113

Figure 5-11: galvanostatic discharge/charge profile of a Grade-M-15/Grade-C-750
hybrid bilayer GnP paper electrode at 110th cycle; 10 A/g current rate

Current rate (A/gm)

10th cycle capacitance
(F/gm)

110th cycle capacitance
(F/gm)

1
10

27
23

27
23

Table 5-1: rate performance, cycle performance and capacitance values of Grade-M15/Grade-C-750 hybrid bilayer GnP paper electrode; electrode weight for 1A/g
measurement: 5.3 mg; electrode weight for 10A/g measurement: 5.1 mg

5.4.5 Capacitance Determination of Grade-C-750 GnP
In section 5.4.4, the capacitance of hybrid bilayer paper electrode has been determined.
In this section the capacitance contributions from individual components in the hybrid
paper (i.e. Grade-M-15 GnP and Grade-C-750 GnP) have been separated. Grade-C750 GnP having a much higher surface area compared to Grade-M-15 GnP, is likely to

114

be the major contributor of capacitance in the hybrid paper. The capacitive current from
a GnP-M(1) paper electrode as revealed by its CV trajectory (section 5.4.3) is one order
of magnitude lower compared to that of the hybrid paper electrode. The CV trajectory of
a GnP-M(1) paper electrode can also be considered as the CV trajectory of the bottom
layer of a Grade-M-15/Grade-C-750 hybrid bilayer GnP paper. The charge/discharge
trajectory of a GnP-M(1) paper electrode reveals that the discharge time is only a
fraction of that for the hybrid paper electrode. Figure 5-12 is a 10th cycle
charge/discharge trajectory of a GnP-M(1) paper electrode at 1A/g (normalized with
respect to the weight of a single electrode) current rate. The current rates and
capacitance values are listed in table 5-2. The hybrid bilayer paper electrode as a whole
remains stable over hundreds of cycles as already seen in table 5-1. Thus, it is
reasonable to assume that the capacitance value of Grade-C-750 GnP alone estimated
from 10th cycle would remain stable over hundreds of cycle. If the total capacitance of a
GnP-M(1) paper electrode is deducted from the total capacitance of a Grade-M15/Grade-C-750 hybrid bilayer paper electrode and normalized with respect to the
weight of Grade-C-750 GnP, it would yield the specific capacitance of Grade-C-750
GnP alone. A comparison of the capacitive performances of hybrid paper and GnP-M(1)
paper from table 5-1 and table 5-2 at 1A/g current rate would yield a specific
capacitance of 66 F/gm for Grade-C-750 GnP. It has already been seen that the
capacitance of a hybrid paper reduces only by 15% as a result of tenfold increase of
current rate. This implies that Grade-C-750 GnP would retain a specific capacitance of
more than 50 F/gm at a current rate of 10A/gm. However, it has been seen in chapter 3
and chapter 4 that the weight of hybrid paper electrode and GnP-M(1) paper electrode

115

slightly varies within a range due to the fabrication process and the fact that the paper
electrodes are not pressed. The capacitance of Grade-C-750 GnP is deduced from the
capacitance of a Grade-M-15/Grade-C-750 hybrid bilayer GnP paper by using a GradeM-15 GnP-M(1) paper as the reference. Thus, the actual value of the specific
capacitance of Grade-C-750 GnP may be slightly different from what is calculated in
table 5-2.

Figure 5-12: galvanostatic discharge/charge trajectory of a Grade-M-15 GnP paper
electrode at 10th cycle; 1 A/g current rate
Current rate (A/gm)

1

10th cycle capacitance
(F/gm GnP-M(1) paper
electrode)
3

10th cycle capacitance (F/gm
Grade-C-750 GnP)
66

Table 5-2: current rate and capacitance values of GnP-M(1) paper electrode and
Grade-C-750 GnP; GnP-M(1) paper electrode weight: 3.3 mg

116

5.4.6 Impedance Spectroscopy of Grade-M-15/Grade-C-750 Hybrid Bilayer GnP
Paper Electrode
Impedance spectroscopy is a well-recognized method for understanding the
fundamentals of electrochemical capacitor materials [11, 17, 20, 21]. The impedance
spectrum has a high frequency intercept on the x-axis. This is followed by a semicircle
and straight line at the end of the semicircle. The high frequency intercept represents
the equivalent series resistance (ESR) of the electrode which includes intrinsic ionic
resistance of the electrolyte; intrinsic resistance of active materials and contact
resistance at active material/current collector interface. The diameter of the semicircle is
a parallel combination of charge transfer resistance (Rct) and double-layer capacitance
at the electrode/electrolyte interface. Diffusion resistance of ion in the electrolyte is
represented by the slope of the straight line that follows the semicircle. Relatively higher
slope closer to 90° is indicative of capacitive behavior; whereas, a relatively less slope
indicates diffusion control. Figure 5-13 is an impedance spectrum of a freshly prepared
EDLC cell with Grade-M-15/Grade-C-750 hybrid bilayer GnP paper electrode. The ESR
value is only 0.053 Ω. This explains why the CV profile remains almost rectangular
representing near-ideal double-layer capacitive behavior even at a scan rate as high as
1 V/sec.

117

ESR

Rct

Figure 5-13: impedance spectroscopy of a freshly prepared EDLC cell with GradeM-15/Grade-C-750 hybrid bilayer GnP paper electrode; the inset shows the initial
part of the impedance spectrum

Figure 5-14 is an impedance spectrum of a GnP-M(1) paper electrode (table 3-1,
chapter 3). Compared to the impedance spectrum of a Grade-M-15/Grade-C-750 hybrid
bilayer GnP paper (figure 5-13), the GnP-M(1) paper electrode has almost twice the
ESR (0.103 Ω). The diameter of the semicircle is also much higher compared to GradeM-15/Grade-C-750 hybrid bilayer GnP paper electrode (figure 5-13). Higher ESR and
higher charge transfer resistance of GnP-M(1) paper corroborates with its relatively poor
capacitive performance compared to a hybrid bilayer GnP paper electrode.

118

ESR

Rct

Figure 5-14: impedance spectroscopy of a freshly prepared EDLC cell with GnPM(1) paper electrode; the inset shows the initial part of the impedance spectrum

5.5 Conclusions
High surface area Grade-C-750 GnP aggtegate shows excellent capacitive performance
as an EDLC electrode material. The capacitance is almost independent of potential
scan rate. High capacitive performance of Grade-C-750 GnP can be attributed to its low
ESR and charge transfer resistance. Compared to high surface area Grade-C-750 GnP,
the capacitive property of relatively low surface area Grade-M-15 GnP is negligible.
However, a paper like structure fabricated with large particle size Grade-M-15 GnP can
be used as the substrate for high surface area Grade-C-750 GnP. This results in an
enhancement of the energy density of an EDLC electrode by doing away with
electrochemically inactive and much higher weight metal current collectors.

119

5.6 Acknowledgements
The author acknowledges the support of Composite Materials and Structures Center
(CMSC) research staff at Michigan State University in accomplishing this work.

120

REFERENCES

121

REFERENCES

1. Binder-free manganese oxide/carbon nanomaterials thin film electrode for
supercapacitors, ACS Appl. Mater. Interfaces, 3, 4185-4189 (2011)
2. Electrochemical characterization of carbon nanotubes as electrode in electrochemical
double-layer capacitors, Carbon, 40, 1193-1197 (2002)
3. Layer-by-layer assembly of all carbon nanotube ultrathin films for electrochemical
applications, J. Am. Chem. Soc., 131, 671-679 (2009)
4. Compact-designed supercapacitors using free-standing single-walled carbon
nanotube films, Energy Environ. Sci., 4, 1440-1446 (2011)
5. High-power supercapacitor electrodes from single-walled carbon nanohorn/nanotube
composite, ACS Nano, 5 (2), 811-819 (2011)
6. Nanocomposite electrodes for high-performance supercapacitors, J. Phys. Chem.
Lett. , 2, 655-660 (2011)
7. Studies on mechanism of carbon nanotube and manganese oxide nanosheet selfsustained thin film for electrochemical capacitor, Solid State Ionics, 181, 1690-1696
(2010)
8. Multilayered nano-architecture of variable sized graphene nanosheets for enhanced
supercapacitor electrode performance, ACS Applied Materials & Interfaces, 2 (8), 22932300 (2010)
9. Liquid-mediated dense integration of graphene materials for compact capacitive
energy storage, Science, 341, 534 (2013)
10. Multilayered nanoarchitecture of graphene nanosheets and polypyrrole nanowires
for high performance supercapacitor electrodes, Chem. Mater., 22, 5667-5671 (2010)
11. Electrochemical properties of graphene nanosheet/carbon black composites as
electrodes for supercapacitors, Carbon, 48, 1731-1737 (2010)
12. A self-assembled hierarchical nanostructure comprising carbon spheres and
graphene nanosheets for enhanced supercapacitor performance, Energy Environ. Sci.,
4, 4504 (2011)
13. Carbon materials for the electrochemical storage of energy in capacitors, Carbon,
39, 937-950 (2001)

122

14. Supercapacitance from cellulose and carbon nanotube nanocomposite fibers, ACS
Appl. Mater. Interfaces, 5 (20), 9983-9990 (2013)
15. Low temperature exfoliated graphenes: vacuum-promoted exfoliation and
electrochemical energy storage, ACS Nano, 3 (11), 3730-3736 (2009)
16. Graphene-based supercapacitor with an ultrahigh energy density, Nano Lett., 10
(12), 4863-4868 (2010)
17. Graphene based ultracapacitors, Nano Lett., 8 (10), 3498-3502 (2008)
18. Asymmetric supercapacitors based on graphene/MnO2 and activated carbon
nanofiber electrodes with high power and energy density, Adv. Funct. Mater. , 21, 23662375 (2011)
19. Compact-designed supercapacitors using free-standing single-walled carbon
nanotube films, Energy Environ. Sci., 4, 1440 (2011)
20. Electrochemical characteristics and impedance spectroscopy studies of carboncarbon supercapacitors, Journal of The Electrochemical Society, 150 (3), A292-A300
(2003)
21. Hydrothermal synthesis of MnO2/CNT nanocomposite with a CNT core/porous
MnO2 sheath hierarchy architecture for supercapacitors, Nanoscale Research Letters,
7:33 (2012)
22. Nanostructured carbon for energy storage and conversion, Nano Energy, 1, 195220 (2012)
23. Graphene for energy conversion and storage in fuel cells and supercapacitors,
Nano Energy, 1, 534-551 (2012)
24. Electrochemical supercapacitors: scientific fundamentals and technological
applications, chapter 6, B.E. Conway, Springer (1999)
25. Preparation of functionalized graphene sheets by a low temperature thermal
exfoliation approach and their electrochemical supercapacitive behaviors,
Electrochimica Acta, 55, 3897-3903 (2010)
26. Self-assembled graphene/carbon nanotube hybrid films for supercapacitors, J.
Phys. Chem. Lett. , 1, 467 (2010)
27. Electrochemical double-layer capacitors using carbon nanotube electrode
structures, Signorelli, R., Proceedings of the IEEE, 97.11, 1837-1847 (2009)

123

28. Effects of electrode geometry and dielectric media on dielectric breakdown strength
of polycrystalline alumina, Master’s Thesis, Youngjin Kim, The University of Utah (2010)

124

Chapter 6
Manganese Oxides for Electrochemical Energy Storage
Applications: Effect of Graphene Nanoplatelet Substrate and
Polymer Surfactant

6.1 Abstract
Manganese oxides (MnOx) have been synthesized on Grade-M-15 (120-150 m2/gm
surface area, 15 µm average particle size) graphene nanoplatelet (GnP) surface though
a wet-chemical method by the redox reaction of two manganese salts (KMnO4 and
MnSO4.H2O). Ratio of reagent salts has been varied to obtain different morphology and
crystal phases of MnO2. It has been found that GnP substrate plays a significant role in
preventing the agglomeration of MnO2 while retaining the crystal phase. Two types of
MnO2 have been synthesized, birnessite-MnO2 and γ-MnO2. Composites of carbon
materials with these MnO2 crystal phases have the potential to find applications as high
performing pseudocapacitor electrodes. Reduction of agglomeration is of significance
for reducing transport limitations in the desired applications. Further modification of the
morphology and crystal phase of manganese oxide has been achieved through the
incorporation of a polyethylenimine (PEI) polymer surfactant in the synthesis system. A
change in morphology from a continuous network-like structure to nanoparticle clusters
has been observed as an effect of polymer surfactant introduction in the synthesis
system. The polymer surfactant also affects the crystal phase of manganese oxide by
reducing the oxidation state of manganese to yield Mn3O4. Composites of graphitic

125

materials with Mn3O4 have the potential to find applications as high performing Lithiumion battery anode. The synthesis process of Mn3O4 through the PEI mediated method
developed in this research is believed to be much simpler compared to other existing
methods for synthesizing the same.

6.2 Introduction
GnP with flat platelet-like morphology and high open surface area may be an ideal
substrate for metal nanoparticles deposition. Substrate plays an important role in
reducing the agglomeration of metal nanoparticles. High electrical conductivity,
oxidation and corrosion resistance of GnP might be of interest for some applications as
well. In this research surface decoration of GnP with manganese oxides through a
bottom up approach has been investigated. Out of many routes for manganese oxides
synthesis one is to synthesize through the redox reaction of two manganese salts.
Synthesis of nanowires and nanorods has been investigated through this route by Wang
et al. [1]. A wet-chemical method based on the redox reactions of MnSO4.H2O and
KMnO4 or (NH4)2S2O8 has been followed to synthesize MnO2 single crystal nanowires
or nanorods. In this type of synthesis process, one of the reagents acts as an oxidizing
material for the other reagent to generate manganese oxide crystals. In the research by
Wang et al. [1], either KMnO4 or (NH4)S2O8 has been used as oxidizing agent for
MnSO4.H2O. The oxidizing agent doesn’t always have to be a salt. Kijima et al. [2] have
synthesized MnO2 by ozone oxidation of manganese salt MnSO 4.H2O. In this chapter
the synthesis of manganese oxides through a wet-chemical redox reaction of
manganese salts has been tried with GnP as substrate for metal oxide deposition.

126

Surfactants, when introduced in the system, have been known to act both as a reducing
agent for a salt to generate manganese oxides crystals as well as to control crystal size,
morphology and phase [4-9]. Tuning of crystal morphology and crystal phase has their
implications in the desired applications and the optimization of reaction kinetics through
the elimination of transport limitations. Polyethylenimine (PEI) is a cationic polymer that
can also help disperse hydrophobic GnP in an aqueous medium [10]. The effect of PEI
on tuning the morphology of manganese oxide crystal has been investigated.

6.3 Method
6.3.1 Materials
Manganese sulfate monohydrate (MnSO4.H2O) (ReagentPlus>99%) and potassium
permanganate (KMnO4) (ACS reagent>99%) were purchased from Sigma-Aldrich.
Branched polyethylenimine was also purchased from Sigma-Aldrich (average Mw
25,000 by LS, average Mn 10,000 by GPC). Grade-M-15 (120-150 m2/gm surface area,
15 μm average particle size) GnP was used as the substrate for manganese oxides
deposition.

6.3.2 Synthesis of Manganese Oxides without PEI
First, 1.0 gm GnP is added to 150 ml reverse osmosis water and sonicated for 2
minutes. This is followed by heating to 80°C under constant stirring in the container
attached to the recirculating heater as shown in figure 6-1. The container has a water
jacket. The heater pumps heated water into the container jacket with an inlet pipe and
reverses it back to the heater through an outlet pipe continuously. The heat is

127

transferred through the water jacket to the water stirring inside the container. The
temperature of the heater is set slightly above 80°C to account for the heat loss in the
process of heat transfer. The water container is placed on a magnetic stirrer plate. The
plate stirs a magnetic stirrer bar which is placed at the bottom of the water container
and thus stirring the water at a uniform speed. MnSO 4.H2O and KMnO4 are dissolved
separately in 25 ml reverse osmosis water by stirring and heating to 80°C. Chapter 1
has discussed different oxidation states and crystal phases of manganese oxides.
Synthesis of manganese oxides in air from the redox reaction of two manganese salts in
neutral aqueous solution is likely to yield the crystal phase with highest oxidation state
+4; which is manganese (IV) oxide (MnO2). Synthesis of other manganese oxides with
lower oxidation states would require special conditions in the system such as presence
of highly reducing agents [11]; wet-chemical synthesis over a long time [12] or postsynthesis treatments of MnO2 such as treatment at high temperature in air or in
reducing gases such as hydrogen and methane [13]. MnO2 may have different
polymorphs [14] depending upon synthesis conditions such as time, temperature and
concentration of reagents [1]; presence of acids in the system [2]; and the nature of the
anion in the acid molecule [2]. Since the objective of this research is not to gain a
detailed understanding of different crystal phases of manganese oxides but to evaluate
the potential of GnP as a substrate for reduction of metal oxide agglomeration, only two
compositions of reagents are chosen that are likely to yield MnO2 polymorphs. The
compositions along with synthesis conditions are listed in table 6-1. For ‘Comp-a’ the
ratio is adopted from Cheng et al. [3]. For ‘Comp-b’ the ratio corresponds to the

128

stoichiometric ratio of KMnO4 and MnSO4 (KMnO4:MnSO4=2:3=0.66:1) that yields MnO2
according to the following equation.

2KMnO4 + 3MnSO4 + 2H O → 5MnO + K SO4 + 2H SO4

(1)

Same protocol was followed for each synthesis process. First, MnSO 4.H2O solution
was added drop-wise with a glass pipet to the GnP dispersion. The significance of slow
addition is that it prevents the solution temperature from deviating from 80°C. This is
followed by slow addition of KMnO4 solution with a glass pipet. After addition of both salt
solutions, the container is covered with a lid and the stirring continues for 1 hour. After 1
hour, the stirrer plate and the heater are turned off and the solution is allowed to cool for
2 hours. After cooling, the solution is filtered with a filter paper to collect the solid
residue; a composite of GnP and manganese oxides. The composite is washed multiple
times with reverse osmosis water. This is followed by a pre-drying step where the
composite is dried in air under a hood. This is followed by a second drying step in a
vacuum oven at 120°C for about 12 hours. The vacuum oven is allowed to cool naturally
by turning off the heater before it is filled with air and the composite is taken out.
Manganese oxides without GnP substrate were also prepared following the same
protocol for comparison with the composites.

129

Grade
Comp
-a
Comp
-b
MnOxa
MnOxb

Water
Volume
(ml.)
150

KMnO4 MnSO4.H2O
(gm)
(gm)

Molar ratio
(KMnO4:MnSO4.H2O)

GnP
(gm)

Temperatu
-re (°C)

1.1

0.4

2.9:1

1

80

150

0.25

0.4

2:3=0.66:1

1

80

150

1.1

0.4

2.9:1

0

80

150

0.25

0.4

2:3=0.66:1

0

80

Table 6-1: composites and manganese oxides; MnOx-a refers to the synthesis of
manganese oxide without GnP substrate corresponding to Comp-a; same
explanation for MnOx-b

Recirculating
chiller/heater

Container for
composite synthesis
with water jacket

Stirrer plate

Figure 6-1: setup for GnP-manganese oxides composites synthesis

130

6.3.3 PEI Mediated Manganese Oxide Synthesis
The synthesis process remains same except for the preparation of GnP/PEI/water
dispersion. First, 1.0 gm GnP is sonicated for 2 minutes in 125 ml. reverse osmosis
water. PEI is dissolved in 25 ml reverse osmosis water; added to the GnP solution and
stirred for 12 hours to prepare the dispersion. This dispersion is then heated to 80°C as
described before and the rest of the steps for composite synthesis remain same. To
investigate the effect of PEI on manganese oxide crystal phase and morphology, the
composition corresponding to ‘Comp-a’ (table 6-1) was chosen only. This is designated
as ‘Comp-c’ in table 6-2.
Grad
e
Comp
-c

Water
Volume
(ml.)
150

KMnO4
(gm)

MnSO4.H2O
(gm)

1.1

0.4

Molar ratio
(KMnO4:MnSO4.H
2O)
2.9:1

GnP
(gm)

PEI
(gm)

Temperature (°C)

1

2

80

Table 6-2: composites synthesized through PEI mediated method
6.3.4 Characterization of Composites
Crystal structures of the synthesized manganese oxides and GnP-manganese oxides
composites were determined with powder X-Ray Diffraction (XRD) (Bruker).
Morphologies of the manganese oxides and GnP-manganese oxides composites were
observed with Scanning Electron Microscopy (SEM) (Carl Zeiss Auriga 39).

6.4 Results and Discussion
6.4.1 Synthesis of Manganese Oxides without PEI
Figure 6-2 shows an SEM image of composite ‘a’ (Comp-a) from table 6-1. The low
magnification image 6-2(a) indicates the presence of a continuous network-like structure

131

deposited on the GnP surface. However, the high magnification image 6-2(b) reveals
that the structure is not only continuous and network-like but also has a certain depth to
it and is protruding vertically from the GnP surface.

(a)

Figure 6-2: SEM images of ‘Comp-a’ from table 6-1: (a) low magnification image
shows presence of a continuous network-like structure on GnP surface; (b) high
magnification image shows that the structure grows vertically on GnP surface and
forms a continuous structure

132

Figure 6-2 (cont’d)
(b)

Figure 6-3 and figure 6-4 are the XRD patterns of composite ‘a’ (Comp-a) and MnOx-a
from table 6-1 respectively. A survey of literature reveals that the XRD patterns have
diffraction peaks characteristic of birnessite-MnO2 [15-20]. Birnessite-MnO2 has
monoclinic structure which consists of 2-D, edge-shared MnO6 octahedral layers with K+
cations and water molecules in the interlayer spaces [15]. The peaks in-between 2θ
20°-30°, 50°-60°, and 70°-80° in figure 6-3 arise from the GnP substrate [27]. Other
diffraction peaks at 2θ 12°, 37°, and 66° in figure 6-3 can be indexed as (001), (111),
and (020) reflections of monoclinic birnessite-MnO2 [17]. In figure 6-4 all of these

133

diffraction peaks from birnessite-MnO2 are visible. Birnessite-MnO2 also shows a peak
at about 25° [15] which is visible in figure 6-4 and might have overlapped with GnP
diffraction peak in figure 6-3. Note that the graphitic peaks from GnP are clearly visible
although some literature have reported disappearance of substrate peaks after the
deposition of manganese oxide [16]. In figure 6-4 apparently there is also a broad
diffraction peak in-between 2θ 50°-60° which can be related to γ-MnO2 [21-23]. It is not
clear why this diffraction peak has appeared in the spectrum. All other peaks, as
discussed already, can be related to birnessite-MnO2 only.

Figure 6-3: XRD of composite ‘a’ (Comp-a) from table 6-1; diffraction peaks from
birnessite- MnO2 are marked with asterisks; inset shows the complete XRD spectrum

134

Figure 6-4 XRD of MnOx-a from table 6-1; diffraction peaks from birnessite-MnO2 are
marked with asterisks
Apparently, the diffraction peaks in both figure 6-3 and figure 6-4 have similar width and
have the same 2θ values. Peak broadening may take place due to crystallite size
reduction or a combination of residual tensile and compressive stress in the material
[28]. Presence of uniform tensile or compressive stress would result in a shift of
diffraction peak positions only [28].
Processes for manganese oxide synthesis on carbon substrate have been discussed in
literature [3, 29]. The synthesis process adopted by Cheng [3] has been used in this
research. However, this method doesn’t explain well the mechanism by which
manganese oxides particles attach themselves to the carbon substrate. Delocalized πelectrons on carbon surface are capable of undergoing the following reaction [30]:

+ 2H O →

H O + OH

(2)
135

Ma et al. [29] and Li et al. [31] have synthesized MnO2 on carbon nanotubes through
the direct reduction of MnO4 – with carbon nanotubes in the presence of acids (H+ ions).
In this type of synthesis method a reducing agent MnSO 4 as adopted by Cheng [3] and
as done in this research has not been used. Instead, carbon nanotube itself has acted
as the reducing agent (i.e. the electron donor) to undergo the following reaction [29]:

MnO4 -+ 4H++3e- (from carbon) → MnO2 + 2H2O

(3)

Π-electrons, after complexation with water, are capable of attracting negative ions such
as MnO4 – (from KMnO4); which are then reduced to MnO2 as shown in reaction (3). In
the synthesis system adopted by Cheng [3] and in this research, in the absence of acids
(i.e. H+ ions), MnSO4 acts as the reducing agent for MnO4– ions after they
electrostatically attach themselves to positively charged π-complexes. To further verify
that MnO2 doesn’t deposit through a reduction by the graphene basal plane surface, the
same composite synthesis reactions in table 6-1 were carried out without any MnSO4
under identical conditions. The SEM images (figure 6-5 and figure 6-6) of GnP surface
after undergoing the reactions do not reveal the presence of any substantial
depositions. Thermogravimetric analysis (figure 6-7 and figure 6-8) reveal that the TGA
trajectory is that of clean GnP, except that the decomposition temperature has been
lowered. The disappearance of diffraction peak from CNT after the deposition of MnO 2
through the direct reduction of MnO4

–

ions has been reported in literature [16]. The

researchers have concluded that this type of disappearance happens due to the coating
effect of MnO2. In the work by Ma et al. the intensity of diffraction peak from CNT has
been reduced along with a broadening of peak [29]. However, in the synthesis method
adopted here, the graphite peak is clearly visible in the composite. These findings may

136

suggest that a direct reduction of MnO4- ions to synthesize MnO2 damages the structure
of the substrate as the substrate itself acts as an electron donor; whereas, the reduction
method by MnSO4 preserves the substrate characteristics.

Figure 6-5: SEM image of GnP surface after reaction of GnP with KMnO 4 only
(without MnSO4 as a reducing agent) corresponding to ‘Comp-a’ in table 6-1

137

Figure 6-6: SEM image of GnP surface after reaction of GnP with KMnO4
only (without MnSO4 as a reducing agent) corresponding to ‘Comp-b’ in
table 6-1

Figure 6-7: TGA of GnP after reaction of GnP with KMnO4 only (without MnSO4 as a
reducing agent) corresponding to ‘Comp-a’ in table 6-1

138

Figure 6-8: TGA of GnP after reaction of GnP with KMnO4 only (without MnSO4 as a
reducing agent) corresponding to ‘Comp-b’ in table 6-1

Figure 6-9 is an SEM image of MnOx-a from table 6-1. The structure agrees with
previously reported crumpled flower like structure of birnessite-MnO2 made up of
nanoribbons or nanoflakes [16, 18, 20]. It is clear that when the same birnessite-MnO2
is developed on a GnP substrate (figure 6-2), instead of crumpling up (figure 6-9), the
nanoribbons are uniformly distributed on GnP surface developing a continuous network
like structure. As discussed already in chapter 1, materials tend to reduce their surface
energy by forming agglomerates. In the presence of GnP substrate, MnO2 nucleation
takes place at π-sites followed by growth along the GnP surface to form a continuous
structure. This attenuates their tendency to form crumpled agglomerates for the
reduction of surface energy while retaining the crystal phase. Composites of GnP and
birnessite-MnO2 have the potential to exhibit high pseudocapacitive performance [1518, 20]. Composites of MnO2 with carbon black (CB) and carbon nanotube (CNT) have

139

delivered capacitance values of 186 F/g at 2 mV/sec scan rate [15] and 223 F/g at 10
mV/sec scan rate [16] respectively. Composites of MnO2 with onion-like carbon and
graphene foam have delivered capacitance values of 177.5 F/g at a current rate of 2 A/g
[18] and 560 F/g at a current rate of 0.2 A/g [20]. Reduction of agglomeration through
the deposition on GnP surface makes for easy access of electrolyte to MnO2 surface,
shortens diffusion length of electrolyte ion as well as may enhance the overall electrical
conductivity of the electrode compared to bare MnO 2 attributed to the high electrical
conductivity of GnP substrate.

Figure 6-9: SEM image of ‘MnOx-a’ from table 6-1

140

Figure 6-10 shows an SEM image of composite ‘b’ (Comp-b) from table 6-1. Image 610(a) indicates the presence of two different types of structures deposited on the GnP
surface. It is revealed that they are either flat disc-like (figure 6-10(b)) structures or a
combination of acicular and flat disc-like structures (figure 6-10(c)).

(a)

Figure 6-10: SEM images of ‘Comp-b’ from table 1: (a) presence of two different
structures (left and right); (b) flat disk-like structure; (c) combination of acicular and
flat disk-like structure

141

Figure 6-10 (cont’d)
(b)

(c)

142

Figure 6-11 and 6-12 are the XRD of composite ‘b’ (Comp-b) and MnOx-b from table 61. A survey of literature reveals that both diffraction spectra have diffraction peaks
characteristic of γ-MnO2 [21-23]. γ-MnO2 is an intergrowth between rutile-type pyrolusite
and ramsdellite (α-MnO2) [21]. Peaks at about 2θ 22°, 36°, 41°, and 55° in figure 6-11
are reflections from (120), (131), (300), and (160) planes of γ-MnO2 [23]. Additional
peaks in-between 2θ 20°-30°, 50°-60°, and 70°-80° in figure 6-11 arise from the GnP
substrate [27]. The same γ-MnO2 peaks are also present in figure 6-12. In addition,
there is a peak at about 2θ 66° from (003) plane in figure 6-12, which doesn’t appear in
the composite. Difference in peak width, especially for the peak at 2θ about 22° is also
observed. It has already been mentioned that peak broadening may take place both due
to a change in crystallite size and through the introduction of stress. Thus, the crystal
structure may have some difference in the composite from that of bare γ-MnO2;
however, most the structural characteristics remain same.

143

Figure 6-11 XRD of composite ‘b’ (Comp-b) from table 6-1; diffraction peaks from γMnO2 are marked with asterisks; inset shows the complete XRD spectrum

Figure 6-12 XRD of MnOx-b from table 6-1; diffraction peaks from γ-MnO2 are
marked with asterisks

144

Figure 6-13 is an SEM image of MnOx-b from table 6-1. γ-MnO2 has been reported to
have rod or filament like structure (i.e. high aspect ratio) [21-23], which is in agreement
with the aggregate of acicular structures that is observed in MnO x-b (figure 6-13). It
seems that GnP substrate has helped disperse the acicular aggregate (figure 6-10 (c)),
although the origin of the flat disc-like structure is not well understood. Composite of
GnP and γ-MnO2 has the potential to exhibit high pseuodocapacitive performance [22].
A composite of γ-MnO2 and CNT has delivered a capacitance value of 292 F/g at a
scan rate of 5 mV/sec [22].

Figure 6-13: SEM image of ‘MnOx-b’ from table 6-1

145

6.4.2 Polyethylenimine (PEI) Mediated Manganese Oxides Synthesis
6.4.2.1 Morphology of polyethylenimine (PEI) mediated manganese oxides
Figure 6-14 is an SEM image of composite ‘c’ (Comp-c) from table 6-2. Figure 6-14(a)
shows that the morphology of the deposits on GnP surface has changed drastically
(from figure 6-2) in presence of PEI. Islands of clustered particles are visible on GnP
surface. High magnification images (figure 6-14(b) and 6-14(c)) reveal that the clusters
are made up of small particles in the order of 20 nm. At low magnification the particles
appear to be spherical; however, at high magnification (figure 6-14(c)) it is observed that
the particles are nearly cubic or pyramid shaped.
(a)

Figure 6-14: SEM images of ‘Comp-c’ from table 6-2: (a) low magnification image of
islands of clustered particles on GnP surface; (b, c) high magnification images of
islands of clustered particles on GnP surface

146

Figure 6-14 (cont’d)
(b)

(c)

147

6.4.2.2 XRD of polyethylenimine (PEI) mediated manganese oxides
Figure 6-15 is the XRD of composite ‘c’ (Comp-c) from table 6-2. A survey of literature
reveals that the XRD pattern has several peaks that agree well with Mn3O4 (II,III) [8, 11,
12, 24-26, 31]. The peaks in-between 2θ 20°-30°, 50°-60°, and 70°-80° in figure 6-15
arise from the GnP substrate [27]. Rest of the peaks can be attributed to various planes
of Mn3O4 (II,III). Mn3O4 (II,III) is a manganese oxide in multivalent state. It is a
combination of MnO and Mn2O3.

Figure 6-15: XRD of composite ‘c’ (Comp-c) from table 6-2; diffraction peaks from
Mn3O4 are marked with asterisks; inset shows the complete XRD spectrum

148

6.4.2.3 Effect of polyethylenimine (PEI) on the morphology and crystal phase of
manganese oxides
PEI adsorbed on GnP surface acts as a capping agent; controls the nucleation and
growth of manganese oxide and changes the morphology from a continuous ribbon-like
structure to nanoparticle clusters. Since the oxidation state of manganese in Mn 3O4
(II,III) is lower than that of MnO2 (IV), PEI has played a role as a reducing agent as well.
Adsorption of polyelectrolytes (such as polyethylenimine) on a substrate surface and its
interaction with the substrate is a complicated and much researched subject [32-34].
Generally, the amine group on polyethylenimine and the induction of charge on this
group is associated with its ability to reduce metal precursor to generate metal particles
[33, 34]. However, various other factors such as the structure of the polyelectrolyte
(linear or branched), molecular weight, chain length, pH of the solution, ionic strength of
the precursor solution, solution temperature also play different roles in determining the
chemical structure and the morphology of the metal particle. Further investigation needs
to be carried out to understand the capping and reduction mechanism of PEI.

6.4.2.4 Significance Mn3O4 synthesis through polyethylenimine (PEI) mediated
method and application
Composites of Mn3O4 with graphene based materials have recently attracted much
attention as high performing Lithium-ion battery anodes [8, 11, 12, 26]. These research
works reveal capacity values up to 900 mAh/gm of composites. Since Mn3O4 has poor
electronic conductivity, deposition of Mn3O4 on high electronic conductivity substrates
such as graphene has been of particular interest [8, 11, 12, 24, 25, 26, 35]. Synthesis

149

of Mn3O4 has been a complicated process often including hazardous reducing agents,
high temperature treatments, inert gas atmosphere and long processing time. The
synthesis of crystalline Mn3O4 nanoparticles through a simple one-pot synthesis as
done in this research is unique.
Various composites of manganese oxides with GnP, structures and their potential
applications are listed in table 6-3.
Composites
GnP-Birnessite MnO2

Structures of manganese
oxides
Network of nano-ribbons

GnP-γ MnO2

Dispersed acicular structure

Potential applications
High performing
pseudocapacitor
electrode

High performing
pseudocapacitor
electrode
GnP-Mn3O4
Clusters of nano-particles
High performing Lithiumion battery anode
Table 6-3: various composites of manganese oxides with GnP, structures and their
potential applications

6.5 Conclusion
Composites of GnP-manganese oxides find many electrochemical energy storage
applications. Composites of different polymorphs of MnO2 and GnP that have the
potential to find application as pseudocapacitor materials can be synthesized by tuning
the molar ratio of reagent manganese salts. Substrate plays an important role in
prevention of agglomeration of synthesized manganese oxide crystals. Although
substrate reduces agglomeration, the crystal phases are retained nearly same.
Reduction of agglomeration might be of significance in overcoming kinetic limitations

150

during

electrochemical

applications.

Introduction

of

polymer

chain

such

as

polyethylenimine in the synthesis system acts as capping and reducing agent.
Polyethylinimine changes crystal morphology from a continuous ribbon-like structure to
clusters of nanoparticles and reduces the oxidation state of manganese to produce
Mn3O4. GnP-Mn3O4 composite has the potential to find application as high performing
Lithium-ion battery anode. Further investigation needs to be conducted to understand
the capping and reduction mechanism of polyethylinimine.

6.6 Acknowledgements
The author acknowledges the support Composite Material and Structures Center
(CMSC) research staff in accomplishing this work.

151

REFERENCES

152

REFERENCES

1. Synthesis and formation of mechanism of manganese oxide nanowires/nanorods,
Chem. Eur. J., 9 (1), 300-306 (2003)
2. Preparation and characterization of open tunnel oxide α-MnO2 precipitated by ozone
oxidation, Journal of Solid State Chemistry, 159, 94-102 (2011)
3. Carbon-supported manganese oxide nanocatalysts for rechargeable lithium-air
batteries, Journal of Power Sources, 195, 1370-1374 (2010)
4. Direct synthesis of palladium nanoparticles on Mn 3O4 modified multi-walled carbon
nanotubes: a highly active catalyst for methanol electro-oxidation in alkaline media,
Journal of Power Sources, 218, 320-330 (2012)
5. Facile synthesis of α-MnO2/graphene nanocomposites and their high performance as
lithium-ion battery anode, Materials Letters, 65, 2104-2106 (2011)
6. Carbon-supported nanostructured manganese oxide composite electrode for
electrochemical supercapacitor, Journal of Power Sources, 173, 1024-1028 (2007)
7. Microstructure control of MnO2/CNT hybrids under in-situ hydrothermal conditions,
Solid State Sciences, 12 (9), 1677-1682 (2010)
8. High gravimetric capacity and long cycle life in Mn3O4/graphene platelet/LiCMC
composite lithium-ion battery anodes, Journal of Power Sources, 213, 249-254 (2012)
9. Chelating agent and surfactant assisted synthesis of manganese oxide/carbon
nanotube composite for electrochemical capacitor, Materials Research Bulletin, 48,
1204-1212 (2013)
10. Ph.D. Dissertation, Jinglei Xiang, Michigan State University (2012)
11. Electrochemical and ex-situ analysis on manganese oxide/graphene hybrid anode
for lithium rechargeable batteries, J. Mater. Res., 26 (20), 2665-2671 (2011)
12. Mn3O4-Graphene hybrid as a high capacity anode material for lithium ion batteries,
J. Am. Chem. Soc., 132 (40), 13978-13980 (2010)
13. The reduction and oxidation behavior of manganese oxides, Catalysis Today, 47,
161-167 (1999)
14. Manganese oxide minerals: crystal structures and economic and environmental
significance, Proc. Natl. Acad. Sci. USA, 96, 3447-3454 (1999)

153

15. Three-dimensional nano MnO2/CB composite and its application for electrochemical
capacitor, Material Letters, 78, 127-130 (2012)
16. Hydrothermal synthesis of MnO2/CNT nanocomposite with a CNT core/porous
MnO2 sheath hierarchy architecture for supercapacitors, Nanoscale Research Letters,
7:33 (2012)
17. Effect of temperature on the pseudo-capacitive behavior of free standing
MnO2/carbon nanofibers composites electrodes in mild electrolyte, Journal of Power
Sources, 224, 86-92 (2013)
18. MnO2/onion-like carbon nanocomposites for pseudo-capacitors, J. Mater. Chem.,
22, 17584 (2012)
19. MnO2-graphene hybrid as an alternative cathodic catalyst to platinum in microbial
fuel cells, Journal of Power Sources, 216, 187-191 (2012)
20. Synthesis of a MnO2-graphene foam hybrid with controlled MnO2 particle shape and
its use as a supercapacitor electrode, Carbon, 50, 4865-4870 (2012)
21. The electrochemical behavior and catalytic activity for oxygen reduction of MnO 2/Ctoray gas diffusion electrodes, Journal of The Electrochemical Society, 159 (2), F23-F24
(2011)
22. In-situ hydrothermal synthesis of three-dimensional MnO2-CNT nanocomposites and
their electrochemical properties, Journal of Alloys and Compounds, 499, 259-264
(2010)
23. Preparation of γ-MnO2/carbon composite material by a wet chemical method,
Materials Research Bulletin, 36, 541-546 (2001)
24. Easy single-step route to manganese oxide nanoparticles embedded in carbon and
their magnetic properties, J. Phys. Chem. C, 112, 15752-15758 (2008)
25. Microwave-assisted hydrothermal synthesis of Mn3O4/reduced graphene oxide
composites for high power supercapacitors, Journal of Power Sources 217, 184-192
(2012)
26. Rapid microwave-assisted synthesis of Mn3O4-graphene nanocomposite and its
lithium storage properties, J. Mater. Chem., 22, 3600 (2012)
27. Porous graphite/silicon micro-sphere prepared by in-situ carbothermal reduction and
spray drying for Lithium ion batteries, ECS Solid State Letters, 1 (2), M5-M7 (2012)
28. X-ray characterization of materials, Eric Lifshin, Wiley-VCH (2008)

154

29. Synthesis and characterization of manganese dioxide spontaneous coated on
carbon nanotubes, Carbon, 45, 375-382 (2007)
30. Carbon-electrochemical and physicochemical properties, John Wiley & Sons (1988)
31. Mn3O4 nanoparticles embedded into graphene nanosheets: preparation,
characterization, and electrochemical properties for supercapacitors, Electrochimica
Acta, 55, 6812-6817 (2010)
32. Adsorption and electrokinetic properties of polyethylenimine on silica surfaces,
Langmuir, 18, 6164-6169 (2002)
33. One-step synthesis and characterization of polyelectrolyte-protected gold
nanoparticles through a thermal process, Polymer, 45, 2181-2184 (2004)
34. Synthesis of positively charged silver nanoparticles via photoreduction of AgNO 3 in
branched polyethylenimine/HEPES solutions, Langmuir, 23, 9836-9843 (2007)
35. Mn3O4 nanoparticles embedded into graphene nanosheets: preparation,
characterization, and electrochemical properties for for supercapacitors, Electrochimica
Acta, 55, 6812-6817 (2010)

155

Chapter 7
Summary and Recommendation

7.1 Summary
Graphene nanoplatelets (GnP) can be prepared with high open surface area and
different particle sizes through the exfoliation of graphite. The high surface area of GnP
predominantly consists of graphene basal planes. GnP possesses excellent electrical;
thermal; mechanical and barrier properties and rivals single or few layer graphene. The
primary objective of this research has been to explore the value of high open surface
area of GnP in various energy storage technologies.
Lithium-air batteries have the potential to exceed the energy density of Lithium-ion
batteries since the cathode material, oxygen, is not stored inside the cell but is
continuously provided from the atmosphere. Lithium oxides are generated as reaction
products at the cathode. High surface area porous carbon materials are generally
investigated as Lithium-air cathodes to provide adequate sites for lithium oxides
deposition. However, one disadvantage of porous carbons is that the pore orifice gets
clogged with deposited reaction products, leaving unutilized volume within the pore. An
alternative could be to use high open surface area materials. In this research, Grade-C750 (750 m2/gm surface area, aggregates of sub-micron platelets) GnP delivers higher
discharge capacity with little overpotential compared to Grade-M-15 (120-150 m2/gm
surface area, 15 μm particle size) GnP. This research shows that by creating high open
surface area instead of introducing a high density of small diameter pores, it’s possible

156

to improve discharge capacity. Paper-like, binder free, self-standing structures have the
potential to be used as flexible electrodes as well as enhance electrode energy density
by doing away with electrochemically inactive metal current collector. A GnP paper can
be prepared with large particle size Grade-M-15 GnP. However, the discharge capacity
of Grade-M-15 GnP paper electrode is reduced as the thickness of the paper electrode
and the amount of electrode material increases. This trend of capacity reduction has
been reversed through hybridization with sub-micron size platelets of Grade-C-750 GnP
by fabricating a bilayer GnP paper and increasing the net surface area of a paper
electrode. Higher discharge capacity at high electrode loading is of significance as it is
the only way to deliver current for a long time that meets the needs of practical
applications.
Supercapacitor technology bridges the gap between secondary batteries and traditional
capacitors. They deliver higher capacitance than a traditional dielectric based capacitor
and can be charged or discharged in seconds; at a much higher rate compared to
secondary batteries. The capacitance of a supercapacitor electrode is proportional to its
electro-active surface area. The potential of high surface area Grade-C-750 GnP as
electric double layer capacitor (EDLC) electrodes combined with aqueous KOH
electrolyte has been investigated in this research. Grade-C-750 GnP exhibits near-ideal
double layer capacitive behavior at a high scan rate of 1 V/sec. Excellent capacitive
characteristics of Grade-C-750 GnP can be attributed to its low equivalent series
resistance (ESR) and charge transfer resistance (Rct). In comparison relatively low
surface area Grade-M-15 GnP has negligible capacitance. However, large particle size
Grade-M-15 GnP can be structured in to a paper which then acts as the substrate for

157

sub-micron platelet shaped Grade-C-750 GnP. By doing away with much higher density
and hence weight metal current collector, an energy density can be increased.
Metal oxides and metal oxides composites with carbon/graphitic materials find various
applications including electrochemical energy storage. A GnP composite system
consisting of high open surface area GnP combined with manganese oxides have been
investigated in this research. Composites of polymorphs of MnO 2 and GnP that are of
interest as pseudocapacitor materials are synthesized through a wet-chemical route.
GnP plays a significant role in reduction of agglomeration of MnO 2 crystals while
retaining the crystal phase. This is of significance as agglomeration imposes kinetic
limitations on materials. Polymers such as polyethylenimine, when introduced in the
system, acts as capping and reducing agent. Manganese oxide morphology is changed
from a continuous ribbon-like structure to clusters of nanoparticles and the oxidation
state of manganese is reduced to form Mn3O4. Synthesis of Mn3O4 nanoparticles
through a simple polymer mediated method is unique. GnP-Mn3O4 composite has the
potential to find application as high performing Lithium-ion battery anode.

7.2 Recommendation
GnP-manganese oxides composites have been shown to offer significant performance
advantages. Composites of Grade-M-15 GnP with birnessite-MnO2 and γ-MnO2 should
be investigated as pseudocapacitor electrodes. MnO2 would improve specific
capacitance; while the high electrical conductivity of GnP would prevent the CV
trajectory from distortion at high scan rate. In addition, the net surface area of GnPMnO2 composite is likely to be higher compared to pristine GnP. This should be a

158

contributing factor towards higher specific capacitance. The composite of Grade-M-15
GnP with Mn3O4 should also be investigated as a Lithium-ion battery anode. Small
particle size and high crystallinity of synthesized Mn 3O4 combined with the high
electronic conductivity of the GnP substrate might make the composite a high
performing Lithium-ion battery anode. Further research is recommended to understand
how polyethylenimine acts as capping and reducing agent and whether discrete
nanoparticles could be obtained instead of nanoparticle clusters.

159